Release v0.5.4

* origin/master:
  Added API call getSnatched
  Potential fix for #2055

.gitignore

@@ -6,12 +6,13 @@
 *.pyproj
 *.sln
 
-# Logs and databases #
+# Headphones files #
 ######################
 *.log
 *.db*
 *.db-journal
 *.ini
+version.lock
 logs/*
 cache/*
 
@@ -19,6 +20,7 @@ cache/*
 ##################
 *.crt
 *.key
+*.csr
 
 # OS generated files #
 ######################
@@ -64,6 +66,3 @@ _ReSharper*/
 /logs
 .project
 .pydevproject
-
-
-headphones_docs

.travis.yml

@@ -8,9 +8,10 @@ language: python
 python:
   - "2.6"
   - "2.7"
+# pylint 1.4 does not run under python 2.6
 install:
  - pip install pyOpenSSL
- - pip install pylint
+ - pip install pylint==1.3.1
  - pip install pyflakes
  - pip install pep8
 script:

API.md

@@ -68,6 +68,8 @@ Unmark album as wanted / i.e. mark as skipped
 force search for wanted albums - not launched in a separate thread so it may take a bit to complete
 ### forceProcess
 Force post process albums in download directory - also not launched in a separate thread
+### forceActiveArtistsUpdate
+force Active Artist Update - also not launched in a separate thread
 
 ### getVersion
 Returns some version information: git_path, install_type, current_version, installed_version, commits_behind

CHANGELOG

@@ -1,7 +0,0 @@
-v0.5 Released 10 Nov 2014
--------------------------
-- Several bug fixes (please retest your posted issues)
-- Cue splitter
-- Other improvements
-
- 

CHANGELOG.md

@@ -0,0 +1,158 @@
+# Changelog
+
+## v0.5.4
+Released 05 February 2015
+
+Highlights:
+* Added: backported 'Scan Artist' feature from `sarakha63/headphones`
+* Fixed: change file permissions of cache files according to settings (#2102)
+* Fixed: hide Songkick if no regional events are available
+* Fixed: only reschedule jobs if timeout changed (#2099)
+* Fixed: limit dialog height (#2096)
+* Improved: upgraded requests to 2.5.1
+* Improved: upgraded unidecode to 0.04.17
+* Improved: upgraded tzlocal to 1.1.2
+* Improved: upgraded PyTZ to 2014.10
+* Improved: upgraded mako to 1.0.1
+* Improved: upgraded mutagen to 1.27
+* Improved: upgraded beets to 1.3.10
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.5.3...v0.5.4).
+
+## v0.5.3
+Released 15 January 2015
+
+Highlights:
+* Added: update active artists via API (#2075)
+* Fixed: queue instance lacks `add` method (#2055)
+* Improved: better SSL error messages (#2058)
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.5.2...v0.5.3).
+
+## v0.5.2
+Released 28 December 2014
+
+Highlights:
+* Added: advanced option to ignore certain folders by patterns. (#2037)
+* Added: advanced option to ignore certain files by patterns (library only)
+* Added: specify optional paths to CUE splitting tools (#1938)
+* Added: experimental support for OPB (provide details yourself!)
+* Fixed: magnet to torrent conversion (#1926)
+* Fixed: new KAT URL (#2043)
+* Fixed: LMS notifications (#1564)
+* Improved: notify user of SSL-related warnings, instead of silently failing.
+* Improved: show all search results for 'Choose Specific Release'
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.5.1...v0.5.2).
+
+## v0.5.1
+Released 13 November 2014
+
+Highlights:
+* Added: allow one to disable interval tasks (#2002)
+* Added: script to downgrade Headphones to last version that started (Linux only)
+* Fixed: SSL issues in CherryPy. Self-generated certificates will be 2048 now (#1995)
+* Fixed: Transmission URL detection (#1998)
+* Fixed: missing dependencies for APScheduler and CherryPy (#2001)
+* Improved: symlink infinite recursion detection
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.5...v0.5.1).
+
+## v0.5
+Released 10 November 2014
+
+Highlights:
+* Added: CUE splitter
+* Added: filter search result by MusicBrainz Release Group ID
+* Added: follow symlinks while scanning library (#1953)
+* Fixed: crash during post processing (#1897)
+* Fixed: embedding lyrics (#1896)
+* Fixed: HTTP errors with older versions of Python 2.6
+* Fixed: jump back to top of page (#1948)
+* Improved: parse MusicBrainz RGID first when post processing (#1952)
+* Improved: Growl unicode characters (#1695)
+* Improved: search handling for PB and KAT
+* Improved: Last.FM API support (#1877)
+* Improved: upgraded CherryPy to version 3.6.0
+* Improved: upgraded Requests to version 2.4.1
+* Improved: upgraded APScheduler to version 3.0.1
+* Improved: lot of code refactoring
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.4...v0.5).
+
+## v0.4
+Released 20 September 2014
+
+Highlights:
+* Added: support for libav-tools (which replaces FFmpeg under Ubuntu)
+* Added: option to freeze library when post processing
+* Added: Songkick per area
+* Added: rename original NFO file (#1797)
+* Removed: removed dead search providers
+* Fixed: removed left-overs of old packages
+* Improved: rename by original folder (#1811)
+* Improved: uTorrent fixess
+* Improved: delayed loading of album art
+* Improved: search result parser
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.3.4...v0.4).
+
+## v0.3.4
+Released 15 May 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.3.3...v0.3.4).
+
+## v0.3.3
+Released 05 May 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.3.2...v0.3.3).
+
+## v0.3.2
+Released 17 April 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.3.1...v0.3.2).
+
+## v0.3.1
+Released 11 April 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.3...v0.3.1).
+
+## v0.3
+Released 09 April 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.2.3...v0.3).
+
+## v0.2.3
+Released 28 March 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.2.2...v0.2.3).
+
+## v0.2.2
+Released 15 January 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.2.1...v0.2.2).
+
+## v0.2.1
+Released 13 January 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.2...v0.2.1).
+
+## v0.2
+Released 13 January 2014
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.1.2...v0.2).
+
+## v0.1.2
+Released 11 October 2013
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.1.1...v0.1.2).
+
+## v0.1.1
+Released 26 August 2013
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/v0.1...v0.1.1).
+
+## v0.1
+Released 05 August 2013
+
+The full list of commits can be found [here](https://github.com/rembo10/headphones/compare/2156e1341405d07c5bcfbe994f6b354b32d94cda...v0.1).

CONTRIBUTING.md

@@ -1,26 +1,28 @@
 # Contributing to Headphones
 
-## For users
+## Issues
 In case you read this because you are posting an issue, please take a minute and conside the things below. The issue tracker is not a support forum. It is primarily intended to submit bugs, improvements or feature requests. However, we are glad to help you, and make sure the problem is not caused by Headphones, but don't expect step-by-step answers.
 
-* Use the search function. Chances are that your problem is already discussed.
+* Use the search function. Chances are that your problem is already discussed. Do not append to (closed) issues if your problem does not fit the discussion.
 * Visit the [Troubleshooting](../../wiki/TroubleShooting) wiki first.
 * Use [proper formatting](https://help.github.com/articles/github-flavored-markdown/). Paste your logs in code blocks.
 * Close your issue if you resolved it.
 
-## For developers
-If you think you can contribute code to the Headphones repository, do not hesitate to submit a pull request. 
+## Pull Requests
+If you think you can contribute code to the Headphones repository, do not hesitate to submit a pull request.
 
 ### Branches
-All pull requests should be based on the `develop` branch. When you want to develop a new feature, clone the repository with `git clone origin/develop -b FEATURE_NAME`. Use meaningful commit messages.
+All pull requests should be based on the `develop` branch, to minimize cross merges. When you want to develop a new feature, clone the repository with `git clone origin/develop -b FEATURE_NAME`. Use meaningful commit messages.
 
-### Code compatibility
+### Python Code
+
+#### Compatibility
 The code should work with Python 2.6 and 2.7. Note that Headphones runs on different platforms, including Network Attached Storage devices such as Synology.
 
 Re-use existing code. Do not hesitate to add logging in your code. You can the logger module `headphones.logger.*` for this. Web requests are invoked via `headphones.request.*` and derived ones. Use these methods to automatically add proper and meaningful error handling.
 
-### Code conventions
-Altough Headphones did not adapt a code convention in the past, we try to follow the [PEP8](http://legacy.python.org/dev/peps/pep-0008/) conventions for future code. A short summary to remind you (copied from http://wiki.ros.org/PyStyleGuide):
+#### Code conventions
+Although Headphones did not adapt a code convention in the past, we try to follow the [PEP8](http://legacy.python.org/dev/peps/pep-0008/) conventions for future code. A short summary to remind you (copied from http://wiki.ros.org/PyStyleGuide):
 
  * 4 space indentation
  * 80 characters per line
@@ -31,5 +33,20 @@ Altough Headphones did not adapt a code convention in the past, we try to follow
  * `_private_something`
  * `self.__really_private_field`
  * `_global`
- 
-Document your code!
+
+#### Documentation
+Document your code. Use docstrings See [PEP-257](https://www.python.org/dev/peps/pep-0257/) for more information.
+
+#### Continuous Integration
+Headphones has a configuration file for [travis-ci](https://travis-ci.org/). You can add your forked repo to Travis to have it check your code against PEP8, PyLint, and PyFlakes for you. Your pull request will show a green check mark or a red cross on each tested commit, depending on if linting passes.
+
+### HTML/Template code
+
+#### Compatibility
+HTML5 compatible browsers are targetted. There is no specific mobile version of Headphones yet.
+
+#### Conventions
+* 4 space indentation
+* `methodName`
+* `variableName`
+* `ClassName`

Headphones.py

@@ -89,6 +89,10 @@ def main():
     if args.quiet:
         headphones.QUIET = True
 
+    # Do an intial setup of the logger.
+    logger.initLogger(console=not headphones.QUIET, log_dir=False,
+        verbose=headphones.VERBOSE)
+
     if args.daemon:
         if sys.platform == 'win32':
             sys.stderr.write(
@@ -103,8 +107,8 @@ def main():
         # If the pidfile already exists, headphones may still be running, so
         # exit
         if os.path.exists(headphones.PIDFILE):
-            sys.exit(
-                "PID file '" + headphones.PIDFILE + "' already exists. Exiting.")
+            raise SystemExit("PID file '%s' already exists. Exiting." %
+                headphones.PIDFILE)
 
         # The pidfile is only useful in daemon mode, make sure we can write the
         # file properly
@@ -115,11 +119,10 @@ def main():
                 with open(headphones.PIDFILE, 'w') as fp:
                     fp.write("pid\n")
             except IOError as e:
-                raise SystemExit(
-                    "Unable to write PID file: %s [%d]", e.strerror, e.errno)
+                raise SystemExit("Unable to write PID file: %s", e)
         else:
-            logger.warn(
-                "Not running in daemon mode. PID file creation disabled.")
+            logger.warn("Not running in daemon mode. PID file creation " \
+                "disabled.")
 
     # Determine which data directory and config file to use
     if args.datadir:
@@ -164,6 +167,16 @@ def main():
     else:
         http_port = int(headphones.CONFIG.HTTP_PORT)
 
+    # Check if pyOpenSSL is installed. It is required for certificate generation
+    # and for CherryPy.
+    if headphones.CONFIG.ENABLE_HTTPS:
+        try:
+            import OpenSSL
+        except ImportError:
+            logger.warn("The pyOpenSSL module is missing. Install this " \
+                "module to enable HTTPS. HTTPS will be disabled.")
+            headphones.CONFIG.ENABLE_HTTPS = False
+
     # Try to start the server. Will exit here is address is already in use.
     web_config = {
         'http_port': http_port,
@@ -178,13 +191,14 @@ def main():
     }
     webstart.initialize(web_config)
 
+    # Start the background threads
+    headphones.start()
+
+    # Open webbrowser
     if headphones.CONFIG.LAUNCH_BROWSER and not args.nolaunch:
         headphones.launch_browser(headphones.CONFIG.HTTP_HOST, http_port,
                                   headphones.CONFIG.HTTP_ROOT)
 
-    # Start the background threads
-    headphones.start()
-
     # Wait endlessy for a signal to happen
     while True:
         if not headphones.SIGNAL:

README.md

@@ -10,7 +10,7 @@ You are free to join the Headphones support community on IRC where you can ask q
 
 ## Installation and Notes
 
-* [Installation page](../../wiki/Usage-guide) shows you how to install Headphones.
+* [Installation page](../../wiki/Installation) shows you how to install Headphones.
 * [Usage guide](../../wiki/Usage-guide) introduces you to Headphones.
 * [Troubleshooting page](../../wiki/TroubleShooting) in the wiki can help you with common problems.
 

contrib/clean_pyc.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+# Display information
+echo "This script will remove *.pyc files. These files are generated by Python, but they can cause conflicts after an upgrade. It's safe to remove them, because they will be regenerated."
+echo "Press enter to continue, or CTRL + C to quit."
+read
+
+# Remove the *.pyc
+find "`dirname $0`/.." -type f -name "*.pyc" -exec rm -rf {} \;

contrib/downgrade.sh

@@ -0,0 +1,32 @@
+#!/bin/bash
+
+# Parameter check
+if [ -z "$1" ]; then
+    echo "Syntax: $0 <data directory>"
+    exit 1
+fi
+
+# Version file check
+if [ ! -s "$1/version.lock" ]; then
+    echo "Missing the version.lock file in the data folder, or the file is empty. Did you start Headphones at least once?"
+    exit 1
+fi
+
+# Git installation check
+if [ ! -x "$(command -v git)" ]; then
+    echo "Git is required to downgrade."
+    exit 1
+fi
+
+# Display information
+HASH=$(cat $1/version.lock)
+
+echo "This script will try to downgrade Headphones to the last version that started, version $HASH. Make sure you have a backup of your config file and database, just in case!"
+echo "Press enter to continue, or CTRL + C to quit."
+read
+
+# Downgrade
+cd "`dirname $0`/.."
+git reset --hard "$HASH"
+
+echo "All done, Headphones should be downgraded to the last version that started."

data/interfaces/default/album.html

@@ -48,44 +48,44 @@
 					</div>
 				</div>
 			<a class="menu_link_edit" id="edit_search_term" href="#"><i class="fa fa-pencil"></i> Edit Search Term</a>
-				<div id="dialog2" title="Enter your own search term for this album" style="display:none" class="configtable">
-					<form action="editSearchTerm" method="GET" id="editSearchTerm">
-						<input type="hidden" name="AlbumID" value="${album['AlbumID']}">
-						<div class="row">
-							<%
-								if not album['SearchTerm']:
-									search_term = ""
-								else:
-									search_term = album['SearchTerm']
-								
-							%>
-							<input type="text" value="${search_term}" name="SearchTerm" size="40" />
-						</div>
-						<input type="button" value="Save changes" onclick="doAjaxCall('editSearchTerm',$(this),'tabs',true);return false;" data-success="Search term updated"/>
-					</form>
-				</div>	
+			<div id="dialog2" title="Enter your own search term for this album" style="display:none" class="configtable">
+				<form action="editSearchTerm" method="GET" id="editSearchTerm">
+					<input type="hidden" name="AlbumID" value="${album['AlbumID']}">
+					<div class="row">
+						<%
+							if not album['SearchTerm']:
+								search_term = ""
+							else:
+								search_term = album['SearchTerm']
+						%>
+						<input type="text" value="${search_term}" name="SearchTerm" size="40" />
+					</div>
+					<input type="button" value="Save changes" onclick="doAjaxCall('editSearchTerm',$(this),'tabs',true);return false;" data-success="Search term updated"/>
+				</form>
+			</div>
 			<a class="menu_link_edit" id="choose_specific_download" href="#" onclick="getAvailableDownloads()"><i class="fa fa-search"></i> Choose Specific Download</a>
-				<div id="choose_specific_download_dialog" title="Choose a specific download for this album" style="display:none" class="configtable">
-						<table class="display" id="downloads_table">
-							<thead>
-								<tr>
-									<th id="title">Title</th>
-									<th id="size">Size</th>
-									<th id="provider">Provider</th>
-									<th id="kind">Kind</th>
-								</tr>
-							</thead>
-							<tbody id="downloads_table_body">
-							</tbody>
-						</table>
-				</div>	
+			<div id="choose_specific_download_dialog" title="Choose a specific download for this album" style="display:none" class="configtable">
+				<table class="display" id="downloads_table">
+					<thead>
+						<tr>
+							<th id="title">Title</th>
+							<th id="size">Size</th>
+							<th id="provider">Provider</th>
+							<th id="kind">Kind</th>
+							<th id="matches" title="Matches quality and/or seed settings.">Matches</th>
+						</tr>
+					</thead>
+					<tbody id="downloads_table_body">
+					</tbody>
+				</table>
+			</div>
 		</div>
-	</div>	
+	</div>
 </%def>
 
 <%def name="body()">
 	<div class="table_wrapper">
-		<div id="albumheader" class="clearfix">	
+		<div id="albumheader" class="clearfix">
 			<div id="albumImg">
 				<img height="200" alt="" class="albumArt" src="artwork/album/${album['AlbumID']}">
 			</div>
@@ -197,14 +197,20 @@
 		var elem = $("#albumInfo");
 		getInfo(elem,id,'album');
 	}
-	
+
 	function initThisPage() {
 		$('#album_chooser').click(function() {
-			$('#dialog').dialog({ width: "500px" });
+			$('#dialog').dialog({
+				width: 500,
+				maxHeight: 500
+			});
 			return false;
 		});
 		$('#edit_search_term').click(function() {
-			$('#dialog2').dialog({ width: "500px" });
+			$('#dialog2').dialog({
+				width: 500,
+				maxHeight: 500
+			});
 			return false;
 		});
 		$('#refresh_artist').click(function() {
@@ -213,8 +219,8 @@
 		initActions();
 		setTimeout(function(){
 			initFancybox();
-			},1000);
-		
+		}, 1000);
+
 		$('#track_table').dataTable({
 			"aaSorting": [],
 			"bFilter": false,
@@ -231,21 +237,26 @@
 			feedback.fadeOut();
 			search_results = data
 			for( var i = 0, len = data.length; i < len; i++ ) {
-    			$('#downloads_table_body').append('<tr><td id="title"><a href="#" onclick="downloadSpecificRelease('+i+')">'+data[i].title+'</a></td><td id="size"><span title='+data[i].size+'></span>'+(data[i].size / (1024*1024)).toFixed(2)+' MB</td><td id="provider">'+data[i].provider+'</td><td id="kind">'+data[i].kind+'</td></tr>');
+    			$('#downloads_table_body').append('<tr><td id="title"><a href="#" onclick="downloadSpecificRelease('+i+')">'+data[i].title+'</a></td><td id="size"><span title='+data[i].size+'></span>'+(data[i].size / (1024*1024)).toFixed(2)+' MB</td><td id="provider">'+data[i].provider+'</td><td id="kind">'+data[i].kind+'</td><td id="matches">'+data[i].matches+'</td></tr>');
     		}
 			$('#downloads_table').dataTable({
 				"aoColumns": [
 					null,
-					{ "sType": "title-numeric"},
+					{"sType": "title-numeric"},
 					null,
-					null
+					null,
+					{"sType": "string"}
 				],
-				"aaSorting": [],
+				"aaSorting": [[ 4, 'desc']],
 				"bFilter": false,
 				"bInfo": false,
 				"bPaginate": false,
+				"bDestroy": true
 			});
-    		$("#choose_specific_download_dialog").dialog({ width: "1000px" });
+    		$("#choose_specific_download_dialog").dialog({
+    			width: "80%",
+    			maxHeight: 500
+    		});
     		return false;
 		});
 	}

data/interfaces/default/artist.html

@@ -10,6 +10,7 @@
 		<div id="subhead_menu">
 			<a id="menu_link_refresh" onclick="doSimpleAjaxCall('refreshArtist?ArtistID=${artist['ArtistID']}')" href="#"><i class="fa fa-refresh"></i> Refresh Artist</a>
 			<a id="menu_link_delete" href="deleteArtist?ArtistID=${artist['ArtistID']}"><i class="fa fa-trash-o"></i> Delete Artist</a>
+			<a id="menu_link_scan" onclick="doAjaxCall('scanArtist?ArtistID=${artist['ArtistID']}', $(this)),'table'" href="#" data-success="'${artist['ArtistName']}' was scanned"><i class="fa fa-refresh"></i> Scan Artist</a>
 			%if artist['Status'] == 'Paused':
 				<a id="menu_link_resume" href="#" onclick="doAjaxCall('resumeArtist?ArtistID=${artist['ArtistID']}',$(this),true)" data-success="${artist['ArtistName']} resumed"><i class="fa fa-play"></i> Resume Artist</a>
 			%else:
@@ -189,45 +190,45 @@
 		$.getJSON("https://api.songkick.com/api/3.0/artists/mbid:${artist['ArtistID']}/calendar.json?apikey=${headphones.CONFIG.SONGKICK_APIKEY}&jsoncallback=?",
 			function(data){
 				if (data['resultsPage'].totalEntries >= 1) {
-
-                    if( ${songkick_filter_enabled} ) {
+                    if (${songkick_filter_enabled}) {
                         data.resultsPage.results.event = $.grep(data.resultsPage.results.event, function(element,index){
-                        return element.venue.metroArea.id == ${songkick_location};
-                    });
+                    	    return element.venue.metroArea.id == ${songkick_location};
+                    	});
                     }
 
-					var tourDate;
+                    if (data.resultsPage.results.event.length > 0) {
+						var tourDate;
 
-					calendarDomNode.show();
-					$("#artistImg").addClass('on-tour');
+						calendarDomNode.show();
+						$("#artistImg").addClass('on-tour');
 
-					jQuery.each( data['resultsPage'].results.event, function( i, event ) {
-						tourDate = template;
-						tourDate = tourDate.replace('URI',event.uri);
-						tourDate = tourDate.replace('NAME',event.displayName);
-						tourDate = tourDate.replace('LOC',event.location.city);
+						jQuery.each(data.resultsPage.results.event, function(i, event) {
+							tourDate = template;
+							tourDate = tourDate.replace('URI',event.uri);
+							tourDate = tourDate.replace('NAME',event.displayName);
+							tourDate = tourDate.replace('LOC',event.location.city);
 
-						calendarDomNode.append(tourDate);
-					});
+							calendarDomNode.append(tourDate);
+						});
 
-					calendarDomNode.append('<li><img src="interfaces/default/images/songkick.png" alt="concerts by songkick" class="sk-logo" /></li>');
-
-				$(function() {
-				  $("#artistCalendar").each(function() {
-					$("li:gt(4)", this).hide(); /* :gt() is zero-indexed */
-					$("li:nth-child(5)", this).after("<br><li class='more'><a href='#'>More...</a></li>"); /* :nth-child() is one-indexed */
-				  });
-				  $("li.more a").live("click", function() {
-					var li = $(this).parents("li:first");
-					li.parent().children().show();
-					li.remove();
-					return false;
-				  });
-				});
+						calendarDomNode.append('<li><img src="interfaces/default/images/songkick.png" alt="concerts by songkick" class="sk-logo" /></li>');
 
+						$(function() {
+						  $("#artistCalendar").each(function() {
+							$("li:gt(4)", this).hide(); /* :gt() is zero-indexed */
+							$("li:nth-child(5)", this).after("<br><li class='more'><a href='#'>More...</a></li>"); /* :nth-child() is one-indexed */
+						  });
+						  $("li.more a").live("click", function() {
+							var li = $(this).parents("li:first");
+							li.parent().children().show();
+							li.remove();
+							return false;
+						  });
+						});
+					}
 				}
-			// data is JSON response object
-		});
+			}
+		);
 	}
 
 	var loadingMessage = false;

data/interfaces/default/base.html

@@ -52,7 +52,7 @@
 				<li><a href="extras">extras</a></li>
 				<li><a href="manage">manage</a></li>
 				<li><a href="history">history</a></li>
-				<li><a href="logs" class="log">logs</a></li>
+				<li><a href="logs">logs</a></li>
 				<li><a href="config" class="config"><i class="fa fa-gear fa-lg"></i></a></li>
 			</ul>
 			<div id="searchbar">
@@ -139,4 +139,4 @@
 
 <%def name="javascriptIncludes()"></%def>
 <%def name="headIncludes()"></%def>
-<%def name="headerIncludes()"></%def>
+<%def name="headerIncludes()"></%def>

data/interfaces/default/config.html

@@ -101,6 +101,7 @@
 
                             <fieldset>
                                 <legend>Interval</legend>
+                                <small>An interval of 0 will disable a task.</small>
                                 <div class="row">
                                     <label title="Time between two searches for new downloads.">
                                         Search Interval
@@ -137,7 +138,7 @@
                     </tr>
                     <tr>
                         <div class="configmessage">
-                            <i class="fa fa-info-circle"></i> Web Interface changes require a restart to take effect
+                            <i class="fa fa-info-circle"></i> Web Interface changes require a restart to take effect. Saving settings will restart intervals if changed.
                         </div>
                     </tr>
                 </table>
@@ -266,7 +267,7 @@
                                         prio_force = ''
                                 %>
                                 <div class="row">
-                                    <label>NZBget Priority:</label>
+                                    <label>NZBget Priority</label>
                                     <select name="nzbget_priority" id="nzbget_priority">
                                         <option value="-100" ${prio_verylow}>Very Low</option>
                                         <option value="-50" ${prio_low}>Low</option>
@@ -294,7 +295,7 @@
                                         Music Download Directory:
                                     </label>
                                     <input type="text" name="download_dir" value="${config['download_dir']}" size="50">
-                                    <small>Full path where SAB or NZBget downloads your music. e.g. /Users/name/Downloads/music</small>
+                                    <small>Full path where SAB or NZBget downloads your music, e.g. /Users/name/Downloads/music</small>
                                 </div>
                                 <div class="row">
                                     <label title="Number of days of retention your usenet provider provides.">
@@ -381,14 +382,14 @@
                             </fieldset>
                             <fieldset id="general_torrent_options">
                                 <div class="row">
-                                    <label>Minimum seeders:</label>
+                                    <label>Minimum seeders</label>
                                     <input type="text" name="numberofseeders" value="${config['numberofseeders']}" size="5">
                                     <small>Number of minimum seeders a torrent must have to be accepted</small>
                                 </div>
                                 <div class="row">
                                     <label>Music Download Directory</label>
                                     <input type="text" name="download_torrent_dir" value="${config['download_torrent_dir']}" size="50">
-                                    <small>Full path where your torrent client downloads your music e.g. '/Users/name/Downloads/music'</small>
+                                    <small>Full path where your torrent client downloads your music e.g. /Users/name/Downloads/music.</small>
                                 </div>
                                 <div class="row checkbox">
                                     <label>Keep Files for Seeding</label>
@@ -526,12 +527,30 @@
                                     </div>
                                     <div class="config">
                                         <div class="row">
-                                            <label>Proxy URL (Optional): </label>
+                                            <label>Proxy URL</label>
                                             <input type="text" name="piratebay_proxy_url" value="${config['piratebay_proxy_url']}" size="36">
+                                            <small>Optional. Leave empty for default.</small>
                                         </div>
                                         <div class="row">
-                                            <label>Seed Ratio: </label>
-                                            <input type="text" class="override-float" name="piratebay_ratio" value="${config['piratebay_ratio']}" size="10" title="Stop seeding when ratio met, 0 = unlimited. Scheduled job will remove torrent when post processed and finished seeding">
+                                            <label>Seed Ratio</label>
+                                            <input type="text" class="override-float" name="piratebay_ratio" value="${config['piratebay_ratio']}" size="10" title="Stop seeding when ratio met, 0 = unlimited. Scheduled job will remove torrent when post processed and finished seeding.">
+                                        </div>
+                                    </div>
+                                </fieldset>
+
+                                <fieldset>
+                                    <legend>Old Pirate Bay</legend>
+                                    <div class="row checkbox">
+                                        <input id="use_oldpiratebay" type="checkbox" name="use_oldpiratebay" value="1" ${config['use_oldpiratebay']} /><label>Use Old Pirate Bay</label>
+                                    </div>
+                                    <div class="config">
+                                        <div class="row">
+                                            <label>URL</label>
+                                            <input type="text" name="oldpiratebay_url" value="${config['oldpiratebay_url']}" size="36">
+                                        </div>
+                                        <div class="row">
+                                            <label>Seed Ratio</label>
+                                            <input type="text" class="override-float" name="oldpiratebay_ratio" value="${config['oldpiratebay_ratio']}" size="10" title="Stop seeding when ratio met, 0 = unlimited. Scheduled job will remove torrent when post processed and finished seeding.">
                                         </div>
                                     </div>
                                 </fieldset>
@@ -543,11 +562,12 @@
                                     </div>
                                     <div class="config">
                                         <div class="row">
-                                            <label>Proxy URL (Optional): </label>
+                                            <label>Proxy URL</label>
                                             <input type="text" name="kat_proxy_url" value="${config['kat_proxy_url']}" size="36">
+                                            <small>Optional. Leave empty for default.</small>
                                         </div>
                                         <div class="row">
-                                            <label>Seed Ratio: </label>
+                                            <label>Seed Ratio</label>
                                             <input type="text" class="override-float" name="kat_ratio" value="${config['kat_ratio']}" size="10" title="Stop seeding when ratio met, 0 = unlimited. Scheduled job will remove torrent when post processed and finished seeding">
                                         </div>
                                     </div>
@@ -560,15 +580,15 @@
                                     </div>
                                     <div class="config">
                                         <div class="row">
-                                            <label>UID Number: </label>
+                                            <label>UID Number</label>
                                             <input type="text" name="waffles_uid" value="${config['waffles_uid']}" size="36">
                                         </div>
                                         <div class="row">
-                                            <label>Passkey: </label>
+                                            <label>Passkey</label>
                                             <input type="password" name="waffles_passkey" value="${config['waffles_passkey']}" size="36">
                                         </div>
                                         <div class="row">
-                                            <label>Seed Ratio: </label>
+                                            <label>Seed Ratio</label>
                                             <input type="text" class="override-float" name="waffles_ratio" value="${config['waffles_ratio']}" size="10" title="Stop seeding when ratio met, 0 = unlimited. Scheduled job will remove torrent when post processed and finished seeding">
                                         </div>
                                     </div>
@@ -581,15 +601,15 @@
                                     </div>
                                     <div class="config">
                                         <div class="row">
-                                            <label>User Name: </label>
+                                            <label>Userame</label>
                                             <input type="text" name="rutracker_user" value="${config['rutracker_user']}" size="36">
                                         </div>
                                         <div class="row">
-                                            <label>Password: </label>
+                                            <label>Password</label>
                                             <input type="password" name="rutracker_password" value="${config['rutracker_password']}" size="36">
                                         </div>
                                         <div class="row">
-                                            <label>Seed Ratio: </label>
+                                            <label>Seed Ratio</label>
                                             <input type="text" class="override-float" name="rutracker_ratio" value="${config['rutracker_ratio']}" size="10" title="Stop seeding when ratio met, 0 = unlimited. Scheduled job will remove torrent when post processed and finished seeding">
                                         </div>
                                     </div>
@@ -688,7 +708,7 @@
                             </fieldset>
                             <fieldset>
                                 <legend>Search Words</legend>
-                                <small>Separate words with a comma, e.g. "word1,word2,word3"</small>
+                                <small>Separate words with a comma, e.g. "word1,word2,word3".</small>
                                 <div class="row">
                                     <label>Ignored Words</label>
                                     <input type="text" name="ignored_words" value="${config['ignored_words']}" size="50">
@@ -714,6 +734,7 @@
                                         Split single file albums into multiple tracks
                                         <input type="checkbox" name="cue_split" id="cue_split" value="1" ${config['cue_split']} />
                                     </label>
+                                </div>
                                 <div class="row checkbox left clearfix">
                                     <label title="Freeze the database, so new artists won't be added automatically. Use this if Headphones adds artists because due to wrong snatches. This check is skipped when the folder name is appended with release group ID.">
                                         Freeze database for adding new artist
@@ -767,14 +788,18 @@
                                     </label>
                                 </div>
                                 <div class="row">
-                                    <label>Path to Destination Folder</label>
+                                    <label>
+                                        Destination Directory:
+                                    </label>
                                     <input type="text" name="destination_dir" value="${config['destination_dir']}" size="50">
-                                    <small>e.g. /Users/name/Music/iTunes or /Volumes/share/music</small>
+                                    <small>The directory where Headphones will move file to after post processing, e.g. /Volumes/share/music.</small>
                                 </div>
                                 <div class="row">
-                                    <label>Path to Lossless Destination folder (optional)</label>
+                                    <label>
+                                        Lossless Destination Directory:
+                                    </label>
                                     <input type="text" name="lossless_destination_dir" value="${config['lossless_destination_dir']}" size="50">
-                                    <small>Set this if you have a separate directory for lossless music</small>
+                                    <small>Optional. Set this if you have a separate directory for lossless music.</small>
                                 </div>
                             </fieldset>
                         </td>
@@ -816,7 +841,7 @@
                                         <input type="checkbox" name="prowl_onsnatch" value="1" ${config['prowl_onsnatch']} /><label>Notify on snatch?</label>
                                     </div>
                                     <div class="row">
-                                        <label>Priority (-2,-1,0,1 or 2):</label>
+                                        <label>Priority (-2,-1,0,1 or 2)</label>
                                         <input type="text" name="prowl_priority" value="${config['prowl_priority']}" size="2">
                                     </div>
                                 </div>
@@ -991,11 +1016,11 @@
                                         <input type="checkbox" name="pushover_onsnatch" value="1" ${config['pushover_onsnatch']} /><label>Notify on snatch?</label>
                                     </div>
                                     <div class="row">
-                                        <label>Priority (-1,0, or 1):</label>
+                                        <label>Priority (-1,0, or 1)</label>
                                         <input type="text" name="pushover_priority" value="${config['pushover_priority']}" size="2">
                                     </div>
                                     <div class="row">
-                                        <label>API Token (leave blank to use Headphones default):</label><input type="text" name="pushover_apitoken" value="${config['pushover_apitoken']}" size="50">
+                                        <label>API Token (leave blank to use Headphones default)</label><input type="text" name="pushover_apitoken" value="${config['pushover_apitoken']}" size="50">
                                     </div>
                                 </div>
                             </fieldset>
@@ -1163,7 +1188,7 @@
                             </div>
                             <div class="row">
                                 <div class="row">
-                                    <label>Multi-core count:</label>
+                                    <label>Multi-core count</label>
                                     <input type="text" name="encoder_multicore_count" value="${config['encoder_multicore_count']}" size="7">
                                     <small>Set equal to the number of cores, or 0 for auto</small>
                                 </div>
@@ -1307,15 +1332,15 @@
                                     <label>Automatically mark manually added albums as wanted</label>
                                 </div>
                                 <div class="row">
-                                    <label>Folder Permissions:</label>
+                                    <label>Folder Permissions</label>
                                     <input type="text" name="folder_permissions" value="${config['folder_permissions']}" size="7">
                                 </div>
                                 <div class="row">
-                                    <label>File Permissions:</label>
+                                    <label>File Permissions</label>
                                     <input type="text" name="file_permissions" value="${config['file_permissions']}" size="7">
                                 </div>
                                 <div class="row">
-                                    <label>Cache Size (in MB):</label>
+                                    <label>Cache Size (in MB)</label>
                                     <input type="text" name="cache_sizemb" value="${config['cache_sizemb']}" size="7">
                                 </div>
                             </fieldset>
@@ -1336,11 +1361,11 @@
                                     </select>
                                 </div>
                                 <div class="row">
-                                    <label>Log Directory:</label>
+                                    <label>Log Directory</label>
                                     <input type="text" name="log_dir" value="${config['log_dir']}" size="50">
                                 </div>
                                 <div class="row">
-                                    <label>Cache Directory:</label>
+                                    <label>Cache Directory</label>
                                     <input type="text" name="cache_dir" value="${config['cache_dir']}" size="50">
                                 </div>
                             </fieldset>
@@ -1352,7 +1377,7 @@
                                 </div>
                                 <div id="songkickoptions">
                                   <div class="row">
-                                      <label>API Key:</label>
+                                      <label>API Key</label>
                                       <input type="text" name="songkick_apikey" value="${config['songkick_apikey']}" size="50">
                                   </div>
                                   <div class="row checkbox">
@@ -1360,7 +1385,7 @@
                                   </div>
                                   <div id="songkick_filteroptions">
                                     <div class="row">
-                                      <label>Metro Area ID:</label>
+                                      <label>Metro Area ID</label>
                                       <input type="text" name="songkick_location" value="${config['songkick_location']}" size="10" title="Enter the Metro Area ID, e.g. the ID for London is 24426, this can be found by clicking the link and searching/selecting the city, e.g. London should find http://www.songkick.com/metro_areas/24426-uk-london">
                                         <a target="_blank" href="http://www.songkick.com/developer/location-search">Find Area ID</a>
                                     </div>
@@ -2042,6 +2067,7 @@
                 initConfigCheckbox("#use_omgwtfnzbs");
                 initConfigCheckbox("#use_kat");
                 initConfigCheckbox("#use_piratebay");
+                initConfigCheckbox("#use_oldpiratebay");
                 initConfigCheckbox("#use_mininova");
                 initConfigCheckbox("#use_waffles");
                 initConfigCheckbox("#use_rutracker");

data/interfaces/default/history.html

@@ -10,9 +10,10 @@
 			<a id="menu_link_delete" href="#" onclick="doAjaxCall('clearhistory?type=all',$(this),'table')" data-success="All History cleared"><i class="fa fa-trash-o"></i> Clear All History</a>
 			<a id="menu_link_delete" href="#" onclick="doAjaxCall('clearhistory?type=Processed',$(this),'table')" data-success="All Processed cleared"><i class="fa fa-trash-o"></i> Clear Processed</a>
 			<a id="menu_link_delete" href="#" onclick="doAjaxCall('clearhistory?type=Unprocessed',$(this),'table')" data-success="All Unprocessed cleared"><i class="fa fa-trash-o"></i> Clear Unprocessed</a>
+			<a id="menu_link_delete" href="#" onclick="doAjaxCall('clearhistory?type=Frozen',$(this),'table')" data-success="All Frozen cleared"><i class="fa fa-trash-o"></i> Clear Frozen</a>
 			<a id="menu_link_delete" href="#" onclick="doAjaxCall('clearhistory?type=Snatched',$(this),'table')" data-success="All Snatched cleared"><i class="fa fa-trash-o"></i> Clear Snatched</a>
 		</div>
-	</div>	
+	</div>
 </%def>
 
 <%def name="body()">
@@ -39,9 +40,11 @@
 					grade = 'C'
 				elif item['Status'] == 'Unprocessed':
 					grade = 'X'
+				elif item['Status'] == 'Frozen':
+					grade = 'X'
 				else:
 					grade = 'U'
-					
+
 				fileid = 'unknown'
 				if item['URL'].find('nzb') != -1:
 					fileid = 'nzb'

data/interfaces/default/managemanual.html

@@ -17,7 +17,7 @@
 <%def name="body()">
 	<div class="table_wrapper">
 	<div id="manageheader" class="title">
-		<h1 class="clearfix"><i class="fa fa-music"></i> Manage Manually Changed Albums</h1>
+		<h1 class="clearfix"><i class="fa fa-music"></i> Manage Manually Matched Albums</h1>
 	</div>
 
 	<table class="display" id="artist_table">
@@ -37,30 +37,30 @@
 				old_album_clean = album['AlbumTitle'].replace('&','%26').replace('+', '%2B').replace("'","%27")
 			%>
 			<td id="artist">${album['ArtistName']}<BR>
-				<button id="restore_artist${count_albums}" onClick="restore_Artist(this.id)">(<-) Restore Artist</button>
-				<div id="restore_artist_dialog${count_albums}" title="Restore Artist" style="display:none">
+				<button id="reset_artist${count_albums}" onClick="reset_Artist(this.id)">(<-) Reset Artist</button>
+				<div id="reset_artist_dialog${count_albums}" title="Reset Artist" style="display:none">
 					<table>
-					<tr><td>Are you sure you want to restore Local Artist: ${album['ArtistName']} to unmatched?</td></tr>
+					<tr><td>Are you sure you want to reset Local Artist: ${album['ArtistName']} to unmatched?</td></tr>
 					<tr><td align="right"><BR>
 					%if album['AlbumStatus'] == "Ignored":
-						<button href="#" onclick="doAjaxCall('markManual?action=unignoreArtist&existing_artist=${old_artist_clean}', $(this), 'page');" data-success="Successfully restored ${album['ArtistName']} to unmatched">Restore Artist</button>
+						<button href="#" onclick="doAjaxCall('markManual?action=unignoreArtist&existing_artist=${old_artist_clean}', $(this), 'page');" data-success="Successfully reset ${album['ArtistName']} to unmatched">Reset Artist</button>
 					%elif album['AlbumStatus'] == "Matched":
-						<button href="#" onclick="doAjaxCall('markManual?action=unmatchArtist&existing_artist=${old_artist_clean}', $(this), 'page');" data-success="Successfully restored ${album['ArtistName']} to unmatched">Restore Artist</button>
+						<button href="#" onclick="doAjaxCall('markManual?action=unmatchArtist&existing_artist=${old_artist_clean}', $(this), 'page');" data-success="Successfully restored ${album['ArtistName']} to unmatched">Reset Artist</button>
 					%endif
 					</td></tr>
 					</table>
 				</div>
 			</td>
 			<td id="album">${album['AlbumTitle']}<BR>
-				<button id="restore_album${count_albums}" onClick="restore_Album(this.id)">(<-) Restore Album</button>
-				<div id="restore_album_dialog${count_albums}" title="Restore Album" style="display:none">
+				<button id="reset_album${count_albums}" onClick="reset_Album(this.id)">(<-) Reset Album</button>
+				<div id="reset_album_dialog${count_albums}" title="Reset Album" style="display:none">
 					<table>
-					<tr><td>Are you sure you want to restore Local Album: ${album['AlbumTitle']} to unmatched?</td></tr>
+					<tr><td>Are you sure you want to reset Local Album: ${album['AlbumTitle']} to unmatched?</td></tr>
 					<tr><td align="right"><BR>
 					%if album['AlbumStatus'] == "Ignored":
-						<button href="#" onclick="doAjaxCall('markManual?action=unignoreAlbum&existing_artist=${old_artist_clean}&existing_album=${old_album_clean}', $(this), 'page');" data-success="Successfully restored ${album['AlbumTitle']} to unmatched">Restore Album</button>
+						<button href="#" onclick="doAjaxCall('markManual?action=unignoreAlbum&existing_artist=${old_artist_clean}&existing_album=${old_album_clean}', $(this), 'page');" data-success="Successfully reset ${album['AlbumTitle']} to unmatched">Reset Album</button>
 					%elif album['AlbumStatus'] == "Matched":
-						<button href="#" onclick="doAjaxCall('markManual?action=unmatchAlbum&existing_artist=${old_artist_clean}&existing_album=${old_album_clean}', $(this), 'page');" data-success="Successfully restored ${album['AlbumTitle']} to unmatched">Restore Album</button>
+						<button href="#" onclick="doAjaxCall('markManual?action=unmatchAlbum&existing_artist=${old_artist_clean}&existing_album=${old_album_clean}', $(this), 'page');" data-success="Successfully reset ${album['AlbumTitle']} to unmatched">Reset Album</button>
 					%endif
 					</td></tr>
 					</table>
@@ -107,15 +107,15 @@
 		initActions();
 	});
 
-	function restore_Artist(clicked_id) {
-		n=clicked_id.replace("restore_artist","");
-		$("#restore_artist_dialog"+n).dialog();
+	function reset_Artist(clicked_id) {
+		n=clicked_id.replace("reset_artist","");
+		$("#reset_artist_dialog"+n).dialog();
 		return false;
 	}
 
-	function restore_Album(clicked_id) {
-		n=clicked_id.replace("restore_album","");
-		$("#restore_album_dialog"+n).dialog();
+	function reset_Album(clicked_id) {
+		n=clicked_id.replace("reset_album","");
+		$("#reset_album_dialog"+n).dialog();
 		return false;
 	}
 

headphones/__init__.py

@@ -23,6 +23,7 @@ import threading
 import webbrowser
 import sqlite3
 import cherrypy
+import datetime
 
 from apscheduler.schedulers.background import BackgroundScheduler
 from apscheduler.triggers.interval import IntervalTrigger
@@ -64,6 +65,7 @@ CREATEPID = False
 PIDFILE = None
 
 SCHED = BackgroundScheduler()
+SCHED_LOCK = threading.Lock()
 
 INIT_LOCK = threading.Lock()
 _INITIALIZED = False
@@ -125,12 +127,15 @@ def initialize(config_file):
             try:
                 os.makedirs(CONFIG.LOG_DIR)
             except OSError:
-                if VERBOSE:
-                    sys.stderr.write(
-                        'Unable to create the log directory. Logging to screen only.\n')
+                CONFIG.LOG_DIR = None
+
+                if not QUIET:
+                    sys.stderr.write("Unable to create the log directory. " \
+                        "Logging to screen only.\n")
 
         # Start the logger, disable console if needed
-        logger.initLogger(console=not QUIET, verbose=VERBOSE)
+        logger.initLogger(console=not QUIET, log_dir=CONFIG.LOG_DIR,
+            verbose=VERBOSE)
 
         if not CONFIG.CACHE_DIR:
             # Put the cache dir in the data dir for now
@@ -138,13 +143,12 @@ def initialize(config_file):
         if not os.path.exists(CONFIG.CACHE_DIR):
             try:
                 os.makedirs(CONFIG.CACHE_DIR)
-            except OSError:
-                logger.error(
-                    'Could not create cache dir. Check permissions of datadir: %s', DATA_DIR)
+            except OSError as e:
+                logger.error("Could not create cache dir '%s': %s", DATA_DIR, e)
 
         # Sanity check for search interval. Set it to at least 6 hours
-        if CONFIG.SEARCH_INTERVAL < 360:
-            logger.info("Search interval too low. Resetting to 6 hour minimum")
+        if CONFIG.SEARCH_INTERVAL and CONFIG.SEARCH_INTERVAL < 360:
+            logger.info("Search interval too low. Resetting to 6 hour minimum.")
             CONFIG.SEARCH_INTERVAL = 360
 
         # Initialize the database
@@ -154,10 +158,23 @@ def initialize(config_file):
         except Exception as e:
             logger.error("Can't connect to the database: %s", e)
 
-        # Get the currently installed version - returns None, 'win32' or the git hash
-        # Also sets INSTALL_TYPE variable to 'win', 'git' or 'source'
+        # Get the currently installed version. Returns None, 'win32' or the git
+        # hash.
         CURRENT_VERSION, CONFIG.GIT_BRANCH = versioncheck.getVersion()
 
+        # Write current version to a file, so we know which version did work.
+        # This allowes one to restore to that version. The idea is that if we
+        # arrive here, most parts of Headphones seem to work.
+        if CURRENT_VERSION:
+            version_lock_file = os.path.join(DATA_DIR, "version.lock")
+
+            try:
+                with open(version_lock_file, "w") as fp:
+                    fp.write(CURRENT_VERSION)
+            except IOError as e:
+                logger.error("Unable to write current version to file '%s': %s",
+                             version_lock_file, e)
+
         # Check for new versions
         if CONFIG.CHECK_GITHUB_ON_STARTUP:
             try:
@@ -244,36 +261,83 @@ def launch_browser(host, port, root):
         logger.error('Could not launch browser: %s', e)
 
 
+def initialize_scheduler():
+    """
+    Start the scheduled background tasks. Re-schedule if interval settings changed.
+    """
+
+    from headphones import updater, searcher, librarysync, postprocessor, \
+        torrentfinished
+
+    with SCHED_LOCK:
+
+        # Check if scheduler should be started
+        start_jobs = not len(SCHED.get_jobs())
+
+        # Regular jobs
+        minutes = CONFIG.SEARCH_INTERVAL
+        schedule_job(searcher.searchforalbum, 'Search for Wanted', hours=0, minutes=minutes)
+
+        minutes = CONFIG.DOWNLOAD_SCAN_INTERVAL
+        schedule_job(postprocessor.checkFolder, 'Download Scan', hours=0, minutes=minutes)
+
+        hours = CONFIG.LIBRARYSCAN_INTERVAL
+        schedule_job(librarysync.libraryScan, 'Library Scan', hours=hours, minutes=0)
+
+        hours = CONFIG.UPDATE_DB_INTERVAL
+        schedule_job(updater.dbUpdate, 'MusicBrainz Update', hours=hours, minutes=0)
+
+        #Update check
+        if CONFIG.CHECK_GITHUB_INTERVAL:
+            minutes = CONFIG.CHECK_GITHUB_INTERVAL
+        else:
+            minutes = 0
+        schedule_job(versioncheck.checkGithub, 'Check GitHub for updates', hours=0, minutes=minutes)
+
+        # Remove Torrent + data if Post Processed and finished Seeding
+        minutes = CONFIG.TORRENT_REMOVAL_INTERVAL
+        schedule_job(torrentfinished.checkTorrentFinished, 'Torrent removal check', hours=0, minutes=minutes)
+
+        # Start scheduler
+        if start_jobs and len(SCHED.get_jobs()):
+            try:
+                SCHED.start()
+            except Exception as e:
+                logger.info(e)
+
+        # Debug
+        #SCHED.print_jobs()
+
+
+def schedule_job(function, name, hours=0, minutes=0):
+    """
+    Start scheduled job if starting or restarting headphones.
+    Reschedule job if Interval Settings have changed.
+    Remove job if if Interval Settings changed to 0
+
+    """
+
+    job = SCHED.get_job(name)
+    if job:
+        if hours == 0 and minutes == 0:
+            SCHED.remove_job(name)
+            logger.info("Removed background task: %s", name)
+        elif job.trigger.interval != datetime.timedelta(hours=hours, minutes=minutes):
+            SCHED.reschedule_job(name, trigger=IntervalTrigger(
+                hours=hours, minutes=minutes))
+            logger.info("Re-scheduled background task: %s", name)
+    elif hours > 0 or minutes > 0:
+        SCHED.add_job(function, id=name, trigger=IntervalTrigger(
+            hours=hours, minutes=minutes))
+        logger.info("Scheduled background task: %s", name)
+
+
 def start():
 
     global started
 
     if _INITIALIZED:
-
-        # Start our scheduled background tasks
-        from headphones import updater, searcher, librarysync, postprocessor, torrentfinished
-        SCHED.add_job(updater.dbUpdate, trigger=IntervalTrigger(
-            hours=CONFIG.UPDATE_DB_INTERVAL))
-        SCHED.add_job(searcher.searchforalbum, trigger=IntervalTrigger(
-            minutes=CONFIG.SEARCH_INTERVAL))
-        SCHED.add_job(librarysync.libraryScan, trigger=IntervalTrigger(
-            hours=CONFIG.LIBRARYSCAN_INTERVAL))
-
-        if CONFIG.CHECK_GITHUB:
-            SCHED.add_job(versioncheck.checkGithub, trigger=IntervalTrigger(
-                minutes=CONFIG.CHECK_GITHUB_INTERVAL))
-
-        if CONFIG.DOWNLOAD_SCAN_INTERVAL > 0:
-            SCHED.add_job(postprocessor.checkFolder, trigger=IntervalTrigger(
-                minutes=CONFIG.DOWNLOAD_SCAN_INTERVAL))
-
-        # Remove Torrent + data if Post Processed and finished Seeding
-        if CONFIG.TORRENT_REMOVAL_INTERVAL > 0:
-            SCHED.add_job(torrentfinished.checkTorrentFinished, trigger=IntervalTrigger(
-                minutes=CONFIG.TORRENT_REMOVAL_INTERVAL))
-
-        SCHED.start()
-
+        initialize_scheduler()
         started = True
 
 

headphones/api.py

@@ -13,16 +13,17 @@
 #  You should have received a copy of the GNU General Public License
 #  along with Headphones.  If not, see <http://www.gnu.org/licenses/>.
 
-from headphones import db, mb, importer, searcher, cache, postprocessor, versioncheck, logger
+from headphones import db, mb, updater, importer, searcher, cache, postprocessor, versioncheck, logger
 
 import headphones
 import json
 
-cmd_list = ['getIndex', 'getArtist', 'getAlbum', 'getUpcoming', 'getWanted', 'getSimilar', 'getHistory', 'getLogs',
+cmd_list = ['getIndex', 'getArtist', 'getAlbum', 'getUpcoming', 'getWanted', 'getSnatched', 'getSimilar', 'getHistory', 'getLogs',
             'findArtist', 'findAlbum', 'addArtist', 'delArtist', 'pauseArtist', 'resumeArtist', 'refreshArtist',
-            'addAlbum', 'queueAlbum', 'unqueueAlbum', 'forceSearch', 'forceProcess', 'getVersion', 'checkGithub',
-            'shutdown', 'restart', 'update', 'getArtistArt', 'getAlbumArt', 'getArtistInfo', 'getAlbumInfo',
-            'getArtistThumb', 'getAlbumThumb', 'choose_specific_download', 'download_specific_release']
+            'addAlbum', 'queueAlbum', 'unqueueAlbum', 'forceSearch', 'forceProcess', 'forceActiveArtistsUpdate', 
+            'getVersion', 'checkGithub','shutdown', 'restart', 'update', 'getArtistArt', 'getAlbumArt', 
+            'getArtistInfo', 'getAlbumInfo', 'getArtistThumb', 'getAlbumThumb', 
+            'choose_specific_download', 'download_specific_release']
 
 
 class Api(object):
@@ -165,6 +166,11 @@ class Api(object):
             "SELECT * from albums WHERE Status='Wanted'")
         return
 
+    def _getSnatched(self, **kwargs):
+        self.data = self._dic_from_query(
+            "SELECT * from albums WHERE Status='Snatched'")
+        return
+        
     def _getSimilar(self, **kwargs):
         self.data = self._dic_from_query('SELECT * from lastfmcloud')
         return
@@ -321,6 +327,9 @@ class Api(object):
             self.dir = kwargs['dir']
         postprocessor.forcePostProcess(self.dir)
 
+    def _forceActiveArtistsUpdate(self, **kwargs):
+        updater.dbUpdate()
+
     def _getVersion(self, **kwargs):
         self.data = {
             'git_path': headphones.CONFIG.GIT_PATH,

headphones/cache.py

@@ -23,20 +23,22 @@ LASTFM_API_KEY = "690e1ed3bc00bc91804cd8f7fe5ed6d4"
 
 class Cache(object):
     """
-    This class deals with getting, storing and serving up artwork (album
-    art, artist images, etc) and info/descriptions (album info, artist descrptions)
-    to and from the cache folder. This can be called from within a web interface,
-    for example, using the helper functions getInfo(id) and getArtwork(id), to utilize the cached
-    images rather than having to retrieve them every time the page is reloaded.
+    This class deals with getting, storing and serving up artwork (album art,
+    artist images, etc) and info/descriptions (album info, artist descrptions)
+    to and from the cache folder. This can be called from within a web
+    interface. For example, using the helper functions `getInfo(id)` and
+    `getArtwork(id)`, to utilize the cached images rather than having to
+    retrieve them every time the page is reloaded.
 
-    So you can call cache.getArtwork(id) which will return an absolute path
-    to the image file on the local machine, or if the cache directory
-    doesn't exist, or can not be written to, it will return a url to the image.
+    You can call `getArtwork(id)` which will return an absolute path to the
+    image file on the local machine, or if the cache directory doesn't exist,
+    or can not be written to, it will return a url to the image.
 
-    Call cache.getInfo(id) to grab the artist/album info; will return the text description
+    Call `getInfo(id)` to grab the artist/album info. This will return the
+    text description.
 
-    The basic format for art in the cache is <musicbrainzid>.<date>.<ext>
-    and for info it is <musicbrainzid>.<date>.txt
+    The basic format for art in the cache is `<musicbrainzid>.<date>.<ext>`
+    and for info it is `<musicbrainzid>.<date>.txt`
     """
 
     path_to_art_cache = os.path.join(headphones.CONFIG.CACHE_DIR, 'artwork')
@@ -197,10 +199,10 @@ class Cache(object):
             return info_dict
 
     def get_image_links(self, ArtistID=None, AlbumID=None):
-        '''
+        """
         Here we're just going to open up the last.fm url, grab the image links and return them
         Won't save any image urls, or save the artwork in the cache. Useful for search results, etc.
-        '''
+        """
         if ArtistID:
 
             self.id_type = 'artist'
@@ -271,9 +273,10 @@ class Cache(object):
                     logger.warn('Error deleting file from the cache: %s', thumb_file)
 
     def _update_cache(self):
-        '''
+        """
         Since we call the same url for both info and artwork, we'll update both at the same time
-        '''
+        """
+
         myDB = db.DBConnection()
 
         # Since lastfm uses release ids rather than release group ids for albums, we have to do a artist + album search for albums
@@ -306,7 +309,6 @@ class Cache(object):
                 logger.debug('No artist thumbnail image found')
 
         else:
-
             dbalbum = myDB.action('SELECT ArtistName, AlbumTitle, ReleaseID FROM albums WHERE AlbumID=?', [self.id]).fetchone()
             if dbalbum['ReleaseID'] != self.id:
                 data = lastfm.request_lastfm("album.getinfo", mbid=dbalbum['ReleaseID'], api_key=LASTFM_API_KEY)
@@ -339,7 +341,8 @@ class Cache(object):
             if not thumb_url:
                 logger.debug('No album thumbnail image found')
 
-        #Save the content & summary to the database no matter what if we've opened up the url
+        # Save the content & summary to the database no matter what if we've
+        # opened up the url
         if self.id_type == 'artist':
             controlValueDict = {"ArtistID": self.id}
         else:
@@ -365,7 +368,8 @@ class Cache(object):
             else:
                 myDB.action('UPDATE albums SET ThumbURL=? WHERE AlbumID=?', [thumb_url, self.id])
 
-        # Should we grab the artwork here if we're just grabbing thumbs or info?? Probably not since the files can be quite big
+        # Should we grab the artwork here if we're just grabbing thumbs or
+        # info? Probably not since the files can be quite big
         if image_url and self.query_type == 'artwork':
             artwork = request.request_content(image_url, timeout=20)
 
@@ -374,12 +378,13 @@ class Cache(object):
                 if not os.path.isdir(self.path_to_art_cache):
                     try:
                         os.makedirs(self.path_to_art_cache)
-                    except Exception as e:
+                        os.chmod(self.path_to_art_cache, int(headphones.CONFIG.FOLDER_PERMISSIONS, 8))
+                    except OSError as e:
                         logger.error('Unable to create artwork cache dir. Error: %s', e)
                         self.artwork_errors = True
                         self.artwork_url = image_url
 
-                #Delete the old stuff
+                # Delete the old stuff
                 for artwork_file in self.artwork_files:
                     try:
                         os.remove(artwork_file)
@@ -392,12 +397,15 @@ class Cache(object):
                 try:
                     with open(artwork_path, 'wb') as f:
                         f.write(artwork)
-                except IOError as e:
+
+                    os.chmod(artwork_path, int(headphones.CONFIG.FILE_PERMISSIONS, 8))
+                except (OSError, IOError) as e:
                     logger.error('Unable to write to the cache dir: %s', e)
                     self.artwork_errors = True
                     self.artwork_url = image_url
 
-        # Grab the thumbnail as well if we're getting the full artwork (as long as it's missing/outdated
+        # Grab the thumbnail as well if we're getting the full artwork (as long
+        # as it's missing/outdated.
         if thumb_url and self.query_type in ['thumb', 'artwork'] and not (self.thumb_files and self._is_current(self.thumb_files[0])):
             artwork = request.request_content(thumb_url, timeout=20)
 
@@ -406,16 +414,17 @@ class Cache(object):
                 if not os.path.isdir(self.path_to_art_cache):
                     try:
                         os.makedirs(self.path_to_art_cache)
-                    except Exception as e:
+                        os.chmod(self.path_to_art_cache, int(headphones.CONFIG.FOLDER_PERMISSIONS, 8))
+                    except OSError as e:
                         logger.error('Unable to create artwork cache dir. Error: %s' + e)
                         self.thumb_errors = True
                         self.thumb_url = thumb_url
 
-                #Delete the old stuff
+                # Delete the old stuff
                 for thumb_file in self.thumb_files:
                     try:
                         os.remove(thumb_file)
-                    except Exception as e:
+                    except OSError as e:
                         logger.error('Error deleting file from the cache: %s', thumb_file)
 
                 ext = os.path.splitext(image_url)[1]
@@ -424,14 +433,15 @@ class Cache(object):
                 try:
                     with open(thumb_path, 'wb') as f:
                         f.write(artwork)
-                except IOError as e:
+
+                    os.chmod(thumb_path, int(headphones.CONFIG.FILE_PERMISSIONS, 8))
+                except (OSError, IOError) as e:
                     logger.error('Unable to write to the cache dir: %s', e)
                     self.thumb_errors = True
                     self.thumb_url = image_url
 
 
 def getArtwork(ArtistID=None, AlbumID=None):
-
     c = Cache()
     artwork_path = c.get_artwork_from_cache(ArtistID, AlbumID)
 
@@ -446,7 +456,6 @@ def getArtwork(ArtistID=None, AlbumID=None):
 
 
 def getThumb(ArtistID=None, AlbumID=None):
-
     c = Cache()
     artwork_path = c.get_thumb_from_cache(ArtistID, AlbumID)
 
@@ -461,16 +470,13 @@ def getThumb(ArtistID=None, AlbumID=None):
 
 
 def getInfo(ArtistID=None, AlbumID=None):
-
     c = Cache()
-
     info_dict = c.get_info_from_cache(ArtistID, AlbumID)
 
     return info_dict
 
 
 def getImageLinks(ArtistID=None, AlbumID=None):
-
     c = Cache()
     image_links = c.get_image_links(ArtistID, AlbumID)
 

headphones/config.py

@@ -42,6 +42,8 @@ _CONFIG_DEFINITIONS = {
     'CONFIG_VERSION': (str, 'General', '0'),
     'CORRECT_METADATA': (int, 'General', 0),
     'CUE_SPLIT': (int, 'General', 1),
+    'CUE_SPLIT_FLAC_PATH': (str, 'General', ''),
+    'CUE_SPLIT_SHNTOOL_PATH': (str, 'General', ''),
     'CUSTOMHOST': (str, 'General', 'localhost'),
     'CUSTOMPORT': (int, 'General', 5000),
     'CUSTOMSLEEP': (int, 'General', 1),
@@ -91,6 +93,8 @@ _CONFIG_DEFINITIONS = {
     'HTTP_ROOT': (str, 'General', '/'),
     'HTTP_USERNAME': (str, 'General', ''),
     'IGNORED_WORDS': (str, 'General', ''),
+    'IGNORED_FOLDERS': (list, 'Advanced', []),
+    'IGNORED_FILES': (list, 'Advanced', []),
     'INCLUDE_EXTRAS': (int, 'General', 0),
     'INTERFACE': (str, 'General', 'default'),
     'JOURNAL_MODE': (str, 'Advanced', 'wal'),
@@ -146,6 +150,9 @@ _CONFIG_DEFINITIONS = {
     'PIRATEBAY': (int, 'Piratebay', 0),
     'PIRATEBAY_PROXY_URL': (str, 'Piratebay', ''),
     'PIRATEBAY_RATIO': (str, 'Piratebay', ''),
+    'OLDPIRATEBAY': (int, 'Old Piratebay', 0),
+    'OLDPIRATEBAY_URL': (str, 'Old Piratebay', ''),
+    'OLDPIRATEBAY_RATIO': (str, 'Old Piratebay', ''),
     'PLEX_CLIENT_HOST': (str, 'Plex', ''),
     'PLEX_ENABLED': (int, 'Plex', 0),
     'PLEX_NOTIFY': (int, 'Plex', 0),

headphones/cuesplit.py

@@ -72,11 +72,13 @@ CUE_META = None
 
 
 def check_splitter(command):
-    '''Check xld or shntools installed'''
+    '''Check xld or shntool installed'''
     try:
         env = os.environ.copy()
         if 'xld' in command:
             env['PATH'] += os.pathsep + '/Applications'
+        elif headphones.CONFIG.CUE_SPLIT_FLAC_PATH:
+            command = os.path.join(headphones.CONFIG.CUE_SPLIT_SHNTOOL_PATH, 'shntool')
         devnull = open(os.devnull)
         subprocess.Popen([command], stdout=devnull, stderr=devnull, env=env).communicate()
     except OSError as e:
@@ -103,6 +105,8 @@ def split_baby(split_file, split_cmd):
     env = os.environ.copy()
     if 'xld' in split_cmd:
         env['PATH'] += os.pathsep + '/Applications'
+    elif headphones.CONFIG.CUE_SPLIT_FLAC_PATH:
+        env['PATH'] += os.pathsep + headphones.CONFIG.CUE_SPLIT_FLAC_PATH
 
     process = subprocess.Popen(split_cmd, startupinfo=startupinfo,
 
@@ -424,7 +428,7 @@ class CueFile(File):
             elif t >= 1:
                 t_index = self.tracks[t]['index']
                 content += t_index[1]
-                if (t < len(self.tracks) - 1):
+                if t < (len(self.tracks) - 1):
                     content += '\n'
         return content
 
@@ -546,8 +550,8 @@ def split(albumpath):
     for _cue in base_dir.filter('CueFile'):
         for _wave in base_dir.filter('WaveFile'):
             if _cue.header['file'] == _wave.name:
-                logger.info('CUE Sheet found: {0}'.format(_cue.name))
-                logger.info('Music file found: {0}'.format(_wave.name))
+                logger.info('CUE Sheet found: %s', _cue.name)
+                logger.info('Music file found: %s', _wave.name)
                 cue = _cue
                 wave = _wave
     # if no perfect match found then try without extensions
@@ -556,9 +560,9 @@ def split(albumpath):
         for _cue in base_dir.filter('CueFile'):
             for _wave in base_dir.filter('WaveFile'):
                 if ''.join(os.path.splitext(_cue.header['file'])[:-1]) == _wave.name_name:
-                    logger.info('Possible CUE Sheet found: {0}'.format(_cue.name))
-                    logger.info('CUE Sheet refers music file: {0}'.format(_cue.header['file']))
-                    logger.info('Possible Music file found: {0}'.format(_wave.name))
+                    logger.info('Possible CUE Sheet found: %s', _cue.name)
+                    logger.info('CUE Sheet refers music file: %s', _cue.header['file'])
+                    logger.info('Possible Music file found: %s', _wave.name)
                     cue = _cue
                     wave = _wave
                     cue.header['file'] = wave.name
@@ -588,9 +592,9 @@ def split(albumpath):
             splitter = 'shntool'
 
     if splitter == 'shntool' and not check_splitter(splitter):
-            raise ValueError('Command not found, ensure shntools with FLAC or xld (OS X) installed')
+            raise ValueError('Command not found, ensure shntool with FLAC or xld (OS X) installed')
 
-    # Determine if file can be split (only flac allowed for shntools)
+    # Determine if file can be split (only flac allowed for shntool)
     if 'xld' in splitter and wave.name_ext not in WAVE_FILE_TYPE_BY_EXTENSION.keys() or \
             wave.type not in SHNTOOL_COMPATIBLE:
         raise ValueError('Cannot split, audio file has unsupported extension')
@@ -627,7 +631,11 @@ def split(albumpath):
         with open(SPLIT_FILE_NAME, mode='w') as split_file:
             split_file.write(cue.breakpoints())
 
-        cmd = ['shntool']
+        if headphones.CONFIG.CUE_SPLIT_SHNTOOL_PATH:
+            cmd = [os.path.join(headphones.CONFIG.CUE_SPLIT_SHNTOOL_PATH, 'shntool')]
+        else:
+            cmd = ['shntool']
+
         cmd.extend(['split'])
         cmd.extend(['-f'])
         cmd.extend([SPLIT_FILE_NAME])
@@ -641,14 +649,14 @@ def split(albumpath):
         # tag FLAC files
         if split and CUE_META.count_tracks() == len(base_dir.tracks(ext='.flac', split=True)):
             for t in base_dir.tracks(ext='.flac', split=True):
-                logger.info('Tagging {0}...'.format(t.name))
+                logger.info('Tagging %s...', t.name)
                 t.tag()
 
         # rename FLAC files
         if split and CUE_META.count_tracks() == len(base_dir.tracks(ext='.flac', split=True)):
             for t in base_dir.tracks(ext='.flac', split=True):
                 if t.name != t.filename():
-                    logger.info('Renaming {0} to {1}...'.format(t.name, t.filename()))
+                    logger.info('Renaming %s to %s...', t.name, t.filename())
                     os.rename(t.name, t.filename())
 
         os.remove(ALBUM_META_FILE_NAME)

headphones/db.py

@@ -96,11 +96,16 @@ class DBConnection:
 
         genParams = lambda myDict: [x + " = ?" for x in myDict.keys()]
 
-        query = "UPDATE " + tableName + " SET " + ", ".join(genParams(valueDict)) + " WHERE " + " AND ".join(genParams(keyDict))
+        update_query = "UPDATE " + tableName + " SET " + ", ".join(genParams(valueDict)) + " WHERE " + " AND ".join(genParams(keyDict))
 
-        self.action(query, valueDict.values() + keyDict.values())
+        self.action(update_query, valueDict.values() + keyDict.values())
 
         if self.connection.total_changes == changesBefore:
-            query = "INSERT INTO " + tableName + " (" + ", ".join(valueDict.keys() + keyDict.keys()) + ")" + \
-                        " VALUES (" + ", ".join(["?"] * len(valueDict.keys() + keyDict.keys())) + ")"
-            self.action(query, valueDict.values() + keyDict.values())
+            insert_query = (
+                "INSERT INTO " + tableName + " (" + ", ".join(valueDict.keys() + keyDict.keys()) + ")" +
+                " VALUES (" + ", ".join(["?"] * len(valueDict.keys() + keyDict.keys())) + ")"
+            )
+            try:
+                self.action(insert_query, valueDict.values() + keyDict.values())
+            except sqlite3.IntegrityError:
+                logger.info('Queries failed: %s and %s', update_query, insert_query)

headphones/getXldProfile.py

@@ -1,32 +1,31 @@
 import os.path
-import plistlib
-import sys
-import xml.parsers.expat as expat
+import biplist
 from headphones import logger
 
 
 def getXldProfile(xldProfile):
+
     xldProfileNotFound = xldProfile
-    expandedPath = os.path.expanduser('~/Library/Preferences/jp.tmkk.XLD.plist')
-    try:
-        preferences = plistlib.Plist.fromFile(expandedPath)
-    except (expat.ExpatError):
-        os.system("/usr/bin/plutil -convert xml1 %s" % expandedPath)
-        try:
-            preferences = plistlib.Plist.fromFile(expandedPath)
-        except (ImportError):
-            os.system("/usr/bin/plutil -convert binary1 %s" % expandedPath)
-            logger.info('The plist at "%s" has a date in it, and therefore is not useable.', expandedPath)
-            return(xldProfileNotFound, None, None)
-    except (ImportError):
-        logger.info('The plist at "%s" has a date in it, and therefore is not useable.', expandedPath)
-    except:
-        logger.info('Unexpected error: %s', sys.exc_info()[0])
+
+    expanded = os.path.expanduser('~/Library/Preferences/jp.tmkk.XLD.plist')
+    if not os.path.isfile(expanded):
+        logger.warn("Could not find xld preferences at: %s", expanded)
         return(xldProfileNotFound, None, None)
 
-    xldProfile = xldProfile.lower()
-    profiles = preferences.get('Profiles')
+    # Get xld preferences plist
+    try:
+        preferences = biplist.readPlist(expanded)
+    except (biplist.InvalidPlistException, biplist.NotBinaryPlistException), e:
+        logger.error("Error reading xld preferences plist: %s", e)
+        return(xldProfileNotFound, None, None)
 
+    if not isinstance(preferences, dict):
+        logger.error("Error reading xld preferences plist, not a dict: %r", preferences)
+        return(xldProfileNotFound, None, None)
+
+    profiles = preferences.get('Profiles', []) # pylint:disable=E1103
+
+    xldProfile = xldProfile.lower()
     for profile in profiles:
 
         profilename = profile.get('XLDProfileManager_ProfileName')

headphones/helpers.py

@@ -13,17 +13,19 @@
 #  You should have received a copy of the GNU General Public License
 #  along with Headphones.  If not, see <http://www.gnu.org/licenses/>.
 
-import os
-import re
-import time
-import shutil
-import datetime
-import headphones
-import unicodedata
-import sys
+from beets.mediafile import MediaFile, FileTypeError, UnreadableFileError
 
 from operator import itemgetter
-from beets.mediafile import MediaFile, FileTypeError, UnreadableFileError
+
+import unicodedata
+import headphones
+import datetime
+import fnmatch
+import shutil
+import time
+import sys
+import re
+import os
 
 # Modified from https://github.com/Verrus/beets-plugin-featInTitle
 RE_FEATURING = re.compile(r"[fF]t\.|[fF]eaturing|[fF]eat\.|\b[wW]ith\b|&|vs\.")
@@ -65,7 +67,8 @@ def latinToAscii(unicrap):
     """
     From couch potato
     """
-    xlate = {0xc0: 'A', 0xc1: 'A', 0xc2: 'A', 0xc3: 'A', 0xc4: 'A', 0xc5: 'A',
+    xlate = {
+        0xc0: 'A', 0xc1: 'A', 0xc2: 'A', 0xc3: 'A', 0xc4: 'A', 0xc5: 'A',
         0xc6: 'Ae', 0xc7: 'C',
         0xc8: 'E', 0xc9: 'E', 0xca: 'E', 0xcb: 'E', 0x86: 'e',
         0xcc: 'I', 0xcd: 'I', 0xce: 'I', 0xcf: 'I',
@@ -90,7 +93,7 @@ def latinToAscii(unicrap):
         0xb9: '{^1}', 0xba: '{^o}', 0xbb: '>>',
         0xbc: '{1/4}', 0xbd: '{1/2}', 0xbe: '{3/4}', 0xbf: '?',
         0xd7: '*', 0xf7: '/'
-        }
+    }
 
     r = ''
     for i in unicrap:
@@ -168,13 +171,20 @@ def mb_to_bytes(mb_str):
 def piratesize(size):
     split = size.split(" ")
     factor = float(split[0])
-    unit = split[1]
+    unit = split[1].upper()
+    
     if unit == 'MiB':
         size = factor * 1048576
+    elif unit == 'MB':
+        size = factor * 1000000
     elif unit == 'GiB':
         size = factor * 1073741824
+    elif unit == 'GB':
+        size = factor * 1000000000
     elif unit == 'KiB':
         size = factor * 1024
+    elif unit == 'KB':
+        size = factor * 1000
     elif unit == "B":
         size = factor
     else:
@@ -329,6 +339,44 @@ def expand_subfolders(f):
     return media_folders
 
 
+def path_match_patterns(path, patterns):
+    """
+    Check if a path matches one or more patterns. The whole path will be
+    matched be matched against the patterns.
+    """
+
+    for pattern in patterns:
+        if fnmatch.fnmatch(path, pattern):
+            return True
+
+    # No match
+    return False
+
+
+def path_filter_patterns(paths, patterns, root=None):
+    """
+    Scan for ignored paths based on glob patterns. Note that the whole path
+    will be matched, therefore paths should only contain the relative paths.
+
+    The root is optional, and only used for producing meaningful debug info.
+    """
+
+    from headphones import logger
+
+    ignored = 0
+
+    for path in paths[:]:
+        if path_match_patterns(path, patterns):
+            logger.debug("Path ignored by pattern: %s",
+                os.path.join(root or "", path))
+
+            ignored += 1
+            paths.remove(path)
+
+    # Return number of ignored paths
+    return ignored
+
+
 def extract_data(s):
 
     s = s.replace('_', ' ')
@@ -473,8 +521,8 @@ def get_downloaded_track_list(albumpath):
 
 def preserve_torrent_direcory(albumpath):
     """
-     Copy torrent directory to headphones-modified to keep files for seeding.
-     """
+    Copy torrent directory to headphones-modified to keep files for seeding.
+    """
     from headphones import logger
     new_folder = os.path.join(albumpath, 'headphones-modified'.encode(headphones.SYS_ENCODING, 'replace'))
     logger.info("Copying files to 'headphones-modified' subfolder to preserve downloaded files for seeding")
@@ -604,6 +652,43 @@ def smartMove(src, dest, delete=True):
     except Exception as e:
         logger.warn('Error moving file %s: %s', filename.decode(headphones.SYS_ENCODING, 'replace'), e)
 
+def walk_directory(basedir, followlinks=True):
+    """
+    Enhanced version of 'os.walk' where symlink directores are traversed, but
+    with care. In case a folder is already processed, don't traverse it again.
+    """
+
+    import logger
+
+    # Add the base path, because symlinks poiting to the basedir should not be
+    # traversed again.
+    traversed = [os.path.abspath(basedir)]
+
+    def _inner(root, directories, files):
+        for directory in directories:
+            path = os.path.join(root, directory)
+
+            if followlinks and os.path.islink(path):
+                real_path = os.path.abspath(os.readlink(path))
+
+                if real_path in traversed:
+                    logger.debug("Skipping '%s' since it is a symlink to "\
+                        "'%s', which is already visited.", path, real_path)
+                else:
+                    traversed.append(real_path)
+
+                    for args in os.walk(real_path):
+                        for result in _inner(*args):
+                            yield result
+
+        # Pass on actual result
+        yield root, directories, files
+
+    # Start traversing
+    for args in os.walk(basedir):
+        for result in _inner(*args):
+            yield result
+
 #########################
 #Sab renaming functions #
 #########################
@@ -660,44 +745,41 @@ def sab_sanitize_foldername(name):
 
     return name
 
-
 def split_string(mystring, splitvar=','):
     mylist = []
     for each_word in mystring.split(splitvar):
         mylist.append(each_word.strip())
     return mylist
 
-
 def create_https_certificates(ssl_cert, ssl_key):
     """
-    Stolen from SickBeard (http://github.com/midgetspy/Sick-Beard):
-    Create self-signed HTTPS certificares and store in paths 'ssl_cert' and 'ssl_key'
+    Create a pair of self-signed HTTPS certificares and store in them in
+    'ssl_cert' and 'ssl_key'. Method assumes pyOpenSSL is installed.
+
+    This code is stolen from SickBeard (http://github.com/midgetspy/Sick-Beard).
     """
+
     from headphones import logger
 
-    try:
-        from OpenSSL import crypto
-        from certgen import createKeyPair, createCertRequest, createCertificate, TYPE_RSA, serial
-    except:
-        logger.warn("pyOpenSSL module missing, please install to enable HTTPS")
-        return False
+    from OpenSSL import crypto
+    from certgen import createKeyPair, createCertRequest, createCertificate, \
+        TYPE_RSA, serial
 
     # Create the CA Certificate
-    cakey = createKeyPair(TYPE_RSA, 1024)
-    careq = createCertRequest(cakey, CN='Certificate Authority')
+    cakey = createKeyPair(TYPE_RSA, 2048)
+    careq = createCertRequest(cakey, CN="Certificate Authority")
     cacert = createCertificate(careq, (careq, cakey), serial, (0, 60 * 60 * 24 * 365 * 10)) # ten years
 
-    cname = 'Headphones'
-    pkey = createKeyPair(TYPE_RSA, 1024)
-    req = createCertRequest(pkey, CN=cname)
+    pkey = createKeyPair(TYPE_RSA, 2048)
+    req = createCertRequest(pkey, CN="Headphones")
     cert = createCertificate(req, (cacert, cakey), serial, (0, 60 * 60 * 24 * 365 * 10)) # ten years
 
     # Save the key and certificate to disk
     try:
-        with open(ssl_key, 'w') as f:
-            f.write(crypto.dump_privatekey(crypto.FILETYPE_PEM, pkey))
-        with open(ssl_cert, 'w') as f:
-            f.write(crypto.dump_certificate(crypto.FILETYPE_PEM, cert))
+        with open(ssl_key, "w") as fp:
+            fp.write(crypto.dump_privatekey(crypto.FILETYPE_PEM, pkey))
+        with open(ssl_cert, "w") as fp:
+            fp.write(crypto.dump_certificate(crypto.FILETYPE_PEM, cert))
     except IOError as e:
         logger.error("Error creating SSL key and certificate: %s", e)
         return False

headphones/importer.py

@@ -48,7 +48,7 @@ def artistlist_to_mbids(artistlist, forced=False):
 
     for artist in artistlist:
 
-        if not artist and not (artist == ' '):
+        if not artist and artist != ' ':
             continue
 
         # If adding artists through Manage New Artists, they're coming through as non-unicode (utf-8?)
@@ -466,7 +466,7 @@ def addArtisttoDB(artistid, extrasonly=False, forcefull=False):
                     myDB.action('UPDATE albums SET Status=? WHERE AlbumID=?', ['Downloaded', rg['id']])
                     marked_as_downloaded = True
             else:
-                if ((have_track_count / float(total_track_count)) >= (headphones.CONFIG.ALBUM_COMPLETION_PCT / 100.0)):
+                if (have_track_count / float(total_track_count)) >= (headphones.CONFIG.ALBUM_COMPLETION_PCT / 100.0):
                     myDB.action('UPDATE albums SET Status=? WHERE AlbumID=?', ['Downloaded', rg['id']])
                     marked_as_downloaded = True
 
@@ -600,6 +600,17 @@ def addReleaseById(rid, rgid=None):
             newValueDict['IncludeExtras'] = 1
             newValueDict['Extras'] = headphones.CONFIG.EXTRAS
 
+        if 'title' in release_dict:
+            newValueDict['LatestAlbum'] = release_dict['title']
+        elif 'rg_title' in release_dict:
+            newValueDict['LatestAlbum'] = release_dict['rg_title']
+
+        if 'date' in release_dict:
+            newValueDict['ReleaseDate'] = release_dict['date']
+
+        if rgid:
+            newValueDict['AlbumID'] = rgid
+
         myDB.upsert("artists", newValueDict, controlValueDict)
 
     elif not artist_exists and not release_dict:

headphones/lastfm.py

@@ -13,10 +13,9 @@
 #  You should have received a copy of the GNU General Public License
 #  along with Headphones.  If not, see <http://www.gnu.org/licenses/>.
 
-import time
 import random
-import threading
 import headphones
+import headphones.lock
 
 from headphones import db, logger, request
 
@@ -28,7 +27,7 @@ ENTRY_POINT = "http://ws.audioscrobbler.com/2.0/"
 API_KEY = "395e6ec6bb557382fc41fde867bce66f"
 
 # Required for API request limit
-lock = threading.Lock()
+lastfm_lock = headphones.lock.TimedLock(REQUEST_LIMIT)
 
 
 def request_lastfm(method, **kwargs):
@@ -49,8 +48,7 @@ def request_lastfm(method, **kwargs):
     logger.debug("Calling Last.FM method: %s", method)
     logger.debug("Last.FM call parameters: %s", kwargs)
 
-    data = request.request_json(ENTRY_POINT, timeout=TIMEOUT, params=kwargs,
-        rate_limit=(lock, REQUEST_LIMIT))
+    data = request.request_json(ENTRY_POINT, timeout=TIMEOUT, params=kwargs, lock=lastfm_lock)
 
     # Parse response and check for errors.
     if not data:
@@ -73,7 +71,6 @@ def getSimilar():
 
     for result in results[:12]:
         data = request_lastfm("artist.getsimilar", mbid=result["ArtistId"])
-        time.sleep(10)
 
         if data and "similarartists" in data:
             artists = data["similarartists"]["artist"]

headphones/librarysync.py

@@ -20,10 +20,10 @@ from beets.mediafile import MediaFile, FileTypeError, UnreadableFileError
 
 from headphones import db, logger, helpers, importer, lastfm
 
-# You can scan a single directory and append it to the current library by specifying append=True, ArtistID & ArtistName
-
-
-def libraryScan(dir=None, append=False, ArtistID=None, ArtistName=None, cron=False):
+# You can scan a single directory and append it to the current library by
+# specifying append=True, ArtistID and ArtistName.
+def libraryScan(dir=None, append=False, ArtistID=None, ArtistName=None,
+    cron=False):
 
     if cron and not headphones.CONFIG.LIBRARYSCAN:
         return
@@ -64,35 +64,30 @@ def libraryScan(dir=None, append=False, ArtistID=None, ArtistName=None, cron=Fal
             encoded_track_string = track['Location'].encode(headphones.SYS_ENCODING, 'replace')
             if not os.path.isfile(encoded_track_string):
                 if track['ArtistName']:
-                    #Make sure deleted files get accounted for when updating artist track counts
+                    # Make sure deleted files get accounted for when updating artist track counts
                     new_artists.append(track['ArtistName'])
                 myDB.action('DELETE FROM have WHERE Location=?', [track['Location']])
                 logger.info('File %s removed from Headphones, as it is no longer on disk' % encoded_track_string.decode(headphones.SYS_ENCODING, 'replace'))
-           ###############myDB.action('DELETE from have')
 
     bitrates = []
-
     song_list = []
+    latest_subdirectory = []
+
     new_song_count = 0
     file_count = 0
 
-    latest_subdirectory = []
-
-    for r, d, f in os.walk(dir, followlinks=True):
-        # Need to abuse slicing to get a copy of the list, doing it directly
-        # will skip the element after a deleted one using a list comprehension
-        # will not work correctly for nested subdirectories (os.walk keeps its
-        # original list)
-        for directory in d[:]:
-            if directory.startswith("."):
-                d.remove(directory)
+    for r, d, f in helpers.walk_directory(dir):
+        # Filter paths based on config. Note that these methods work directly
+        # on the inputs
+        helpers.path_filter_patterns(d, headphones.CONFIG.IGNORED_FOLDERS, r)
+        helpers.path_filter_patterns(f, headphones.CONFIG.IGNORED_FILES, r)
 
         for files in f:
             # MEDIA_FORMATS = music file extensions, e.g. mp3, flac, etc
             if any(files.lower().endswith('.' + x.lower()) for x in headphones.MEDIA_FORMATS):
-
                 subdirectory = r.replace(dir, '')
                 latest_subdirectory.append(subdirectory)
+
                 if file_count == 0 and r.replace(dir, '') != '':
                     logger.info("[%s] Now scanning subdirectory %s" % (dir.decode(headphones.SYS_ENCODING, 'replace'), subdirectory.decode(headphones.SYS_ENCODING, 'replace')))
                 elif latest_subdirectory[file_count] != latest_subdirectory[file_count - 1] and file_count != 0:
@@ -316,7 +311,6 @@ def libraryScan(dir=None, append=False, ArtistID=None, ArtistName=None, cron=Fal
         ]
 
         # Update track counts
-
         for artist in artists_checked:
             # Have tracks are selected from tracks table and not all tracks because of duplicates
             # We update the track count upon an album switch to compliment this
@@ -324,13 +318,13 @@ def libraryScan(dir=None, append=False, ArtistID=None, ArtistName=None, cron=Fal
                 len(myDB.select('SELECT TrackTitle from tracks WHERE ArtistName like ? AND Location IS NOT NULL', [artist]))
                 + len(myDB.select('SELECT TrackTitle from have WHERE ArtistName like ? AND Matched = "Failed"', [artist]))
             )
-            #Note, some people complain about having "artist have tracks" > # of tracks total in artist official releases
+            # Note: some people complain about having "artist have tracks" > # of tracks total in artist official releases
             # (can fix by getting rid of second len statement)
             myDB.action('UPDATE artists SET HaveTracks=? WHERE ArtistName=?', [havetracks, artist])
 
         logger.info('Found %i new artists' % len(artist_list))
 
-        if len(artist_list):
+        if artist_list:
             if headphones.CONFIG.AUTO_ADD_ARTISTS:
                 logger.info('Importing %i new artists' % len(artist_list))
                 importer.artistlist_to_mbids(artist_list)
@@ -353,6 +347,7 @@ def libraryScan(dir=None, append=False, ArtistID=None, ArtistName=None, cron=Fal
     if not append:
         update_album_status()
         lastfm.getSimilar()
+
     logger.info('Library scan complete')
 
 #ADDED THIS SECTION TO MARK ALBUMS AS DOWNLOADED IF ARTISTS ARE ADDED EN MASSE BEFORE LIBRARY IS SCANNED

headphones/lock.py

@@ -0,0 +1,87 @@
+"""
+Locking-related classes
+"""
+
+import headphones.logger
+import time
+import threading
+import Queue
+
+
+class TimedLock(object):
+    """
+    Enforce request rate limit if applicable. This uses the lock so there
+    is synchronized access to the API. When N threads enter this method, the
+    first will pass trough, since there there was no last request recorded.
+    The last request time will be set. Then, the second thread will unlock,
+    and see that the last request was X seconds ago. It will sleep
+    (request_limit - X) seconds, and then continue. Then the third one will
+    unblock, and so on. After all threads finish, the total time will at
+    least be (N * request_limit) seconds. If some request takes longer than
+    request_limit seconds, the next unblocked thread will wait less.
+    """
+
+    def __init__(self, minimum_delta=0):
+        """
+        Set up the lock
+        """
+        self.lock = threading.Lock()
+        self.last_used = 0
+        self.minimum_delta = minimum_delta
+        self.queue = Queue.Queue()
+
+    def __enter__(self):
+        """
+        Called when with lock: is invoked
+        """
+        self.lock.acquire()
+        delta = time.time() - self.last_used
+        sleep_amount = self.minimum_delta - delta
+        if sleep_amount >= 0:
+            # zero sleeps give the cpu a chance to task-switch
+            headphones.logger.debug('Sleeping %s (interval)', sleep_amount)
+            time.sleep(sleep_amount)
+        while not self.queue.empty():
+            try:
+                seconds = self.queue.get(False)
+                headphones.logger.debug('Sleeping %s (queued)', seconds)
+                time.sleep(seconds)
+            except Queue.Empty:
+                continue
+            self.queue.task_done()
+
+    def __exit__(self, type, value, traceback):
+        """
+        Called when exiting the with block.
+        """
+        self.last_used = time.time()
+        self.lock.release()
+
+    def snooze(self, seconds):
+        """
+        Asynchronously add time to the next request. Can be called outside
+        of the lock context, but it is possible for the next lock holder
+        to not check the queue until after something adds time to it.
+        """
+        # We use a queue so that we don't have to synchronize
+        # across threads and with or without locks
+        headphones.logger.info('Adding %s to queue', seconds)
+        self.queue.put(seconds)
+
+
+class FakeLock(object):
+    """
+    If no locking or request throttling is needed, use this
+    """
+
+    def __enter__(self):
+        """
+        Do nothing on enter
+        """
+        pass
+
+    def __exit__(self, type, value, traceback):
+        """
+        Do nothing on exit
+        """
+        pass

headphones/logger.py

@@ -39,7 +39,6 @@ logger = logging.getLogger("headphones")
 # Global queue for multiprocessing logging
 queue = None
 
-
 class LogListHandler(logging.Handler):
     """
     Log handler for Web UI.
@@ -112,7 +111,7 @@ def initMultiprocessing():
     threading.current_thread().name = multiprocessing.current_process().name
 
 
-def initLogger(console=False, verbose=False):
+def initLogger(console=False, log_dir=False, verbose=False):
     """
     Setup logging for Headphones. It uses the logger instance with the name
     'headphones'. Three log handlers are added:
@@ -121,7 +120,8 @@ def initLogger(console=False, verbose=False):
     * LogListHandler: for Web UI
     * StreamHandler: for console (if console)
 
-    Console logging is only enabled if console is set to True.
+    Console logging is only enabled if console is set to True. This method can
+    be invoked multiple times, during different stages of Headphones.
     """
 
     # Close and remove old handlers. This is required to reinit the loggers
@@ -139,22 +139,23 @@ def initLogger(console=False, verbose=False):
     logger.propagate = False
     logger.setLevel(logging.DEBUG if verbose else logging.INFO)
 
-    # Setup file logger
-    filename = os.path.join(headphones.CONFIG.LOG_DIR, FILENAME)
-
-    file_formatter = logging.Formatter('%(asctime)s - %(levelname)-7s :: %(threadName)s : %(message)s', '%d-%b-%Y %H:%M:%S')
-    file_handler = handlers.RotatingFileHandler(filename, maxBytes=MAX_SIZE, backupCount=MAX_FILES)
-    file_handler.setLevel(logging.DEBUG)
-    file_handler.setFormatter(file_formatter)
-
-    logger.addHandler(file_handler)
-
     # Add list logger
     loglist_handler = LogListHandler()
     loglist_handler.setLevel(logging.DEBUG)
 
     logger.addHandler(loglist_handler)
 
+    # Setup file logger
+    if log_dir:
+        filename = os.path.join(log_dir, FILENAME)
+
+        file_formatter = logging.Formatter('%(asctime)s - %(levelname)-7s :: %(threadName)s : %(message)s', '%d-%b-%Y %H:%M:%S')
+        file_handler = handlers.RotatingFileHandler(filename, maxBytes=MAX_SIZE, backupCount=MAX_FILES)
+        file_handler.setLevel(logging.DEBUG)
+        file_handler.setFormatter(file_formatter)
+
+        logger.addHandler(file_handler)
+
     # Setup console logger
     if console:
         console_formatter = logging.Formatter('%(asctime)s - %(levelname)s :: %(threadName)s : %(message)s', '%d-%b-%Y %H:%M:%S')

headphones/mb.py

@@ -16,10 +16,9 @@
 
 from headphones import logger, db, helpers
 
-import time
-import threading
 import headphones
 import musicbrainzngs
+import headphones.lock
 
 try:
     # pylint:disable=E0611
@@ -30,7 +29,7 @@ except ImportError:
     # Python 2.6.x fallback, from libs
     from ordereddict import OrderedDict
 
-mb_lock = threading.Lock()
+mb_lock = headphones.lock.TimedLock(0)
 
 # Quick fix to add mirror switching on the fly. Need to probably return the mbhost & mbport that's
 # being used, so we can send those values to the log
@@ -60,11 +59,14 @@ def startmb():
 
     musicbrainzngs.set_useragent("headphones", "0.0", "https://github.com/rembo10/headphones")
     musicbrainzngs.set_hostname(mbhost + ":" + str(mbport))
+
+    # Their rate limiting should be redundant to our lock
     if sleepytime == 0:
         musicbrainzngs.set_rate_limit(False)
     else:
         #calling it with an it ends up blocking all requests after the first
         musicbrainzngs.set_rate_limit(limit_or_interval=float(sleepytime))
+        mb_lock.minimum_delta = sleepytime
 
     # Add headphones credentials
     if headphones.CONFIG.MIRROR == "headphones":
@@ -79,335 +81,315 @@ def startmb():
 
 
 def findArtist(name, limit=1):
+     artistlist = []
+     artistResults = None
 
-    with mb_lock:
-        artistlist = []
-        artistResults = None
+     chars = set('!?*-')
+     if any((c in chars) for c in name):
+         name = '"' + name + '"'
 
-        chars = set('!?*-')
-        if any((c in chars) for c in name):
-            name = '"' + name + '"'
+     criteria = {'artist': name.lower()}
 
-        criteria = {'artist': name.lower()}
+     with mb_lock:
+         try:
+             artistResults = musicbrainzngs.search_artists(limit=limit, **criteria)['artist-list']
+         except musicbrainzngs.WebServiceError as e:
+             logger.warn('Attempt to query MusicBrainz for %s failed (%s)' % (name, str(e)))
+             mb_lock.snooze(5)
 
-        try:
-            artistResults = musicbrainzngs.search_artists(limit=limit, **criteria)['artist-list']
-        except musicbrainzngs.WebServiceError as e:
-            logger.warn('Attempt to query MusicBrainz for %s failed (%s)' % (name, str(e)))
-            time.sleep(5)
-
-        if not artistResults:
-            return False
-        for result in artistResults:
-            if 'disambiguation' in result:
-                uniquename = unicode(result['sort-name'] + " (" + result['disambiguation'] + ")")
-            else:
-                uniquename = unicode(result['sort-name'])
-            if result['name'] != uniquename and limit == 1:
-                logger.info('Found an artist with a disambiguation: %s - doing an album based search' % name)
-                artistdict = findArtistbyAlbum(name)
-                if not artistdict:
-                    logger.info('Cannot determine the best match from an artist/album search. Using top match instead')
-                    artistlist.append({
-                    # Just need the artist id if the limit is 1
-                    #    'name':             unicode(result['sort-name']),
-                    #    'uniquename':        uniquename,
-                        'id': unicode(result['id']),
-                    #    'url':                 unicode("http://musicbrainz.org/artist/" + result['id']),#probably needs to be changed
-                    #    'score':            int(result['ext:score'])
-                        })
-                else:
-                    artistlist.append(artistdict)
-            else:
-                artistlist.append({
-                        'name': unicode(result['sort-name']),
-                        'uniquename': uniquename,
-                        'id': unicode(result['id']),
-                        'url': unicode("http://musicbrainz.org/artist/" + result['id']),#probably needs to be changed
-                        'score': int(result['ext:score'])
-                        })
-        return artistlist
+     if not artistResults:
+         return False
+     for result in artistResults:
+         if 'disambiguation' in result:
+             uniquename = unicode(result['sort-name'] + " (" + result['disambiguation'] + ")")
+         else:
+             uniquename = unicode(result['sort-name'])
+         if result['name'] != uniquename and limit == 1:
+             logger.info('Found an artist with a disambiguation: %s - doing an album based search' % name)
+             artistdict = findArtistbyAlbum(name)
+             if not artistdict:
+                 logger.info('Cannot determine the best match from an artist/album search. Using top match instead')
+                 artistlist.append({
+                 # Just need the artist id if the limit is 1
+                 #    'name':             unicode(result['sort-name']),
+                 #    'uniquename':        uniquename,
+                     'id': unicode(result['id']),
+                 #    'url':                 unicode("http://musicbrainz.org/artist/" + result['id']),#probably needs to be changed
+                 #    'score':            int(result['ext:score'])
+                     })
+             else:
+                 artistlist.append(artistdict)
+         else:
+             artistlist.append({
+                     'name': unicode(result['sort-name']),
+                     'uniquename': uniquename,
+                     'id': unicode(result['id']),
+                     'url': unicode("http://musicbrainz.org/artist/" + result['id']),#probably needs to be changed
+                     'score': int(result['ext:score'])
+                     })
+     return artistlist
 
 
 def findRelease(name, limit=1, artist=None):
+    releaselist = []
+    releaseResults = None
+
+    # additional artist search
+    if not artist and ':' in name:
+        name, artist = name.rsplit(":", 1)
+
+    chars = set('!?*-')
+    if any((c in chars) for c in name):
+        name = '"' + name + '"'
+    if artist and any((c in chars) for c in artist):
+        artist = '"' + artist + '"'
 
     with mb_lock:
-        releaselist = []
-        releaseResults = None
-
-        # additional artist search
-        if not artist and ':' in name:
-            name, artist = name.rsplit(":", 1)
-
-        chars = set('!?*-')
-        if any((c in chars) for c in name):
-            name = '"' + name + '"'
-        if artist and any((c in chars) for c in artist):
-            artist = '"' + artist + '"'
-
         try:
             releaseResults = musicbrainzngs.search_releases(query=name, limit=limit, artist=artist)['release-list']
         except musicbrainzngs.WebServiceError as e: #need to update exceptions
             logger.warn('Attempt to query MusicBrainz for "%s" failed: %s' % (name, str(e)))
-            time.sleep(5)
+            mb_lock.snooze(5)
 
-        if not releaseResults:
-            return False
+    if not releaseResults:
+        return False
 
-        for result in releaseResults:
+    for result in releaseResults:
 
-            title = result['title']
-            if 'disambiguation' in result:
-                title += ' (' + result['disambiguation'] + ')'
+        title = result['title']
+        if 'disambiguation' in result:
+            title += ' (' + result['disambiguation'] + ')'
 
-            # Get formats and track counts
-            format_dict = OrderedDict()
-            formats = ''
-            tracks = ''
-            if 'medium-list' in result:
-                for medium in result['medium-list']:
-                    if 'format' in medium:
-                        format = medium['format']
-                        if format not in format_dict:
-                            format_dict[format] = 0
-                        format_dict[format] += 1
-                    if 'track-count' in medium:
-                        if tracks:
-                            tracks += ' + '
-                        tracks += str(medium['track-count'])
-                for format, count in format_dict.items():
-                    if formats:
-                        formats += ' + '
-                    if count > 1:
-                        formats += str(count) + 'x'
-                    formats += format
+        # Get formats and track counts
+        format_dict = OrderedDict()
+        formats = ''
+        tracks = ''
+        if 'medium-list' in result:
+            for medium in result['medium-list']:
+                if 'format' in medium:
+                    format = medium['format']
+                    if format not in format_dict:
+                        format_dict[format] = 0
+                    format_dict[format] += 1
+                if 'track-count' in medium:
+                    if tracks:
+                        tracks += ' + '
+                    tracks += str(medium['track-count'])
+            for format, count in format_dict.items():
+                if formats:
+                    formats += ' + '
+                if count > 1:
+                    formats += str(count) + 'x'
+                formats += format
 
-            rg_type = ''
-            if 'type' in result['release-group']:
-                rg_type = result['release-group']['type']
-                if rg_type == 'Album' and 'secondary-type-list' in result['release-group']:
-                    secondary_type = result['release-group']['secondary-type-list'][0]
-                    if secondary_type != rg_type:
-                        rg_type = secondary_type
+        rg_type = ''
+        if 'type' in result['release-group']:
+            rg_type = result['release-group']['type']
+            if rg_type == 'Album' and 'secondary-type-list' in result['release-group']:
+                secondary_type = result['release-group']['secondary-type-list'][0]
+                if secondary_type != rg_type:
+                    rg_type = secondary_type
 
-            releaselist.append({
-                        'uniquename': unicode(result['artist-credit'][0]['artist']['name']),
-                        'title': unicode(title),
-                        'id': unicode(result['artist-credit'][0]['artist']['id']),
-                        'albumid': unicode(result['id']),
-                        'url': unicode("http://musicbrainz.org/artist/" + result['artist-credit'][0]['artist']['id']),#probably needs to be changed
-                        'albumurl': unicode("http://musicbrainz.org/release/" + result['id']),#probably needs to be changed
-                        'score': int(result['ext:score']),
-                        'date': unicode(result['date']) if 'date' in result else '',
-                        'country': unicode(result['country']) if 'country' in result else '',
-                        'formats': unicode(formats),
-                        'tracks': unicode(tracks),
-                        'rgid': unicode(result['release-group']['id']),
-                        'rgtype': unicode(rg_type)
-                        })
-        return releaselist
+        releaselist.append({
+                    'uniquename': unicode(result['artist-credit'][0]['artist']['name']),
+                    'title': unicode(title),
+                    'id': unicode(result['artist-credit'][0]['artist']['id']),
+                    'albumid': unicode(result['id']),
+                    'url': unicode("http://musicbrainz.org/artist/" + result['artist-credit'][0]['artist']['id']),#probably needs to be changed
+                    'albumurl': unicode("http://musicbrainz.org/release/" + result['id']),#probably needs to be changed
+                    'score': int(result['ext:score']),
+                    'date': unicode(result['date']) if 'date' in result else '',
+                    'country': unicode(result['country']) if 'country' in result else '',
+                    'formats': unicode(formats),
+                    'tracks': unicode(tracks),
+                    'rgid': unicode(result['release-group']['id']),
+                    'rgtype': unicode(rg_type)
+                    })
+    return releaselist
 
 
 def getArtist(artistid, extrasonly=False):
-
-    with mb_lock:
-        artist_dict = {}
-
-        artist = None
-
-        try:
-            limit = 200
+    artist_dict = {}
+    artist = None
+    try:
+        limit = 200
+        with mb_lock:
             artist = musicbrainzngs.get_artist_by_id(artistid)['artist']
-            newRgs = None
-            artist['release-group-list'] = []
-            while newRgs is None or len(newRgs) >= limit:
-                newRgs = musicbrainzngs.browse_release_groups(artistid, release_type="album", offset=len(artist['release-group-list']), limit=limit)['release-group-list']
-                artist['release-group-list'] += newRgs
-        except musicbrainzngs.WebServiceError as e:
-            logger.warn('Attempt to retrieve artist information from MusicBrainz failed for artistid: %s (%s)' % (artistid, str(e)))
-            time.sleep(5)
-        except Exception as e:
-            pass
+        newRgs = None
+        artist['release-group-list'] = []
+        while newRgs is None or len(newRgs) >= limit:
+            with mb_lock:
+                newRgs = musicbrainzngs.browse_release_groups(
+                    artistid,
+                    release_type="album",
+                    offset=len(artist['release-group-list']),
+                    limit=limit)
+            newRgs = newRgs['release-group-list']
+            artist['release-group-list'] += newRgs
+    except musicbrainzngs.WebServiceError as e:
+        logger.warn('Attempt to retrieve artist information from MusicBrainz failed for artistid: %s (%s)' % (artistid, str(e)))
+        mb_lock.snooze(5)
+    except Exception as e:
+        pass
 
-        if not artist:
-            return False
+    if not artist:
+        return False
 
-        #if 'disambiguation' in artist:
-        #    uniquename = unicode(artist['sort-name'] + " (" + artist['disambiguation'] + ")")
-        #else:
-        #    uniquename = unicode(artist['sort-name'])
+    artist_dict['artist_name'] = unicode(artist['name'])
 
-        artist_dict['artist_name'] = unicode(artist['name'])
+    releasegroups = []
 
-        # Not using the following values anywhere yet so we don't need to grab them.
-        # Was causing an exception to be raised if they didn't exist.
-        #
-        #artist_dict['artist_sortname'] = unicode(artist['sort-name'])
-        #artist_dict['artist_uniquename'] = uniquename
-        #artist_dict['artist_type'] = unicode(artist['type'])
+    if not extrasonly:
+        for rg in artist['release-group-list']:
+            if "secondary-type-list" in rg.keys(): #only add releases without a secondary type
+                continue
+            releasegroups.append({
+                'title': unicode(rg['title']),
+                'id': unicode(rg['id']),
+                'url': u"http://musicbrainz.org/release-group/" + rg['id'],
+                'type': unicode(rg['type'])
+            })
 
-        #artist_dict['artist_begindate'] = None
-        #artist_dict['artist_enddate'] = None
-        #if 'life-span' in artist:
-        #    if 'begin' in artist['life-span']:
-        #        artist_dict['artist_begindate'] = unicode(artist['life-span']['begin'])
-        #    if 'end' in artist['life-span']:
-        #        artist_dict['artist_enddate'] = unicode(artist['life-span']['end'])
+    # See if we need to grab extras. Artist specific extras take precedence over global option
+    # Global options are set when adding a new artist
+    myDB = db.DBConnection()
 
-        releasegroups = []
+    try:
+        db_artist = myDB.action('SELECT IncludeExtras, Extras from artists WHERE ArtistID=?', [artistid]).fetchone()
+        includeExtras = db_artist['IncludeExtras']
+    except IndexError:
+        includeExtras = False
+
+    if includeExtras:
+
+        # Need to convert extras string from something like '2,5.6' to ['ep','live','remix'] (append new extras to end)
+        if db_artist['Extras']:
+            extras = map(int, db_artist['Extras'].split(','))
+        else:
+            extras = []
+        extras_list = headphones.POSSIBLE_EXTRAS
+
+        includes = []
+
+        i = 1
+        for extra in extras_list:
+            if i in extras:
+                includes.append(extra)
+            i += 1
+
+        for include in includes:
+
+            mb_extras_list = []
+
+            try:
+                limit = 200
+                newRgs = None
+                while newRgs is None or len(newRgs) >= limit:
+                    with mb_lock:
+                        newRgs = musicbrainzngs.browse_release_groups(
+                            artistid, release_type=include, offset=len(mb_extras_list), limit=limit)
+                    newRgs = newRgs['release-group-list']
+                    mb_extras_list += newRgs
+            except musicbrainzngs.WebServiceError as e:
+                logger.warn('Attempt to retrieve artist information from MusicBrainz failed for artistid: %s (%s)' % (artistid, str(e)))
+                mb_lock.snooze(5)
+
+            for rg in mb_extras_list:
+                rg_type = rg['type']
+                if rg_type == 'Album' and 'secondary-type-list' in rg:
+                    secondary_type = rg['secondary-type-list'][0]
+                    if secondary_type != rg_type:
+                        rg_type = secondary_type
 
-        if not extrasonly:
-            for rg in artist['release-group-list']:
-                if "secondary-type-list" in rg.keys(): #only add releases without a secondary type
-                    continue
                 releasegroups.append({
-                            'title': unicode(rg['title']),
-                            'id': unicode(rg['id']),
-                            'url': u"http://musicbrainz.org/release-group/" + rg['id'],
-                            'type': unicode(rg['type'])
-                    })
-
-        # See if we need to grab extras. Artist specific extras take precedence over global option
-        # Global options are set when adding a new artist
-        myDB = db.DBConnection()
-
-        try:
-            db_artist = myDB.action('SELECT IncludeExtras, Extras from artists WHERE ArtistID=?', [artistid]).fetchone()
-            includeExtras = db_artist['IncludeExtras']
-        except IndexError:
-            includeExtras = False
-
-        if includeExtras:
-
-            # Need to convert extras string from something like '2,5.6' to ['ep','live','remix'] (append new extras to end)
-            if db_artist['Extras']:
-                extras = map(int, db_artist['Extras'].split(','))
-            else:
-                extras = []
-            extras_list = headphones.POSSIBLE_EXTRAS
-
-            includes = []
-
-            i = 1
-            for extra in extras_list:
-                if i in extras:
-                    includes.append(extra)
-                i += 1
-
-            for include in includes:
-
-                mb_extras_list = []
-
-                try:
-                    limit = 200
-                    newRgs = None
-                    while newRgs is None or len(newRgs) >= limit:
-                        newRgs = musicbrainzngs.browse_release_groups(artistid, release_type=include, offset=len(mb_extras_list), limit=limit)['release-group-list']
-                        mb_extras_list += newRgs
-                except musicbrainzngs.WebServiceError as e:
-                    logger.warn('Attempt to retrieve artist information from MusicBrainz failed for artistid: %s (%s)' % (artistid, str(e)))
-                    time.sleep(5)
-
-                for rg in mb_extras_list:
-
-                    rg_type = rg['type']
-                    if rg_type == 'Album' and 'secondary-type-list' in rg:
-                        secondary_type = rg['secondary-type-list'][0]
-                        if secondary_type != rg_type:
-                            rg_type = secondary_type
-
-                    releasegroups.append({
-                            'title': unicode(rg['title']),
-                            'id': unicode(rg['id']),
-                            'url': u"http://musicbrainz.org/release-group/" + rg['id'],
-                            'type': unicode(rg_type)
-                        })
-
-        artist_dict['releasegroups'] = releasegroups
-
-        return artist_dict
+                    'title': unicode(rg['title']),
+                    'id': unicode(rg['id']),
+                    'url': u"http://musicbrainz.org/release-group/" + rg['id'],
+                    'type': unicode(rg_type)
+                })
+    artist_dict['releasegroups'] = releasegroups
+    return artist_dict
 
 
 def getReleaseGroup(rgid):
     """
     Returns a list of releases in a release group
     """
-    with mb_lock:
+    releaseGroup = None
+    try:
+        with mb_lock:
+            releaseGroup = musicbrainzngs.get_release_group_by_id(
+                rgid, ["artists", "releases", "media", "discids", ])
+            releaseGroup = releaseGroup['release-group']
+    except musicbrainzngs.WebServiceError as e:
+        logger.warn('Attempt to retrieve information from MusicBrainz for release group "%s" failed (%s)' % (rgid, str(e)))
+        mb_lock.snooze(5)
 
-        releaseGroup = None
-
-        try:
-            releaseGroup = musicbrainzngs.get_release_group_by_id(rgid, ["artists", "releases", "media", "discids", ])['release-group']
-        except musicbrainzngs.WebServiceError as e:
-            logger.warn('Attempt to retrieve information from MusicBrainz for release group "%s" failed (%s)' % (rgid, str(e)))
-            time.sleep(5)
-
-        if not releaseGroup:
-            return False
-        else:
-            return releaseGroup['release-list']
+    if not releaseGroup:
+        return False
+    else:
+        return releaseGroup['release-list']
 
 
 def getRelease(releaseid, include_artist_info=True):
     """
     Deep release search to get track info
     """
-    with mb_lock:
+    release = {}
+    results = None
 
-        release = {}
-        results = None
-
-        try:
+    try:
+        with mb_lock:
             if include_artist_info:
                 results = musicbrainzngs.get_release_by_id(releaseid, ["artists", "release-groups", "media", "recordings"]).get('release')
             else:
                 results = musicbrainzngs.get_release_by_id(releaseid, ["media", "recordings"]).get('release')
-        except musicbrainzngs.WebServiceError as e:
-            logger.warn('Attempt to retrieve information from MusicBrainz for release "%s" failed (%s)' % (releaseid, str(e)))
-            time.sleep(5)
+    except musicbrainzngs.WebServiceError as e:
+        logger.warn('Attempt to retrieve information from MusicBrainz for release "%s" failed (%s)' % (releaseid, str(e)))
+        mb_lock.snooze(5)
 
-        if not results:
-            return False
+    if not results:
+        return False
 
-        release['title'] = unicode(results['title'])
-        release['id'] = unicode(results['id'])
-        release['asin'] = unicode(results['asin']) if 'asin' in results else None
-        release['date'] = unicode(results['date']) if 'date' in results else None
-        try:
-            release['format'] = unicode(results['medium-list'][0]['format'])
-        except:
-            release['format'] = u'Unknown'
+    release['title'] = unicode(results['title'])
+    release['id'] = unicode(results['id'])
+    release['asin'] = unicode(results['asin']) if 'asin' in results else None
+    release['date'] = unicode(results['date']) if 'date' in results else None
+    try:
+        release['format'] = unicode(results['medium-list'][0]['format'])
+    except:
+        release['format'] = u'Unknown'
 
-        try:
-            release['country'] = unicode(results['country'])
-        except:
-            release['country'] = u'Unknown'
+    try:
+        release['country'] = unicode(results['country'])
+    except:
+        release['country'] = u'Unknown'
 
-        if include_artist_info:
+    if include_artist_info:
 
-            if 'release-group' in results:
-                release['rgid'] = unicode(results['release-group']['id'])
-                release['rg_title'] = unicode(results['release-group']['title'])
-                try:
-                    release['rg_type'] = unicode(results['release-group']['type'])
+        if 'release-group' in results:
+            release['rgid'] = unicode(results['release-group']['id'])
+            release['rg_title'] = unicode(results['release-group']['title'])
+            try:
+                release['rg_type'] = unicode(results['release-group']['type'])
 
-                    if release['rg_type'] == 'Album' and 'secondary-type-list' in results['release-group']:
-                        secondary_type = unicode(results['release-group']['secondary-type-list'][0])
-                        if secondary_type != release['rg_type']:
-                            release['rg_type'] = secondary_type
+                if release['rg_type'] == 'Album' and 'secondary-type-list' in results['release-group']:
+                    secondary_type = unicode(results['release-group']['secondary-type-list'][0])
+                    if secondary_type != release['rg_type']:
+                        release['rg_type'] = secondary_type
 
-                except KeyError:
-                    release['rg_type'] = u'Unknown'
+            except KeyError:
+                release['rg_type'] = u'Unknown'
 
-            else:
-                logger.warn("Release " + releaseid + "had no ReleaseGroup associated")
+        else:
+            logger.warn("Release " + releaseid + "had no ReleaseGroup associated")
 
-            release['artist_name'] = unicode(results['artist-credit'][0]['artist']['name'])
-            release['artist_id'] = unicode(results['artist-credit'][0]['artist']['id'])
+        release['artist_name'] = unicode(results['artist-credit'][0]['artist']['name'])
+        release['artist_id'] = unicode(results['artist-credit'][0]['artist']['id'])
 
-        release['tracks'] = getTracksFromRelease(results)
+    release['tracks'] = getTracksFromRelease(results)
 
-        return release
+    return release
 
 
 def get_new_releases(rgid, includeExtras=False, forcefull=False):
@@ -418,7 +400,12 @@ def get_new_releases(rgid, includeExtras=False, forcefull=False):
         limit = 100
         newResults = None
         while newResults is None or len(newResults) >= limit:
-            newResults = musicbrainzngs.browse_releases(release_group=rgid, includes=['artist-credits', 'labels', 'recordings', 'release-groups', 'media'], limit=limit, offset=len(results))
+            with mb_lock:
+                newResults = musicbrainzngs.browse_releases(
+                    release_group=rgid,
+                    includes=['artist-credits', 'labels', 'recordings', 'release-groups', 'media'],
+                    limit=limit,
+                    offset=len(results))
             if 'release-list' not in newResults:
                 break #may want to raise an exception here instead ?
             newResults = newResults['release-list']
@@ -426,7 +413,7 @@ def get_new_releases(rgid, includeExtras=False, forcefull=False):
 
     except musicbrainzngs.WebServiceError as e:
         logger.warn('Attempt to retrieve information from MusicBrainz for release group "%s" failed (%s)' % (rgid, str(e)))
-        time.sleep(5)
+        mb_lock.snooze(5)
         return False
 
     if not results or len(results) == 0:
@@ -565,11 +552,11 @@ def get_new_releases(rgid, includeExtras=False, forcefull=False):
                 logger.info('[%s] Existing release %s (%s) updated' % (release['ArtistName'], release['AlbumTitle'], rel_id_check))
             else:
                 logger.info('[%s] New release %s (%s) added' % (release['ArtistName'], release['AlbumTitle'], rel_id_check))
-    if force_repackage1 == 1:
-        num_new_releases = -1
-        logger.info('[%s] Forcing repackage of %s, since dB releases have been removed' % (release['ArtistName'], release_title))
-    else:
-        num_new_releases = num_new_releases
+        if force_repackage1 == 1:
+            num_new_releases = -1
+            logger.info('[%s] Forcing repackage of %s, since dB releases have been removed' % (release['ArtistName'], release_title))
+        else:
+            num_new_releases = num_new_releases
 
     return num_new_releases
 
@@ -614,10 +601,11 @@ def findArtistbyAlbum(name):
     results = None
 
     try:
-        results = musicbrainzngs.search_release_groups(term).get('release-group-list')
+        with mb_lock:
+            results = musicbrainzngs.search_release_groups(term).get('release-group-list')
     except musicbrainzngs.WebServiceError as e:
         logger.warn('Attempt to query MusicBrainz for %s failed (%s)' % (name, str(e)))
-        time.sleep(5)
+        mb_lock.snooze(5)
 
     if not results:
         return False
@@ -656,11 +644,11 @@ def findAlbumID(artist=None, album=None):
             if any((c in chars) for c in album):
                 album = '"' + album + '"'
             criteria = {'release': album.lower()}
-
-        results = musicbrainzngs.search_release_groups(limit=1, **criteria).get('release-group-list')
+        with mb_lock:
+            results = musicbrainzngs.search_release_groups(limit=1, **criteria).get('release-group-list')
     except musicbrainzngs.WebServiceError as e:
         logger.warn('Attempt to query MusicBrainz for %s - %s failed (%s)' % (artist, album, str(e)))
-        time.sleep(5)
+        mb_lock.snooze(5)
 
     if not results:
         return False

headphones/music_encoder.py

@@ -14,12 +14,12 @@
 #  along with Headphones.  If not, see <http://www.gnu.org/licenses/>.
 
 import os
-import headphones
-import shutil
 import time
+import shutil
+import subprocess
+import headphones
 import multiprocessing
 
-import subprocess
 from headphones import logger
 from beets.mediafile import MediaFile
 
@@ -36,6 +36,8 @@ def encode(albumPath):
         if not xldFormat:
             logger.error('Details for xld profile \'%s\' not found, files will not be re-encoded', xldProfile)
             return None
+    else:
+        xldProfile = None
 
     tempDirEncode = os.path.join(albumPath, "temp")
     musicFiles = []
@@ -64,7 +66,7 @@ def encode(albumPath):
                     xldInfoMusic = MediaFile(xldMusicFile)
                     encoderFormat = xldFormat
 
-                if (headphones.CONFIG.ENCODERLOSSLESS):
+                if headphones.CONFIG.ENCODERLOSSLESS:
                     ext = os.path.normpath(os.path.splitext(music)[1].lstrip(".")).lower()
                     if not use_xld and ext == 'flac' or use_xld and (ext != xldFormat and (xldInfoMusic.bitrate / 1000 > 400)):
                         musicFiles.append(os.path.join(r, music))
@@ -116,7 +118,7 @@ def encode(albumPath):
             if not any(music.decode(headphones.SYS_ENCODING, 'replace').lower().endswith('.' + x) for x in ["mp3", "wav"]):
                 logger.warn('Lame cannot encode %s format for %s, use ffmpeg', os.path.splitext(music)[1], music)
             else:
-                if (music.decode(headphones.SYS_ENCODING, 'replace').lower().endswith('.mp3') and (int(infoMusic.bitrate / 1000) <= headphones.CONFIG.BITRATE)):
+                if music.decode(headphones.SYS_ENCODING, 'replace').lower().endswith('.mp3') and (int(infoMusic.bitrate / 1000) <= headphones.CONFIG.BITRATE):
                     logger.info('%s has bitrate <= %skb, will not be re-encoded', music, headphones.CONFIG.BITRATE)
                 else:
                     encode = True
@@ -126,14 +128,14 @@ def encode(albumPath):
                     logger.warn('Cannot re-encode .ogg %s', music.decode(headphones.SYS_ENCODING, 'replace'))
                 else:
                     encode = True
-            elif (headphones.CONFIG.ENCODEROUTPUTFORMAT == 'mp3' or headphones.CONFIG.ENCODEROUTPUTFORMAT == 'm4a'):
-                if (music.decode(headphones.SYS_ENCODING, 'replace').lower().endswith('.' + headphones.CONFIG.ENCODEROUTPUTFORMAT) and (int(infoMusic.bitrate / 1000) <= headphones.CONFIG.BITRATE)):
+            elif headphones.CONFIG.ENCODEROUTPUTFORMAT == 'mp3' or headphones.CONFIG.ENCODEROUTPUTFORMAT == 'm4a':
+                if music.decode(headphones.SYS_ENCODING, 'replace').lower().endswith('.' + headphones.CONFIG.ENCODEROUTPUTFORMAT) and (int(infoMusic.bitrate / 1000) <= headphones.CONFIG.BITRATE):
                     logger.info('%s has bitrate <= %skb, will not be re-encoded', music, headphones.CONFIG.BITRATE)
                 else:
                     encode = True
         # encode
         if encode:
-            job = (encoder, music, musicTempFiles[i], albumPath)
+            job = (encoder, music, musicTempFiles[i], albumPath, xldProfile)
             jobs.append(job)
         else:
             musicFiles[i] = None
@@ -242,22 +244,16 @@ def command_map(args):
         return False
 
 
-def command(encoder, musicSource, musicDest, albumPath):
+def command(encoder, musicSource, musicDest, albumPath, xldProfile):
     """
     Encode a given music file with a certain encoder. Returns True on success,
     or False otherwise.
     """
-    use_xld = headphones.CONFIG.ENCODER == 'xld'
 
     startMusicTime = time.time()
     cmd = []
 
-    # Return if xld details not found
-    if use_xld:
-        (xldProfile, xldFormat, xldBitrate) = getXldProfile.getXldProfile(headphones.CONFIG.XLDPROFILE)
-        if not xldFormat:
-            logger.error('Details for xld profile \'%s\' not found, files will not be re-encoded', xldProfile)
-            return None
+    if xldProfile:
         xldDestDir = os.path.split(musicDest)[0]
         cmd = [encoder]
         cmd.extend([musicSource])

headphones/notifiers.py

@@ -17,6 +17,7 @@ from headphones import logger, helpers, common, request
 
 from xml.dom import minidom
 from httplib import HTTPSConnection
+from urlparse import parse_qsl
 from urllib import urlencode
 from pynma import pynma
 
@@ -33,12 +34,6 @@ import json
 import oauth2 as oauth
 import pythontwitter as twitter
 
-try:
-    from urlparse import parse_qsl
-except ImportError:
-    from cgi import parse_qsl
-
-
 class GROWL(object):
     """
     Growl notifications, for OS X.
@@ -295,7 +290,7 @@ class LMS(object):
             return response['result']
         except:
             logger.warn('LMS returned error: %s' % response['error'])
-            return
+            return response['error']
 
     def update(self):
 
@@ -305,7 +300,7 @@ class LMS(object):
             logger.info('Sending library rescan command to LMS @ ' + host)
             request = self._sendjson(host)
 
-            if not request:
+            if request:
                 logger.warn('Error sending rescan request to LMS')
 
 
@@ -449,16 +444,15 @@ class PUSHBULLET(object):
 
         http_handler = HTTPSConnection("api.pushbullet.com")
 
-        data = {'device_iden': headphones.CONFIG.PUSHBULLET_DEVICEID,
-                'type': "note",
+        data = {'type': "note",
                 'title': "Headphones",
                 'body': message.encode("utf-8")}
 
         http_handler.request("POST",
-                                "/api/pushes",
-                                headers={'Content-type': "application/x-www-form-urlencoded",
+                                "/v2/pushes",
+                                headers={'Content-type': "application/json",
                                             'Authorization': 'Basic %s' % base64.b64encode(headphones.CONFIG.PUSHBULLET_APIKEY + ":")},
-                                body=urlencode(data))
+                                body=json.dumps(data))
         response = http_handler.getresponse()
         request_status = response.status
         logger.debug(u"PushBullet response status: %r" % request_status)
@@ -731,6 +725,7 @@ class OSX_NOTIFY(object):
     def __init__(self):
         try:
             self.objc = __import__("objc")
+            self.AppKit = __import__("AppKit")
         except:
             return False
 
@@ -743,7 +738,7 @@ class OSX_NOTIFY(object):
             signature=old_IMP.signature)
         self.objc.classAddMethod(cls, SEL, new_IMP)
 
-    def notify(self, title, subtitle=None, text=None, sound=True):
+    def notify(self, title, subtitle=None, text=None, sound=True, image=None):
 
         try:
             self.swizzle(self.objc.lookUpClass('NSBundle'),
@@ -767,6 +762,10 @@ class OSX_NOTIFY(object):
                 notification.setInformativeText_(text)
             if sound:
                 notification.setSoundName_("NSUserNotificationDefaultSoundName")
+            if image:
+                source_img = self.AppKit.NSImage.alloc().initByReferencingFile_(image)
+                notification.setContentImage_(source_img)
+                #notification.set_identityImage_(source_img)
             notification.setHasActionButton_(False)
 
             notification_center = NSUserNotificationCenter.defaultUserNotificationCenter()

headphones/nzbget.py

@@ -59,7 +59,7 @@ def sendNZB(nzb):
         return False
 
     except xmlrpclib.ProtocolError, e:
-        if (e.errmsg == "Unauthorized"):
+        if e.errmsg == "Unauthorized":
             logger.error(u"NZBget password is incorrect.")
         else:
             logger.error(u"Protocol Error: " + e.errmsg)

headphones/postprocessor.py

@@ -19,6 +19,7 @@ import shutil
 import uuid
 import beets
 import threading
+import itertools
 import headphones
 
 from beets import autotag
@@ -33,16 +34,14 @@ postprocessor_lock = threading.Lock()
 
 
 def checkFolder():
+    logger.info("Checking download folder for completed downloads (only snatched ones).")
 
     with postprocessor_lock:
-
         myDB = db.DBConnection()
         snatched = myDB.select('SELECT * from snatched WHERE Status="Snatched"')
 
         for album in snatched:
-
             if album['FolderName']:
-
                 if album['Kind'] == 'nzb':
                     download_dir = headphones.CONFIG.DOWNLOAD_DIR
                 else:
@@ -50,13 +49,14 @@ def checkFolder():
 
                 album_path = os.path.join(download_dir, album['FolderName']).encode(headphones.SYS_ENCODING, 'replace')
                 logger.info("Checking if %s exists" % album_path)
+
                 if os.path.exists(album_path):
                     logger.info('Found "' + album['FolderName'] + '" in ' + album['Kind'] + ' download folder. Verifying....')
                     verify(album['AlbumID'], album_path, album['Kind'])
-
             else:
                 logger.info("No folder name found for " + album['Title'])
 
+    logger.info("Checking download folder finished.")
 
 def verify(albumid, albumpath, Kind=None, forced=False):
 
@@ -99,6 +99,11 @@ def verify(albumid, albumpath, Kind=None, forced=False):
                     "but database is frozen. Will skip postprocessing for " \
                     "album with rgid: %s", release_dict['artist_name'],
                     release_dict['artist_id'], albumid)
+
+                myDB.action('UPDATE snatched SET status = "Frozen" WHERE status NOT LIKE "Seed%" and AlbumID=?', [albumid])
+                frozen = re.search(r' \(Frozen\)(?:\[\d+\])?', albumpath)
+                if not frozen:
+                    renameUnprocessedFolder(albumpath, tag="Frozen")
                 return
 
         logger.info(u"Now adding/updating artist: " + release_dict['artist_name'])
@@ -189,7 +194,7 @@ def verify(albumid, albumpath, Kind=None, forced=False):
             myDB.action('UPDATE snatched SET status = "Unprocessed" WHERE status NOT LIKE "Seed%" and AlbumID=?', [albumid])
             processed = re.search(r' \(Unprocessed\)(?:\[\d+\])?', albumpath)
             if not processed:
-                renameUnprocessedFolder(albumpath)
+                renameUnprocessedFolder(albumpath, tag="Unprocessed")
             return
 
     # test #1: metadata - usually works
@@ -273,9 +278,7 @@ def verify(albumid, albumpath, Kind=None, forced=False):
     myDB.action('UPDATE snatched SET status = "Unprocessed" WHERE status NOT LIKE "Seed%" and AlbumID=?', [albumid])
     processed = re.search(r' \(Unprocessed\)(?:\[\d+\])?', albumpath)
     if not processed:
-        renameUnprocessedFolder(albumpath)
-    else:
-        logger.info(u"Already marked as unprocessed: " + albumpath.decode(headphones.SYS_ENCODING, 'replace'))
+        renameUnprocessedFolder(albumpath, tag="Unprocessed")
 
 
 def doPostProcessing(albumid, albumpath, release, tracks, downloaded_track_list, Kind=None):
@@ -329,9 +332,10 @@ def doPostProcessing(albumid, albumpath, release, tracks, downloaded_track_list,
            headphones.CONFIG.MOVE_FILES:
 
             try:
-                with open(downloaded_track, "a+b"):
-                    pass
-            except IOError:
+                with open(downloaded_track, "a+b") as fp:
+                    fp.seek(0)
+            except IOError as e:
+                logger.debug("Write check exact error: %s", e)
                 logger.error("Track file is not writeable. This is required " \
                     "for some post processing steps: %s. Not continuing.",
                     downloaded_track.decode(headphones.SYS_ENCODING, "replace"))
@@ -484,11 +488,15 @@ def doPostProcessing(albumid, albumpath, release, tracks, downloaded_track_list,
         twitter.notify_download(pushmessage)
 
     if headphones.CONFIG.OSX_NOTIFY_ENABLED:
+        from headphones import cache
+        c = cache.Cache()
+        album_art = c.get_artwork_from_cache(None, release['AlbumID'])
         logger.info(u"Sending OS X notification")
         osx_notify = notifiers.OSX_NOTIFY()
         osx_notify.notify(release['ArtistName'],
                           release['AlbumTitle'],
-                          statusmessage)
+                          statusmessage,
+                          image=album_art)
 
     if headphones.CONFIG.BOXCAR_ENABLED:
         logger.info(u"Sending Boxcar2 notification")
@@ -1024,30 +1032,34 @@ def updateFilePermissions(albumpaths):
                     continue
 
 
-def renameUnprocessedFolder(albumpath):
+def renameUnprocessedFolder(path, tag):
+    """
+    Rename a unprocessed folder to a new unique name to indicate a certain
+    status.
+    """
 
-    i = 0
-    while True:
+    for i in itertools.count():
         if i == 0:
-            new_folder_name = albumpath + ' (Unprocessed)'
+            new_path = "%s (%s)" % (path, tag)
         else:
-            new_folder_name = albumpath + ' (Unprocessed)[%i]' % i
+            new_path = "%s (%s[%d])" % (path, tag, i)
 
-        if os.path.exists(new_folder_name):
+        if os.path.exists(new_path):
             i += 1
-
         else:
-            os.rename(albumpath, new_folder_name)
+            os.rename(path, new_path)
             return
 
 
 def forcePostProcess(dir=None, expand_subfolders=True, album_dir=None):
 
+    ignored = 0
+
     if album_dir:
         folders = [album_dir.encode(headphones.SYS_ENCODING, 'replace')]
-
     else:
         download_dirs = []
+
         if dir:
             download_dirs.append(dir.encode(headphones.SYS_ENCODING, 'replace'))
         if headphones.CONFIG.DOWNLOAD_DIR and not dir:
@@ -1066,7 +1078,13 @@ def forcePostProcess(dir=None, expand_subfolders=True, album_dir=None):
             if not os.path.isdir(download_dir):
                 logger.warn('Directory %s does not exist. Skipping', download_dir)
                 continue
-            for folder in os.listdir(download_dir):
+
+            # Scan for subfolders
+            subfolders = os.listdir(download_dir)
+            ignored += helpers.path_filter_patterns(subfolders,
+                headphones.CONFIG.IGNORED_FOLDERS, root=download_dir)
+
+            for folder in subfolders:
                 path_to_folder = os.path.join(download_dir, folder)
 
                 if os.path.isdir(path_to_folder):
@@ -1079,8 +1097,9 @@ def forcePostProcess(dir=None, expand_subfolders=True, album_dir=None):
 
     # Log number of folders
     if folders:
-        logger.info('Found %i folders to process.', len(folders))
         logger.debug('Expanded post processing folders: %s', folders)
+        logger.info('Found %d folders to process (%d ignored).',
+            len(folders), ignored)
     else:
         logger.info('Found no folders to process. Aborting.')
         return
@@ -1221,4 +1240,4 @@ def forcePostProcess(dir=None, expand_subfolders=True, album_dir=None):
         logger.info("Couldn't parse '%s' into any valid format. If adding " \
             "albums from another source, they must be in an 'Artist - Album " \
             "[Year]' format, or end with the musicbrainz release group id.",
-            folder_basename)
+            folder_basename)

headphones/request.py

@@ -18,21 +18,22 @@ from headphones import logger
 from xml.dom import minidom
 from bs4 import BeautifulSoup
 
-import time
 import requests
 import feedparser
 import headphones
+import headphones.lock
 import collections
 
 # Dictionary with last request times, for rate limiting.
 last_requests = collections.defaultdict(int)
+fake_lock = headphones.lock.FakeLock()
 
 
 def request_response(url, method="get", auto_raise=True,
-    whitelist_status_code=None, rate_limit=None, **kwargs):
+                     whitelist_status_code=None, lock=fake_lock, **kwargs):
     """
-    Convenient wrapper for `requests.get', which will capture the exceptions and
-    log them. On success, the Response object is returned. In case of a
+    Convenient wrapper for `requests.get', which will capture the exceptions
+    and log them. On success, the Response object is returned. In case of a
     exception, None is returned.
 
     Additionally, there is support for rate limiting. To use this feature,
@@ -50,40 +51,16 @@ def request_response(url, method="get", auto_raise=True,
     # pose a security issue!
     kwargs["verify"] = bool(headphones.CONFIG.VERIFY_SSL_CERT)
 
-    # Map method to the request.XXX method. This is a simple hack, but it allows
-    # requests to apply more magic per method. See lib/requests/api.py.
+    # Map method to the request.XXX method. This is a simple hack, but it
+    # allows requests to apply more magic per method. See lib/requests/api.py.
     request_method = getattr(requests, method.lower())
 
-    # Enfore request rate limit if applicable. This uses the lock so there
-    # is synchronized access to the API. When N threads enter this method, the
-    # first will pass trough, since there there was no last request recorded.
-    # The last request time will be set. Then, the second thread will unlock,
-    # and see that the last request was X seconds ago. It will sleep
-    # (request_limit - X) seconds, and then continue. Then the third one will
-    # unblock, and so on. After all threads finished, the total time will at
-    # least be (N * request_limit) seconds. If some request takes longer than
-    # request_limit seconds, the next unblocked thread will wait less.
-    if rate_limit:
-        lock, request_limit = rate_limit
-
-        with lock:
-            delta = time.time() - last_requests[lock]
-            limit = int(1.0 / request_limit)
-
-            if delta < request_limit:
-                logger.debug("Sleeping %.2f seconds for request, limit " \
-                    "is %d req/sec.", request_limit - delta, limit)
-
-                # Sleep the remaining time
-                time.sleep(request_limit - delta)
-
-            # Update last request time.
-            last_requests[lock] = time.time()
-
     try:
         # Request URL and wait for response
-        logger.debug("Requesting URL via %s method: %s", method.upper(), url)
-        response = request_method(url, **kwargs)
+        with lock:
+            logger.debug(
+                "Requesting URL via %s method: %s", method.upper(), url)
+            response = request_method(url, **kwargs)
 
         # If status code != OK, then raise exception, except if the status code
         # is white listed.
@@ -92,19 +69,33 @@ def request_response(url, method="get", auto_raise=True,
                 try:
                     response.raise_for_status()
                 except:
-                    logger.debug("Response status code %d is not white " \
-                         "listed, raised exception", response.status_code)
+                    logger.debug(
+                        "Response status code %d is not white "
+                        "listed, raised exception", response.status_code)
                     raise
         elif auto_raise:
             response.raise_for_status()
 
         return response
+    except requests.exceptions.SSLError as e:
+        if kwargs["verify"]:
+            logger.error(
+                "Unable to connect to remote host because of a SSL error. "
+                "It is likely that your system cannot verify the validity"
+                "of the certificate. The remote certificate is either "
+                "self-signed, or the remote server uses SNI. See the wiki for "
+                "more information on this topic.")
+        else:
+            logger.error(
+                "SSL error raised during connection, with certificate "
+                "verification turned off: %s", e)
     except requests.ConnectionError:
-        logger.error("Unable to connect to remote host. Check if the remote " \
+        logger.error(
+            "Unable to connect to remote host. Check if the remote "
             "host is up and running.")
     except requests.Timeout:
-        logger.error("Request timed out. The remote host did not respeond " \
-            "timely.")
+        logger.error(
+            "Request timed out. The remote host did not respond timely.")
     except requests.HTTPError as e:
         if e.response is not None:
             if e.response.status_code >= 500:
@@ -115,7 +106,8 @@ def request_response(url, method="get", auto_raise=True,
                 # I don't think we will end up here, but for completeness
                 cause = "unknown"
 
-            logger.error("Request raise HTTP error with status code %d (%s).",
+            logger.error(
+                "Request raise HTTP error with status code %d (%s).",
                 e.response.status_code, cause)
 
             # Debug response
@@ -157,8 +149,8 @@ def request_json(url, **kwargs):
     Wrapper for `request_response', which will decode the response as JSON
     object and return the result, if no exceptions are raised.
 
-    As an option, a validator callback can be given, which should return True if
-    the result is valid.
+    As an option, a validator callback can be given, which should return True
+    if the result is valid.
     """
 
     validator = kwargs.pop("validator", None)
@@ -223,9 +215,9 @@ def server_message(response):
         # Find body and cleanup common tags to grab content, which probably
         # contains the message.
         message = soup.find("body")
+        elements = ("header", "script", "footer", "nav", "input", "textarea")
 
-        for element in ["header", "script", "footer", "nav", "input",
-            "textarea"]:
+        for element in elements:
 
             for tag in soup.find_all(element):
                 tag.replaceWith("")

headphones/searcher.py

@@ -28,6 +28,7 @@ import re
 import string
 import shutil
 import random
+import urllib
 import headphones
 import subprocess
 import unicodedata
@@ -128,7 +129,8 @@ def read_torrent_name(torrent_file, default_name=None):
 
 def calculate_torrent_hash(link, data=None):
     """
-    Calculate the torrent hash from a magnet link or data.
+    Calculate the torrent hash from a magnet link or data. Raises a ValueError
+    when it cannot create a torrent hash given the input data.
     """
 
     if link.startswith("magnet:"):
@@ -142,7 +144,7 @@ def calculate_torrent_hash(link, data=None):
         raise ValueError("Cannot calculate torrent hash without magnet link " \
             "or data")
 
-    return torrent_hash
+    return torrent_hash.upper()
 
 
 def get_seed_ratio(provider):
@@ -159,6 +161,8 @@ def get_seed_ratio(provider):
         seed_ratio = headphones.CONFIG.WHATCD_RATIO
     elif provider == 'The Pirate Bay':
         seed_ratio = headphones.CONFIG.PIRATEBAY_RATIO
+    elif provider == 'Old Pirate Bay':
+        seed_ratio = headphones.CONFIG.OLDPIRATEBAY_RATIO
     elif provider == 'Waffles.fm':
         seed_ratio = headphones.CONFIG.WAFFLES_RATIO
     elif provider == 'Mininova':
@@ -175,16 +179,18 @@ def get_seed_ratio(provider):
     return seed_ratio
 
 
-def searchforalbum(albumid=None, new=False, losslessOnly=False, choose_specific_download=False):
+def searchforalbum(albumid=None, new=False, losslessOnly=False,
+    choose_specific_download=False):
+
+    logger.info('Searching for wanted albums')
     myDB = db.DBConnection()
 
     if not albumid:
         results = myDB.select('SELECT * from albums WHERE Status="Wanted" OR Status="Wanted Lossless"')
 
         for album in results:
-
             if not album['AlbumTitle'] or not album['ArtistName']:
-                logger.warn('Skipping release %s. No title available' % album['AlbumID'])
+                logger.warn('Skipping release %s. No title available', album['AlbumID'])
                 continue
 
             new = True
@@ -196,7 +202,6 @@ def searchforalbum(albumid=None, new=False, losslessOnly=False, choose_specific_
             do_sorted_search(album, new, losslessOnly)
 
     elif albumid and choose_specific_download:
-
         album = myDB.action('SELECT * from albums WHERE AlbumID=?', [albumid]).fetchone()
         logger.info('Searching for "%s - %s"' % (album['ArtistName'], album['AlbumTitle']))
         results = do_sorted_search(album, new, losslessOnly, choose_specific_download=True)
@@ -207,14 +212,14 @@ def searchforalbum(albumid=None, new=False, losslessOnly=False, choose_specific_
         logger.info('Searching for "%s - %s" since it was marked as wanted' % (album['ArtistName'], album['AlbumTitle']))
         do_sorted_search(album, new, losslessOnly)
 
-    logger.info('Search for Wanted albums complete')
+    logger.info('Search for wanted albums complete')
 
 
 def do_sorted_search(album, new, losslessOnly, choose_specific_download=False):
 
     NZB_PROVIDERS = (headphones.CONFIG.HEADPHONES_INDEXER or headphones.CONFIG.NEWZNAB or headphones.CONFIG.NZBSORG or headphones.CONFIG.OMGWTFNZBS)
     NZB_DOWNLOADERS = (headphones.CONFIG.SAB_HOST or headphones.CONFIG.BLACKHOLE_DIR or headphones.CONFIG.NZBGET_HOST)
-    TORRENT_PROVIDERS = (headphones.CONFIG.KAT or headphones.CONFIG.PIRATEBAY or headphones.CONFIG.MININOVA or headphones.CONFIG.WAFFLES or headphones.CONFIG.RUTRACKER or headphones.CONFIG.WHATCD)
+    TORRENT_PROVIDERS = (headphones.CONFIG.KAT or headphones.CONFIG.PIRATEBAY or headphones.CONFIG.OLDPIRATEBAY or headphones.CONFIG.MININOVA or headphones.CONFIG.WAFFLES or headphones.CONFIG.RUTRACKER or headphones.CONFIG.WHATCD)
 
     results = []
     myDB = db.DBConnection()
@@ -258,6 +263,10 @@ def do_sorted_search(album, new, losslessOnly, choose_specific_download=False):
     if choose_specific_download:
         return results
 
+    # Filter all results that do not comply
+    results = [ result for result in results if result[5] ]
+
+    # Sort the remaining results
     sorted_search_results = sort_search_results(results, album, new, albumlength)
 
     if not sorted_search_results:
@@ -500,9 +509,8 @@ def searchNZB(album, new=False, losslessOnly=False, albumlength=None):
                         title = item.title
                         size = int(item.links[1]['length'])
 
-                        resultlist.append((title, size, url, provider, 'nzb'))
+                        resultlist.append((title, size, url, provider, 'nzb', True))
                         logger.info('Found %s. Size: %s' % (title, helpers.bytes_to_mb(size)))
-
                     except Exception as e:
                         logger.error(u"An unknown error occurred trying to parse the feed: %s" % e)
 
@@ -572,7 +580,7 @@ def searchNZB(album, new=False, losslessOnly=False, albumlength=None):
                             size = int(item.links[1]['length'])
                             if all(word.lower() in title.lower() for word in term.split()):
                                 logger.info('Found %s. Size: %s' % (title, helpers.bytes_to_mb(size)))
-                                resultlist.append((title, size, url, provider, 'nzb'))
+                                resultlist.append((title, size, url, provider, 'nzb', True))
                             else:
                                 logger.info('Skipping %s, not all search term words found' % title)
 
@@ -593,7 +601,7 @@ def searchNZB(album, new=False, losslessOnly=False, albumlength=None):
             logger.info("Album type is audiobook/spokenword. Using audiobook category")
 
         # Request results
-        logger.info('Parsing results from nzbs.org')
+        logger.info('Requesting from nzbs.org')
 
         headers = {'User-Agent': USER_AGENT}
         params = {
@@ -606,9 +614,11 @@ def searchNZB(album, new=False, losslessOnly=False, albumlength=None):
 
         data = request.request_feed(
             url='http://beta.nzbs.org/api',
-            params=params, headers=headers
+            params=params, headers=headers,
+            timeout=5
         )
 
+        logger.info('Parsing results from nzbs.org')
         # Process feed
         if data:
             if not len(data.entries):
@@ -620,7 +630,7 @@ def searchNZB(album, new=False, losslessOnly=False, albumlength=None):
                         title = item.title
                         size = int(item.links[1]['length'])
 
-                        resultlist.append((title, size, url, provider, 'nzb'))
+                        resultlist.append((title, size, url, provider, 'nzb', True))
                         logger.info('Found %s. Size: %s' % (title, helpers.bytes_to_mb(size)))
                     except Exception as e:
                         logger.exception("Unhandled exception while parsing feed")
@@ -667,7 +677,7 @@ def searchNZB(album, new=False, losslessOnly=False, albumlength=None):
                         title = item['release']
                         size = int(item['sizebytes'])
 
-                        resultlist.append((title, size, url, provider, 'nzb'))
+                        resultlist.append((title, size, url, provider, 'nzb', True))
                         logger.info('Found %s. Size: %s', title, helpers.bytes_to_mb(size))
                     except Exception as e:
                         logger.exception("Unhandled exception")
@@ -922,9 +932,15 @@ def send_to_downloader(data, bestqual, album):
         pushalot = notifiers.PUSHALOT()
         pushalot.notify(name, "Download started")
     if headphones.CONFIG.OSX_NOTIFY_ENABLED and headphones.CONFIG.OSX_NOTIFY_ONSNATCH:
+        from headphones import cache
+        c = cache.Cache()
+        album_art = c.get_artwork_from_cache(None, rgid)
         logger.info(u"Sending OS X notification")
         osx_notify = notifiers.OSX_NOTIFY()
-        osx_notify.notify(artist, albumname, 'Snatched: ' + provider + '. ' + name)
+        osx_notify.notify(artist,
+                          albumname,
+                          'Snatched: ' + provider + '. ' + name,
+                          image=album_art)
     if headphones.CONFIG.BOXCAR_ENABLED and headphones.CONFIG.BOXCAR_ONSNATCH:
         logger.info(u"Sending Boxcar2 notification")
         b2msg = 'From ' + provider + '<br></br>' + name
@@ -1002,7 +1018,6 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
     global gazelle  # persistent what.cd api object to reduce number of login attempts
 
     # rutracker login
-
     if headphones.CONFIG.RUTRACKER and album:
         rulogin = rutracker.login(headphones.CONFIG.RUTRACKER_USER, headphones.CONFIG.RUTRACKER_PASSWORD)
         if not rulogin:
@@ -1077,7 +1092,7 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
         if headphones.CONFIG.KAT_PROXY_URL:
             providerurl = fix_url(set_proxy(headphones.CONFIG.KAT_PROXY_URL))
         else:
-            providerurl = fix_url("https://kickass.to")
+            providerurl = fix_url("https://kickass.so")
 
         # Build URL
         providerurl = providerurl + "/usearch/" + ka_term
@@ -1123,10 +1138,13 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
                                 rightformat = False
 
                         if rightformat and size < maxsize and minimumseeders < int(seeders):
-                            resultlist.append((title, size, url, provider, 'torrent'))
+                            match = True
                             logger.info('Found %s. Size: %s' % (title, helpers.bytes_to_mb(size)))
                         else:
+                            match = False
                             logger.info('%s is larger than the maxsize, the wrong format or has too little seeders for this category, skipping. (Size: %i bytes, Seeders: %d, Format: %s)', title, size, int(seeders), rightformat)
+
+                        resultlist.append((title, size, url, provider, 'torrent', match))
                     except Exception as e:
                         logger.exception("Unhandled exception in the KAT parser")
 
@@ -1190,14 +1208,13 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
                         desc_match = re.search(r"Size: (\d+)<", item.description)
                         size = int(desc_match.group(1))
                         url = item.link
-                        resultlist.append((title, size, url, provider, 'torrent'))
+                        resultlist.append((title, size, url, provider, 'torrent', True))
                         logger.info('Found %s. Size: %s', title, helpers.bytes_to_mb(size))
                     except Exception as e:
                         logger.error(u"An error occurred while trying to parse the response from Waffles.fm: %s", e)
 
     # rutracker.org
     if headphones.CONFIG.RUTRACKER and rulogin:
-
         provider = "rutracker.org"
 
         # Ignore if release date not specified, results too unpredictable
@@ -1232,7 +1249,7 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
                     title = ru[0].decode('utf-8')
                     size = ru[1]
                     url = ru[2]
-                    resultlist.append((title, size, url, provider, 'torrent'))
+                    resultlist.append((title, size, url, provider, 'torrent', True))
                     logger.info('Found %s. Size: %s' % (title, helpers.bytes_to_mb(size)))
             else:
                 logger.info(u"No valid results found from %s" % (provider))
@@ -1312,7 +1329,7 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
                                    torrent.size,
                                    gazelle.generate_torrent_link(torrent.id),
                                    provider,
-                                   'torrent'))
+                                   'torrent', True))
 
     # Pirate Bay
     if headphones.CONFIG.PIRATEBAY:
@@ -1342,7 +1359,7 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
         # Request content
         logger.info("Searching The Pirate Bay using term: %s", tpb_term)
 
-        headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/37.0.2049.0 Safari/537.36'}
+        headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2243.2 Safari/537.36'}
         data = request.request_soup(url=providerurl + category, headers=headers)
 
         # Process content
@@ -1378,13 +1395,69 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
                         size = helpers.piratesize(formatted_size)
 
                         if size < maxsize and minimumseeders < seeds and url is not None:
-                            resultlist.append((title, size, url, provider, "torrent"))
+                            match = True
                             logger.info('Found %s. Size: %s' % (title, formatted_size))
                         else:
+                            match = False
                             logger.info('%s is larger than the maxsize or has too little seeders for this category, skipping. (Size: %i bytes, Seeders: %i)' % (title, size, int(seeds)))
+
+                        resultlist.append((title, size, url, provider, "torrent", match))
                     except Exception as e:
                         logger.error(u"An unknown error occurred in the Pirate Bay parser: %s" % e)
 
+    # Old Pirate Bay Compatible
+    if headphones.CONFIG.OLDPIRATEBAY:
+        provider = "Old Pirate Bay"
+        tpb_term = term.replace("!", "")
+
+        # Pick category for torrents
+        if headphones.CONFIG.PREFERRED_QUALITY == 3 or losslessOnly:
+            maxsize = 10000000000
+        elif headphones.CONFIG.PREFERRED_QUALITY == 1 or allow_lossless:
+            maxsize = 10000000000
+        else:
+            maxsize = 300000000
+
+        # Requesting content
+        logger.info("Parsing results from Old Pirate Bay")
+
+        headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2243.2 Safari/537.36'}
+        provider_url = fix_url(headphones.CONFIG.OLDPIRATEBAY_URL) + \
+            "/search.php?" + urllib.urlencode({"q": tpb_term, "iht": 6})
+
+        data = request.request_soup(url=provider_url, headers=headers)
+
+        # Process content
+        if data:
+            rows = data.select('table tbody tr')
+
+            if not rows:
+                logger.info("No results found")
+            else:
+                for item in rows:
+                    try:
+                        links = item.select("td.title-row a")
+
+                        rightformat = True
+                        title = links[1].text
+                        seeds = int(item.select("td.seeders-row")[0].text)
+                        url = links[0]["href"] # Magnet link. The actual download link is not based on the URL
+
+                        formatted_size = item.select("td.size-row")[0].text
+                        size = helpers.piratesize(formatted_size)
+
+                        if size < maxsize and minimumseeders < seeds and url is not None:
+                            match = True
+                            logger.info('Found %s. Size: %s' % (title, formatted_size))
+                        else:
+                            match = False
+                            logger.info('%s is larger than the maxsize or has too little seeders for this category, skipping. (Size: %i bytes, Seeders: %i)' % (title, size, int(seeds)))
+
+                        resultlist.append((title, size, url, provider, "torrent", match))
+                    except Exception as e:
+                        logger.error(u"An unknown error occurred in the Old Pirate Bay parser: %s" % e)
+
+    # Mininova
     if headphones.CONFIG.MININOVA:
         provider = "Mininova"
         providerurl = fix_url("http://www.mininova.org/rss/" + term + "/5")
@@ -1430,12 +1503,15 @@ def searchTorrent(album, new=False, losslessOnly=False, albumlength=None):
                             torrent = request.request_content(url)
                             if not torrent or (int(torrent.find(".mp3")) > 0 and int(torrent.find(".flac")) < 1):
                                 rightformat = False
+
                         if rightformat and size < maxsize and minimumseeders < seeds:
-                            resultlist.append((title, size, url, provider, 'torrent'))
+                            match = True
                             logger.info('Found %s. Size: %s' % (title, helpers.bytes_to_mb(size)))
                         else:
+                            match = False
                             logger.info('%s is larger than the maxsize, the wrong format or has too little seeders for this category, skipping. (Size: %i bytes, Seeders: %i, Format: %s)' % (title, size, int(seeds), rightformat))
 
+                        resultlist.append((title, size, url, provider, 'torrent', match))
                     except Exception as e:
                         logger.exception("Unhandled exception in Mininova Parser")
 
@@ -1474,8 +1550,8 @@ def preprocess(resultlist):
                 headers['Referer'] = 'http://kat.ph/'
             elif result[3] == 'What.cd':
                 headers['User-Agent'] = 'Headphones'
-            elif result[3] == "The Pirate Bay":
-                headers['User-Agent'] = 'Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/37.0.2049.0 Safari/537.36'
+            elif result[3] == "The Pirate Bay" or result[3] == "Old Pirate Bay":
+                headers['User-Agent'] = 'Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2243.2 Safari/537.36'
             return request.request_content(url=result[2], headers=headers), result
 
         else:

headphones/torrentfinished.py

@@ -13,21 +13,21 @@
 #  You should have received a copy of the GNU General Public License
 #  along with Headphones.  If not, see <http://www.gnu.org/licenses/>.
 
+from headphones import db, utorrent, transmission, logger
+
 import threading
 import headphones
-from headphones import db, utorrent, transmission, logger
 
 postprocessor_lock = threading.Lock()
 
-# Remove Torrent + data if Post Processed and finished Seeding
-
-
 def checkTorrentFinished():
+    """
+    Remove Torrent + data if Post Processed and finished Seeding
+    """
 
     logger.info("Checking if any torrents have finished seeding and can be removed")
 
     with postprocessor_lock:
-
         myDB = db.DBConnection()
         results = myDB.select('SELECT * from snatched WHERE Status="Seed_Processed"')
 
@@ -42,3 +42,5 @@ def checkTorrentFinished():
 
             if torrent_removed:
                 myDB.action('DELETE from snatched WHERE status = "Seed_Processed" and AlbumID=?', [albumid])
+
+    logger.info("Checking finished torrents completed")

headphones/transmission.py

@@ -18,6 +18,7 @@ from headphones import logger, request
 import time
 import json
 import base64
+import urlparse
 import headphones
 
 # This is just a simple script to send torrents to transmission. The
@@ -126,7 +127,6 @@ def torrentAction(method, arguments):
     host = headphones.CONFIG.TRANSMISSION_HOST
     username = headphones.CONFIG.TRANSMISSION_USERNAME
     password = headphones.CONFIG.TRANSMISSION_PASSWORD
-    sessionid = None
 
     if not host.startswith('http'):
         host = 'http://' + host
@@ -134,30 +134,22 @@ def torrentAction(method, arguments):
     if host.endswith('/'):
         host = host[:-1]
 
-    # Either the host ends with a port, or some directory, or rpc
-    # If it ends with /rpc we don't have to do anything
-    # If it ends with a port we add /transmission/rpc
-    # anything else we just add rpc
-    if not host.endswith('/rpc'):
-        # Check if it ends in a port number
-        i = host.rfind(':')
-        if i >= 0:
-            possible_port = host[i + 1:]
-            host = host + "/rpc"
-            try:
-                port = int(possible_port)
-                if port:
-                    host = host + "/transmission/rpc"
-            except ValueError:
-                logger.debug('No port, assuming not transmission')
-        else:
-            logger.error('Transmission port missing')
-            return
+    # Fix the URL. We assume that the user does not point to the RPC endpoint,
+    # so add it if it is missing.
+    parts = list(urlparse.urlparse(host))
+
+    if not parts[0] in ("http", "https"):
+        parts[0] = "http"
+
+    if not parts[2].endswith("/rpc"):
+        parts[2] += "/transmission/rpc"
+
+    host = urlparse.urlunparse(parts)
 
     # Retrieve session id
     auth = (username, password) if username and password else None
-
-    response = request.request_response(host, auth=auth, whitelist_status_code=[401, 409])
+    response = request.request_response(host, auth=auth,
+        whitelist_status_code=[401, 409])
 
     if response is None:
         logger.error("Error gettings Transmission session ID")
@@ -166,23 +158,27 @@ def torrentAction(method, arguments):
     # Parse response
     if response.status_code == 401:
         if auth:
-            logger.error("Username and/or password not accepted by Transmission")
+            logger.error("Username and/or password not accepted by " \
+                "Transmission")
         else:
             logger.error("Transmission authorization required")
 
         return
     elif response.status_code == 409:
-        sessionid = response.headers['x-transmission-session-id']
+        session_id = response.headers['x-transmission-session-id']
 
-    if not sessionid:
-        logger.error("Error getting Session ID from Transmission")
-        return
+        if not session_id:
+            logger.error("Expected a Session ID from Transmission")
+            return
 
     # Prepare next request
-    headers = {'x-transmission-session-id': sessionid}
+    headers = {'x-transmission-session-id': session_id}
     data = {'method': method, 'arguments': arguments}
 
-    response = request.request_json(host, method="post", data=json.dumps(data), headers=headers, auth=auth)
+    response = request.request_json(host, method="POST", data=json.dumps(data),
+        headers=headers, auth=auth)
+
+    print response
 
     if not response:
         logger.error("Error sending torrent to Transmission")

headphones/updater.py

@@ -20,11 +20,10 @@ def dbUpdate(forcefull=False):
 
     myDB = db.DBConnection()
 
-    activeartists = myDB.select('SELECT ArtistID, ArtistName from artists WHERE Status="Active" or Status="Loading" order by LastUpdated ASC')
-    logger.info('Starting update for %i active artists' % len(activeartists))
-
-    for artist in activeartists:
+    active_artists = myDB.select('SELECT ArtistID, ArtistName from artists WHERE Status="Active" or Status="Loading" order by LastUpdated ASC')
+    logger.info('Starting update for %i active artists', len(active_artists))
 
+    for artist in active_artists:
         artistid = artist[0]
         importer.addArtisttoDB(artistid=artistid, extrasonly=False, forcefull=forcefull)
 

headphones/utorrent.py

@@ -215,7 +215,7 @@ def dirTorrent(hash, cacheid=None, return_name=None):
     cacheid = torrentList['torrentc']
 
     for torrent in torrents:
-        if (torrent[0].lower() == hash):
+        if torrent[0].upper() == hash:
             if not return_name:
                 return torrent[26], cacheid
             else:

headphones/webserve.py

@@ -23,13 +23,17 @@ from mako import exceptions
 
 from operator import itemgetter
 
-import os
-import sys
+import headphones
+import threading
+import cherrypy
+import urllib2
+import hashlib
+import random
+import urllib
 import json
 import time
-import cherrypy
-import threading
-import headphones
+import sys
+import os
 
 try:
     # pylint:disable=E0611
@@ -57,36 +61,35 @@ def serve_template(templatename, **kwargs):
 
 class WebInterface(object):
 
+    @cherrypy.expose
     def index(self):
         raise cherrypy.HTTPRedirect("home")
-    index.exposed = True
 
+    @cherrypy.expose
     def home(self):
         myDB = db.DBConnection()
         artists = myDB.select('SELECT * from artists order by ArtistSortName COLLATE NOCASE')
         return serve_template(templatename="index.html", title="Home", artists=artists)
-    home.exposed = True
 
+    @cherrypy.expose
     def artistPage(self, ArtistID):
         myDB = db.DBConnection()
         artist = myDB.action('SELECT * FROM artists WHERE ArtistID=?', [ArtistID]).fetchone()
-        albums = myDB.select('SELECT * from albums WHERE ArtistID=? order by ReleaseDate DESC', [ArtistID])
 
         # Don't redirect to the artist page until it has the bare minimum info inserted
         # Redirect to the home page if we still can't get it after 5 seconds
         retry = 0
 
-        while retry < 5:
-            if not artist:
-                time.sleep(1)
-                artist = myDB.action('SELECT * FROM artists WHERE ArtistID=?', [ArtistID]).fetchone()
-                retry += 1
-            else:
-                break
+        while not artist and retry < 5:
+            time.sleep(1)
+            artist = myDB.action('SELECT * FROM artists WHERE ArtistID=?', [ArtistID]).fetchone()
+            retry += 1
 
         if not artist:
             raise cherrypy.HTTPRedirect("home")
 
+        albums = myDB.select('SELECT * from albums WHERE ArtistID=? order by ReleaseDate DESC', [ArtistID])
+
         # Serve the extras up as a dict to make things easier for new templates (append new extras to the end)
         extras_list = headphones.POSSIBLE_EXTRAS
         if artist['Extras']:
@@ -105,13 +108,11 @@ class WebInterface(object):
             i += 1
 
         return serve_template(templatename="artist.html", title=artist['ArtistName'], artist=artist, albums=albums, extras=extras_dict)
-    artistPage.exposed = True
 
+    @cherrypy.expose
     def albumPage(self, AlbumID):
         myDB = db.DBConnection()
         album = myDB.action('SELECT * from albums WHERE AlbumID=?', [AlbumID]).fetchone()
-        tracks = myDB.select('SELECT * from tracks WHERE AlbumID=? ORDER BY CAST(TrackNumber AS INTEGER)', [AlbumID])
-        description = myDB.action('SELECT * from descriptions WHERE ReleaseGroupID=?', [AlbumID]).fetchone()
 
         retry = 0
         while retry < 5:
@@ -125,6 +126,9 @@ class WebInterface(object):
         if not album:
             raise cherrypy.HTTPRedirect("home")
 
+        tracks = myDB.select('SELECT * from tracks WHERE AlbumID=? ORDER BY CAST(TrackNumber AS INTEGER)', [AlbumID])
+        description = myDB.action('SELECT * from descriptions WHERE ReleaseGroupID=?', [AlbumID]).fetchone()
+
         if not album['ArtistName']:
             title = ' - '
         else:
@@ -134,8 +138,8 @@ class WebInterface(object):
         else:
             title = title + album['AlbumTitle']
         return serve_template(templatename="album.html", title=title, album=album, tracks=tracks, description=description)
-    albumPage.exposed = True
 
+    @cherrypy.expose
     def search(self, name, type):
         if len(name) == 0:
             raise cherrypy.HTTPRedirect("home")
@@ -144,13 +148,15 @@ class WebInterface(object):
         else:
             searchresults = mb.findRelease(name, limit=100)
         return serve_template(templatename="searchresults.html", title='Search Results for: "' + name + '"', searchresults=searchresults, name=name, type=type)
-    search.exposed = True
 
+    @cherrypy.expose
     def addArtist(self, artistid):
-        threading.Thread(target=importer.addArtisttoDB, args=[artistid]).start()
+        thread = threading.Thread(target=importer.addArtisttoDB, args=[artistid])
+        thread.start()
+        thread.join(1)
         raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % artistid)
-    addArtist.exposed = True
 
+    @cherrypy.expose
     def getExtras(self, ArtistID, newstyle=False, **kwargs):
         # if calling this function without the newstyle, they're using the old format
         # which doesn't separate extras, so we'll grab all of them
@@ -172,10 +178,12 @@ class WebInterface(object):
         newValueDict = {'IncludeExtras': 1,
                         'Extras': extras}
         myDB.upsert("artists", newValueDict, controlValueDict)
-        threading.Thread(target=importer.addArtisttoDB, args=[ArtistID, True, False]).start()
+        thread = threading.Thread(target=importer.addArtisttoDB, args=[ArtistID, True, False])
+        thread.start()
+        thread.join(1)
         raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
-    getExtras.exposed = True
 
+    @cherrypy.expose
     def removeExtras(self, ArtistID, ArtistName):
         myDB = db.DBConnection()
         controlValueDict = {'ArtistID': ArtistID}
@@ -193,8 +201,8 @@ class WebInterface(object):
             c.remove_from_cache(AlbumID=album['AlbumID'])
         importer.finalize_update(ArtistID, ArtistName)
         raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
-    removeExtras.exposed = True
 
+    @cherrypy.expose
     def pauseArtist(self, ArtistID):
         logger.info(u"Pausing artist: " + ArtistID)
         myDB = db.DBConnection()
@@ -202,8 +210,8 @@ class WebInterface(object):
         newValueDict = {'Status': 'Paused'}
         myDB.upsert("artists", newValueDict, controlValueDict)
         raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
-    pauseArtist.exposed = True
 
+    @cherrypy.expose
     def resumeArtist(self, ArtistID):
         logger.info(u"Resuming artist: " + ArtistID)
         myDB = db.DBConnection()
@@ -211,7 +219,6 @@ class WebInterface(object):
         newValueDict = {'Status': 'Active'}
         myDB.upsert("artists", newValueDict, controlValueDict)
         raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
-    resumeArtist.exposed = True
 
     def removeArtist(self, ArtistID):
         logger.info(u"Deleting all traces of artist: " + ArtistID)
@@ -242,24 +249,39 @@ class WebInterface(object):
         myDB.action('DELETE from descriptions WHERE ArtistID=?', [ArtistID])
         myDB.action('INSERT OR REPLACE into blacklist VALUES (?)', [ArtistID])
 
+    @cherrypy.expose
     def deleteArtist(self, ArtistID):
         self.removeArtist(ArtistID)
         raise cherrypy.HTTPRedirect("home")
-    deleteArtist.exposed = True
 
+    @cherrypy.expose
+    def scanArtist(self, ArtistID):
+        logger.info(u"Scanning artist: %s", ArtistID)
+        myDB = db.DBConnection()
+        artistname = myDB.select('SELECT DISTINCT ArtistName FROM artists WHERE ArtistID=?', [ArtistID])
+        artistfolder = os.path.join(headphones.CONFIG.DESTINATION_DIR, artistname[0][0])
+        try:
+            threading.Thread(target=librarysync.libraryScan(dir=artistfolder)).start()
+        except Exception as e:
+            logger.error('Unable to complete the scan: %s', e)
+        raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
+
+    @cherrypy.expose
     def deleteEmptyArtists(self):
         logger.info(u"Deleting all empty artists")
         myDB = db.DBConnection()
         emptyArtistIDs = [row['ArtistID'] for row in myDB.select("SELECT ArtistID FROM artists WHERE LatestAlbum IS NULL")]
         for ArtistID in emptyArtistIDs:
             self.removeArtist(ArtistID)
-    deleteEmptyArtists.exposed = True
 
+    @cherrypy.expose
     def refreshArtist(self, ArtistID):
-        threading.Thread(target=importer.addArtisttoDB, args=[ArtistID, False, True]).start()
+        thread = threading.Thread(target=importer.addArtisttoDB, args=[ArtistID, False, True])
+        thread.start()
+        thread.join(1)
         raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
-    refreshArtist.exposed = True
 
+    @cherrypy.expose
     def markAlbums(self, ArtistID=None, action=None, **args):
         myDB = db.DBConnection()
         if action == 'WantedNew' or action == 'WantedLossless':
@@ -286,8 +308,8 @@ class WebInterface(object):
             raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
         else:
             raise cherrypy.HTTPRedirect("upcoming")
-    markAlbums.exposed = True
 
+    @cherrypy.expose
     def addArtists(self, action=None, **args):
         if action == "add":
             threading.Thread(target=importer.artistlist_to_mbids, args=[args, True]).start()
@@ -298,8 +320,8 @@ class WebInterface(object):
                 myDB.action('UPDATE have SET Matched="Ignored" WHERE ArtistName=?', [artist.decode(headphones.SYS_ENCODING, 'replace')])
                 logger.info("Artist %s removed from new artist list and set to ignored" % artist)
         raise cherrypy.HTTPRedirect("home")
-    addArtists.exposed = True
 
+    @cherrypy.expose
     def queueAlbum(self, AlbumID, ArtistID=None, new=False, redirect=None, lossless=False):
         logger.info(u"Marking album: " + AlbumID + " as wanted...")
         myDB = db.DBConnection()
@@ -312,41 +334,34 @@ class WebInterface(object):
         myDB.upsert("albums", newValueDict, controlValueDict)
         searcher.searchforalbum(AlbumID, new)
         if ArtistID:
-            raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
-        else:
-            raise cherrypy.HTTPRedirect(redirect)
-    queueAlbum.exposed = True
+            redirect = "artistPage?ArtistID=%s" % ArtistID
+        raise cherrypy.HTTPRedirect(redirect)
 
+    @cherrypy.expose
     def choose_specific_download(self, AlbumID):
         results = searcher.searchforalbum(AlbumID, choose_specific_download=True)
 
         results_as_dicts = []
 
         for result in results:
-
             result_dict = {
                 'title': result[0],
                 'size': result[1],
                 'url': result[2],
                 'provider': result[3],
-                'kind': result[4]
+                'kind': result[4],
+                'matches': result[5]
             }
             results_as_dicts.append(result_dict)
-
         s = json.dumps(results_as_dicts)
         cherrypy.response.headers['Content-type'] = 'application/json'
         return s
 
-    choose_specific_download.exposed = True
-
+    @cherrypy.expose
     def download_specific_release(self, AlbumID, title, size, url, provider, kind, **kwargs):
-
         # Handle situations where the torrent url contains arguments that are parsed
         if kwargs:
-            import urllib
-            import urllib2
             url = urllib2.quote(url, safe=":?/=&") + '&' + urllib.urlencode(kwargs)
-
         try:
             result = [(title, int(size), url, provider, kind)]
         except ValueError:
@@ -360,8 +375,7 @@ class WebInterface(object):
           album = myDB.action('SELECT * from albums WHERE AlbumID=?', [AlbumID]).fetchone()
           searcher.send_to_downloader(data, bestqual, album)
 
-    download_specific_release.exposed = True
-
+    @cherrypy.expose
     def unqueueAlbum(self, AlbumID, ArtistID):
         logger.info(u"Marking album: " + AlbumID + "as skipped...")
         myDB = db.DBConnection()
@@ -369,15 +383,16 @@ class WebInterface(object):
         newValueDict = {'Status': 'Skipped'}
         myDB.upsert("albums", newValueDict, controlValueDict)
         raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
-    unqueueAlbum.exposed = True
 
+    @cherrypy.expose
     def deleteAlbum(self, AlbumID, ArtistID=None):
         logger.info(u"Deleting all traces of album: " + AlbumID)
         myDB = db.DBConnection()
 
         myDB.action('DELETE from have WHERE Matched=?', [AlbumID])
-        album = myDB.action('SELECT ArtistName, AlbumTitle from albums where AlbumID=?', [AlbumID]).fetchone()
+        album = myDB.action('SELECT ArtistID, ArtistName, AlbumTitle from albums where AlbumID=?', [AlbumID]).fetchone()
         if album:
+            ArtistID = album['ArtistID']
             myDB.action('DELETE from have WHERE ArtistName=? AND AlbumTitle=?', [album['ArtistName'], album['AlbumTitle']])
 
         myDB.action('DELETE from albums WHERE AlbumID=?', [AlbumID])
@@ -395,17 +410,18 @@ class WebInterface(object):
             raise cherrypy.HTTPRedirect("artistPage?ArtistID=%s" % ArtistID)
         else:
             raise cherrypy.HTTPRedirect("home")
-    deleteAlbum.exposed = True
 
+    @cherrypy.expose
     def switchAlbum(self, AlbumID, ReleaseID):
-        '''
-        Take the values from allalbums/alltracks (based on the ReleaseID) and swap it into the album & track tables
-        '''
+        """
+        Take the values from allalbums/alltracks (based on the ReleaseID) and
+        swap it into the album & track tables
+        """
         from headphones import albumswitcher
         albumswitcher.switch(AlbumID, ReleaseID)
         raise cherrypy.HTTPRedirect("albumPage?AlbumID=%s" % AlbumID)
-    switchAlbum.exposed = True
 
+    @cherrypy.expose
     def editSearchTerm(self, AlbumID, SearchTerm):
         logger.info(u"Updating search term for albumid: " + AlbumID)
         myDB = db.DBConnection()
@@ -413,27 +429,27 @@ class WebInterface(object):
         newValueDict = {'SearchTerm': SearchTerm}
         myDB.upsert("albums", newValueDict, controlValueDict)
         raise cherrypy.HTTPRedirect("albumPage?AlbumID=%s" % AlbumID)
-    editSearchTerm.exposed = True
 
+    @cherrypy.expose
     def upcoming(self):
         myDB = db.DBConnection()
         upcoming = myDB.select("SELECT * from albums WHERE ReleaseDate > date('now') order by ReleaseDate ASC")
         wanted = myDB.select("SELECT * from albums WHERE Status='Wanted'")
         return serve_template(templatename="upcoming.html", title="Upcoming", upcoming=upcoming, wanted=wanted)
-    upcoming.exposed = True
 
+    @cherrypy.expose
     def manage(self):
         myDB = db.DBConnection()
         emptyArtists = myDB.select("SELECT * FROM artists WHERE LatestAlbum IS NULL")
         return serve_template(templatename="manage.html", title="Manage", emptyArtists=emptyArtists)
-    manage.exposed = True
 
+    @cherrypy.expose
     def manageArtists(self):
         myDB = db.DBConnection()
         artists = myDB.select('SELECT * from artists order by ArtistSortName COLLATE NOCASE')
         return serve_template(templatename="manageartists.html", title="Manage Artists", artists=artists)
-    manageArtists.exposed = True
 
+    @cherrypy.expose
     def manageAlbums(self, Status=None):
         myDB = db.DBConnection()
         if Status == "Upcoming":
@@ -443,14 +459,14 @@ class WebInterface(object):
         else:
             albums = myDB.select('SELECT * from albums')
         return serve_template(templatename="managealbums.html", title="Manage Albums", albums=albums)
-    manageAlbums.exposed = True
 
+    @cherrypy.expose
     def manageNew(self):
         myDB = db.DBConnection()
         newartists = myDB.select('SELECT * from newartists')
         return serve_template(templatename="managenew.html", title="Manage New Artists", newartists=newartists)
-    manageNew.exposed = True
 
+    @cherrypy.expose
     def manageUnmatched(self):
         myDB = db.DBConnection()
         have_album_dictionary = []
@@ -476,8 +492,8 @@ class WebInterface(object):
         unmatchedalbums = [d for d in have_album_dictionary if (cleanName(d['ArtistName']).lower(), cleanName(d['AlbumTitle']).lower()) not in check]
 
         return serve_template(templatename="manageunmatched.html", title="Manage Unmatched Items", unmatchedalbums=unmatchedalbums)
-    manageUnmatched.exposed = True
 
+    @cherrypy.expose
     def markUnmatched(self, action=None, existing_artist=None, existing_album=None, new_artist=None, new_album=None):
         myDB = db.DBConnection()
 
@@ -563,8 +579,7 @@ class WebInterface(object):
             else:
                 logger.info("Artist %s / Album %s already named appropriately; nothing to modify" % (existing_artist, existing_album))
 
-    markUnmatched.exposed = True
-
+    @cherrypy.expose
     def manageManual(self):
         myDB = db.DBConnection()
         manual_albums = []
@@ -583,8 +598,8 @@ class WebInterface(object):
         manual_albums_sorted = sorted(manual_albums, key=itemgetter('ArtistName', 'AlbumTitle'))
 
         return serve_template(templatename="managemanual.html", title="Manage Manual Items", manualalbums=manual_albums_sorted)
-    manageManual.exposed = True
 
+    @cherrypy.expose
     def markManual(self, action=None, existing_artist=None, existing_album=None):
         myDB = db.DBConnection()
         if action == "unignoreArtist":
@@ -635,8 +650,7 @@ class WebInterface(object):
                 librarysync.update_album_status(album_id)
             logger.info("Album: %s successfully restored to unmatched list" % album)
 
-    markManual.exposed = True
-
+    @cherrypy.expose
     def markArtists(self, action=None, **args):
         myDB = db.DBConnection()
         artistsToAdd = []
@@ -657,28 +671,29 @@ class WebInterface(object):
             logger.debug("Refreshing artists: %s" % artistsToAdd)
             threading.Thread(target=importer.addArtistIDListToDB, args=[artistsToAdd]).start()
         raise cherrypy.HTTPRedirect("home")
-    markArtists.exposed = True
 
+    @cherrypy.expose
     def importLastFM(self, username):
         headphones.CONFIG.LASTFM_USERNAME = username
         headphones.CONFIG.write()
         threading.Thread(target=lastfm.getArtists).start()
         raise cherrypy.HTTPRedirect("home")
-    importLastFM.exposed = True
 
+    @cherrypy.expose
     def importLastFMTag(self, tag, limit):
         threading.Thread(target=lastfm.getTagTopArtists, args=(tag, limit)).start()
         raise cherrypy.HTTPRedirect("home")
-    importLastFMTag.exposed = True
 
+    @cherrypy.expose
     def importItunes(self, path):
         headphones.CONFIG.PATH_TO_XML = path
         headphones.CONFIG.write()
-        threading.Thread(target=importer.itunesImport, args=[path]).start()
-        time.sleep(10)
+        thread = threading.Thread(target=importer.itunesImport, args=[path])
+        thread.start()
+        thread.join(10)
         raise cherrypy.HTTPRedirect("home")
-    importItunes.exposed = True
 
+    @cherrypy.expose
     def musicScan(self, path, scan=0, redirect=None, autoadd=0, libraryscan=0):
         headphones.CONFIG.LIBRARYSCAN = libraryscan
         headphones.CONFIG.AUTO_ADD_ARTISTS = autoadd
@@ -693,64 +708,64 @@ class WebInterface(object):
             raise cherrypy.HTTPRedirect(redirect)
         else:
             raise cherrypy.HTTPRedirect("home")
-    musicScan.exposed = True
 
+    @cherrypy.expose
     def forceUpdate(self):
         from headphones import updater
         threading.Thread(target=updater.dbUpdate, args=[False]).start()
         raise cherrypy.HTTPRedirect("home")
-    forceUpdate.exposed = True
 
+    @cherrypy.expose
     def forceFullUpdate(self):
         from headphones import updater
         threading.Thread(target=updater.dbUpdate, args=[True]).start()
         raise cherrypy.HTTPRedirect("home")
-    forceFullUpdate.exposed = True
 
+    @cherrypy.expose
     def forceSearch(self):
         from headphones import searcher
         threading.Thread(target=searcher.searchforalbum).start()
         raise cherrypy.HTTPRedirect("home")
-    forceSearch.exposed = True
 
+    @cherrypy.expose
     def forcePostProcess(self, dir=None, album_dir=None):
         from headphones import postprocessor
         threading.Thread(target=postprocessor.forcePostProcess, kwargs={'dir': dir, 'album_dir': album_dir}).start()
         raise cherrypy.HTTPRedirect("home")
-    forcePostProcess.exposed = True
 
+    @cherrypy.expose
     def checkGithub(self):
         from headphones import versioncheck
         versioncheck.checkGithub()
         raise cherrypy.HTTPRedirect("home")
-    checkGithub.exposed = True
 
+    @cherrypy.expose
     def history(self):
         myDB = db.DBConnection()
         history = myDB.select('''SELECT * from snatched WHERE Status NOT LIKE "Seed%" order by DateAdded DESC''')
         return serve_template(templatename="history.html", title="History", history=history)
-    history.exposed = True
 
+    @cherrypy.expose
     def logs(self):
         return serve_template(templatename="logs.html", title="Log", lineList=headphones.LOG_LIST)
-    logs.exposed = True
 
+    @cherrypy.expose
     def clearLogs(self):
         headphones.LOG_LIST = []
         logger.info("Web logs cleared")
         raise cherrypy.HTTPRedirect("logs")
-    clearLogs.exposed = True
 
+    @cherrypy.expose
     def toggleVerbose(self):
         headphones.VERBOSE = not headphones.VERBOSE
-        logger.initLogger(not headphones.QUIET, headphones.VERBOSE)
+        logger.initLogger(console=not headphones.QUIET,
+            log_dir=headphones.CONFIG.LOG_DIR, verbose=headphones.VERBOSE)
         logger.info("Verbose toggled, set to %s", headphones.VERBOSE)
         logger.debug("If you read this message, debug logging is available")
         raise cherrypy.HTTPRedirect("logs")
-    toggleVerbose.exposed = True
 
+    @cherrypy.expose
     def getLog(self, iDisplayStart=0, iDisplayLength=100, iSortCol_0=0, sSortDir_0="desc", sSearch="", **kwargs):
-
         iDisplayStart = int(iDisplayStart)
         iDisplayLength = int(iDisplayLength)
 
@@ -775,8 +790,8 @@ class WebInterface(object):
             'iTotalRecords': len(headphones.LOG_LIST),
             'aaData': rows,
         })
-    getLog.exposed = True
 
+    @cherrypy.expose
     def getArtists_json(self, iDisplayStart=0, iDisplayLength=100, sSearch="", iSortCol_0='0', sSortDir_0='asc', **kwargs):
         iDisplayStart = int(iDisplayStart)
         iDisplayLength = int(iDisplayLength)
@@ -847,8 +862,8 @@ class WebInterface(object):
         s = json.dumps(dict)
         cherrypy.response.headers['Content-type'] = 'application/json'
         return s
-    getArtists_json.exposed = True
 
+    @cherrypy.expose
     def getAlbumsByArtist_json(self, artist=None):
         myDB = db.DBConnection()
         album_json = {}
@@ -861,7 +876,6 @@ class WebInterface(object):
 
         cherrypy.response.headers['Content-type'] = 'application/json'
         return json_albums
-    getAlbumsByArtist_json.exposed = True
 
     def getArtistjson(self, ArtistID, **kwargs):
         myDB = db.DBConnection()
@@ -871,8 +885,8 @@ class WebInterface(object):
                                     'Status': artist['Status']
                                  })
         return artist_json
-    getArtistjson.exposed = True
 
+    @cherrypy.expose
     def getAlbumjson(self, AlbumID, **kwargs):
         myDB = db.DBConnection()
         album = myDB.action('SELECT * from albums WHERE AlbumID=?', [AlbumID]).fetchone()
@@ -882,8 +896,8 @@ class WebInterface(object):
                                    'Status': album['Status']
         })
         return album_json
-    getAlbumjson.exposed = True
 
+    @cherrypy.expose
     def clearhistory(self, type=None, date_added=None, title=None):
         myDB = db.DBConnection()
         if type:
@@ -897,19 +911,14 @@ class WebInterface(object):
             logger.info(u"Deleting '%s' from history" % title)
             myDB.action('DELETE from snatched WHERE Status NOT LIKE "Seed%" AND Title=? AND DateAdded=?', [title, date_added])
         raise cherrypy.HTTPRedirect("history")
-    clearhistory.exposed = True
 
+    @cherrypy.expose
     def generateAPI(self):
-
-        import hashlib
-        import random
-
         apikey = hashlib.sha224(str(random.getrandbits(256))).hexdigest()[0:32]
         logger.info("New API generated")
         return apikey
 
-    generateAPI.exposed = True
-
+    @cherrypy.expose
     def forceScan(self, keepmatched=None):
         myDB = db.DBConnection()
         #########################################
@@ -928,10 +937,9 @@ class WebInterface(object):
         except Exception as e:
             logger.error('Unable to complete the scan: %s' % e)
         raise cherrypy.HTTPRedirect("home")
-    forceScan.exposed = True
 
+    @cherrypy.expose
     def config(self):
-
         interface_dir = os.path.join(headphones.PROG_DIR, 'data/interfaces/')
         interface_list = [name for name in os.listdir(interface_dir) if os.path.isdir(os.path.join(interface_dir, name))]
 
@@ -1002,6 +1010,9 @@ class WebInterface(object):
             "use_piratebay": checked(headphones.CONFIG.PIRATEBAY),
             "piratebay_proxy_url": headphones.CONFIG.PIRATEBAY_PROXY_URL,
             "piratebay_ratio": headphones.CONFIG.PIRATEBAY_RATIO,
+            "use_oldpiratebay": checked(headphones.CONFIG.OLDPIRATEBAY),
+            "oldpiratebay_url": headphones.CONFIG.OLDPIRATEBAY_URL,
+            "oldpiratebay_ratio": headphones.CONFIG.OLDPIRATEBAY_RATIO,
             "use_mininova": checked(headphones.CONFIG.MININOVA),
             "mininova_ratio": headphones.CONFIG.MININOVA_RATIO,
             "use_waffles": checked(headphones.CONFIG.WAFFLES),
@@ -1029,6 +1040,8 @@ class WebInterface(object):
             "lossless_bitrate_to": headphones.CONFIG.LOSSLESS_BITRATE_TO,
             "freeze_db": checked(headphones.CONFIG.FREEZE_DB),
             "cue_split": checked(headphones.CONFIG.CUE_SPLIT),
+            "cue_split_flac_path": headphones.CONFIG.CUE_SPLIT_FLAC_PATH,
+            "cue_split_shntool_path": headphones.CONFIG.CUE_SPLIT_SHNTOOL_PATH,
             "move_files": checked(headphones.CONFIG.MOVE_FILES),
             "rename_files": checked(headphones.CONFIG.RENAME_FILES),
             "correct_metadata": checked(headphones.CONFIG.CORRECT_METADATA),
@@ -1167,17 +1180,17 @@ class WebInterface(object):
         config["extras"] = extras_dict
 
         return serve_template(templatename="config.html", title="Settings", config=config)
-    config.exposed = True
 
+    @cherrypy.expose
     def configUpdate(self, **kwargs):
         # Handle the variable config options. Note - keys with False values aren't getting passed
 
         checked_configs = [
             "launch_browser", "enable_https", "api_enabled", "use_blackhole", "headphones_indexer", "use_newznab", "newznab_enabled",
-            "use_nzbsorg", "use_omgwtfnzbs", "use_kat", "use_piratebay", "use_mininova", "use_waffles", "use_rutracker", "use_whatcd",
-            "preferred_bitrate_allow_lossless", "detect_bitrate", "freeze_db", "cue_split", "move_files", "rename_files", "correct_metadata",
-            "cleanup_files", "keep_nfo", "add_album_art", "embed_album_art", "embed_lyrics", "replace_existing_folders", "file_underscores",
-            "include_extras", "autowant_upcoming", "autowant_all", "autowant_manually_added", "keep_torrent_files", "music_encoder",
+            "use_nzbsorg", "use_omgwtfnzbs", "use_kat", "use_piratebay", "use_oldpiratebay", "use_mininova", "use_waffles", "use_rutracker",
+            "use_whatcd", "preferred_bitrate_allow_lossless", "detect_bitrate", "freeze_db", "cue_split", "move_files", "rename_files",
+            "correct_metadata", "cleanup_files", "keep_nfo", "add_album_art", "embed_album_art", "embed_lyrics", "replace_existing_folders",
+            "file_underscores", "include_extras", "autowant_upcoming", "autowant_all", "autowant_manually_added", "keep_torrent_files", "music_encoder",
             "encoderlossless", "encoder_multicore", "delete_lossless_files", "growl_enabled", "growl_onsnatch", "prowl_enabled",
             "prowl_onsnatch", "xbmc_enabled", "xbmc_update", "xbmc_notify", "lms_enabled", "plex_enabled", "plex_update", "plex_notify",
             "nma_enabled", "nma_onsnatch", "pushalot_enabled", "pushalot_onsnatch", "synoindex_enabled", "pushover_enabled",
@@ -1241,73 +1254,69 @@ class WebInterface(object):
             headphones.CONFIG.add_extra_newznab(extra_newznab)
 
         # Sanity checking
-        if headphones.CONFIG.SEARCH_INTERVAL < 360:
+        if headphones.CONFIG.SEARCH_INTERVAL and headphones.CONFIG.SEARCH_INTERVAL < 360:
             logger.info("Search interval too low. Resetting to 6 hour minimum")
             headphones.CONFIG.SEARCH_INTERVAL = 360
 
         # Write the config
         headphones.CONFIG.write()
 
-        #reconfigure musicbrainz database connection with the new values
+        # Reconfigure scheduler
+        headphones.initialize_scheduler()
+
+        # Reconfigure musicbrainz database connection with the new values
         mb.startmb()
 
         raise cherrypy.HTTPRedirect("config")
 
-    configUpdate.exposed = True
-
+    @cherrypy.expose
     def do_state_change(self, signal, title, timer):
         headphones.SIGNAL = signal
         message = title + '...'
         return serve_template(templatename="shutdown.html", title=title,
                               message=message, timer=timer)
 
+    @cherrypy.expose
     def shutdown(self):
         return self.do_state_change('shutdown', 'Shutting Down', 15)
-    shutdown.exposed = True
 
+    @cherrypy.expose
     def restart(self):
         return self.do_state_change('restart', 'Restarting', 30)
-    restart.exposed = True
 
+    @cherrypy.expose
     def update(self):
         return self.do_state_change('update', 'Updating', 120)
-    update.exposed = True
 
+    @cherrypy.expose
     def extras(self):
         myDB = db.DBConnection()
         cloudlist = myDB.select('SELECT * from lastfmcloud')
         return serve_template(templatename="extras.html", title="Extras", cloudlist=cloudlist)
-    extras.exposed = True
 
+    @cherrypy.expose
     def addReleaseById(self, rid, rgid=None):
         threading.Thread(target=importer.addReleaseById, args=[rid, rgid]).start()
         if rgid:
             raise cherrypy.HTTPRedirect("albumPage?AlbumID=%s" % rgid)
         else:
             raise cherrypy.HTTPRedirect("home")
-    addReleaseById.exposed = True
 
+    @cherrypy.expose
     def updateCloud(self):
-
         lastfm.getSimilar()
         raise cherrypy.HTTPRedirect("extras")
 
-    updateCloud.exposed = True
-
+    @cherrypy.expose
     def api(self, *args, **kwargs):
-
         from headphones.api import Api
 
         a = Api()
-
         a.checkParams(*args, **kwargs)
 
-        data = a.fetchData()
-
-        return data
-
-    api.exposed = True
+        return a.fetchData()
 
+    @cherrypy.expose
     def getInfo(self, ArtistID=None, AlbumID=None):
 
         from headphones import cache
@@ -1315,25 +1324,22 @@ class WebInterface(object):
 
         return json.dumps(info_dict)
 
-    getInfo.exposed = True
-
+    @cherrypy.expose
     def getArtwork(self, ArtistID=None, AlbumID=None):
 
         from headphones import cache
         return cache.getArtwork(ArtistID, AlbumID)
 
-    getArtwork.exposed = True
-
+    @cherrypy.expose
     def getThumb(self, ArtistID=None, AlbumID=None):
 
         from headphones import cache
         return cache.getThumb(ArtistID, AlbumID)
 
-    getThumb.exposed = True
-
-    # If you just want to get the last.fm image links for an album, make sure to pass a releaseid and not a releasegroupid
+    # If you just want to get the last.fm image links for an album, make sure
+    # to pass a releaseid and not a releasegroupid
+    @cherrypy.expose
     def getImageLinks(self, ArtistID=None, AlbumID=None):
-
         from headphones import cache
         image_dict = cache.getImageLinks(ArtistID, AlbumID)
 
@@ -1350,14 +1356,13 @@ class WebInterface(object):
 
         return json.dumps(image_dict)
 
-    getImageLinks.exposed = True
-
+    @cherrypy.expose
     def twitterStep1(self):
         cherrypy.response.headers['Cache-Control'] = "max-age=0,no-cache,no-store"
         tweet = notifiers.TwitterNotifier()
         return tweet._get_authorization()
-    twitterStep1.exposed = True
 
+    @cherrypy.expose
     def twitterStep2(self, key):
         cherrypy.response.headers['Cache-Control'] = "max-age=0,no-cache,no-store"
         tweet = notifiers.TwitterNotifier()
@@ -1367,8 +1372,8 @@ class WebInterface(object):
             return "Key verification successful"
         else:
             return "Unable to verify key"
-    twitterStep2.exposed = True
 
+    @cherrypy.expose
     def testTwitter(self):
         cherrypy.response.headers['Cache-Control'] = "max-age=0,no-cache,no-store"
         tweet = notifiers.TwitterNotifier()
@@ -1377,8 +1382,8 @@ class WebInterface(object):
             return "Tweet successful, check your twitter to make sure it worked"
         else:
             return "Error sending tweet"
-    testTwitter.exposed = True
 
+    @cherrypy.expose
     def osxnotifyregister(self, app):
         cherrypy.response.headers['Cache-Control'] = "max-age=0,no-cache,no-store"
         from osxnotify import registerapp as osxnotify
@@ -1390,14 +1395,13 @@ class WebInterface(object):
         else:
             logger.warn(msg)
         return msg
-    osxnotifyregister.exposed = True
-
 
 class Artwork(object):
+    @cherrypy.expose
     def index(self):
         return "Artwork"
-    index.exposed = True
 
+    @cherrypy.expose
     def default(self, ArtistOrAlbum="", ID=None):
         from headphones import cache
         ArtistID = None
@@ -1422,16 +1426,15 @@ class Artwork(object):
             cherrypy.response.headers['Content-type'] = 'image/' + fileext
             cherrypy.response.headers['Cache-Control'] = 'max-age=31556926'
 
-        path = os.path.normpath(path)
-        f = open(path, 'rb')
-        return f.read()
-    default.exposed = True
+        with open(os.path.normpath(path), "rb") as fp:
+            return fp.read()
 
     class Thumbs(object):
+        @cherrypy.expose
         def index(self):
             return "Here be thumbs"
-        index.exposed = True
 
+        @cherrypy.expose
         def default(self, ArtistOrAlbum="", ID=None):
             from headphones import cache
             ArtistID = None
@@ -1456,12 +1459,8 @@ class Artwork(object):
                 cherrypy.response.headers['Content-type'] = 'image/' + fileext
                 cherrypy.response.headers['Cache-Control'] = 'max-age=31556926'
 
-            path = os.path.normpath(path)
-            f = open(path, 'rb')
-            return f.read()
-        default.exposed = True
+            with open(os.path.normpath(path), "rb") as fp:
+                return fp.read()
 
     thumbs = Thumbs()
-
-
 WebInterface.artwork = Artwork()

headphones/webstart.py

@@ -23,9 +23,7 @@ from headphones.webserve import WebInterface
 from headphones.helpers import create_https_certificates
 
 
-def initialize(options=None):
-    if options is None:
-        options = {}
+def initialize(options):
 
     # HTTPS stuff stolen from sickbeard
     enable_https = options['enable_https']
@@ -33,16 +31,17 @@ def initialize(options=None):
     https_key = options['https_key']
 
     if enable_https:
-        # If either the HTTPS certificate or key do not exist, make some self-signed ones.
+        # If either the HTTPS certificate or key do not exist, try to make
+        # self-signed ones.
         if not (https_cert and os.path.exists(https_cert)) or not (https_key and os.path.exists(https_key)):
             if not create_https_certificates(https_cert, https_key):
-                logger.warn(u"Unable to create cert/key files, disabling HTTPS")
-                headphones.CONFIG.ENABLE_HTTPS = False
+                logger.warn("Unable to create certificate and key. Disabling " \
+                    "HTTPS")
                 enable_https = False
 
         if not (os.path.exists(https_cert) and os.path.exists(https_key)):
-            logger.warn(u"Disabled HTTPS because of missing CERT and KEY files")
-            headphones.CONFIG.ENABLE_HTTPS = False
+            logger.warn("Disabled HTTPS because of missing certificate and " \
+                "key.")
             enable_https = False
 
     options_dict = {

lib/beets/LICENSE

@@ -0,0 +1,21 @@
+The MIT License
+
+Copyright (c) 2010-2014 Adrian Sampson
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.

lib/beets/README.rst

@@ -0,0 +1,94 @@
+.. image:: https://travis-ci.org/sampsyo/beets.svg?branch=master
+    :target: https://travis-ci.org/sampsyo/beets
+
+.. image:: http://img.shields.io/coveralls/sampsyo/beets.svg
+    :target: https://coveralls.io/r/sampsyo/beets
+
+.. image:: http://img.shields.io/pypi/v/beets.svg
+    :target: https://pypi.python.org/pypi/beets
+
+Beets is the media library management system for obsessive-compulsive music
+geeks.
+
+The purpose of beets is to get your music collection right once and for all.
+It catalogs your collection, automatically improving its metadata as it goes.
+It then provides a bouquet of tools for manipulating and accessing your music.
+
+Here's an example of beets' brainy tag corrector doing its thing::
+
+  $ beet import ~/music/ladytron
+  Tagging:
+      Ladytron - Witching Hour
+  (Similarity: 98.4%)
+   * Last One Standing      -> The Last One Standing
+   * Beauty                 -> Beauty*2
+   * White Light Generation -> Whitelightgenerator
+   * All the Way            -> All the Way...
+
+Because beets is designed as a library, it can do almost anything you can
+imagine for your music collection. Via `plugins`_, beets becomes a panacea:
+
+- Fetch or calculate all the metadata you could possibly need: `album art`_,
+  `lyrics`_, `genres`_, `tempos`_, `ReplayGain`_ levels, or `acoustic
+  fingerprints`_.
+- Get metadata from `MusicBrainz`_, `Discogs`_, or `Beatport`_. Or guess
+  metadata using songs' filenames or their acoustic fingerprints.
+- `Transcode audio`_ to any format you like.
+- Check your library for `duplicate tracks and albums`_ or for `albums that
+  are missing tracks`_.
+- Clean up crufty tags left behind by other, less-awesome tools.
+- Embed and extract album art from files' metadata.
+- Browse your music library graphically through a Web browser and play it in any
+  browser that supports `HTML5 Audio`_.
+- Analyze music files' metadata from the command line.
+- Listen to your library with a music player that speaks the `MPD`_ protocol
+  and works with a staggering variety of interfaces.
+
+If beets doesn't do what you want yet, `writing your own plugin`_ is
+shockingly simple if you know a little Python.
+
+.. _plugins: http://beets.readthedocs.org/page/plugins/
+.. _MPD: http://www.musicpd.org/
+.. _MusicBrainz music collection: http://musicbrainz.org/doc/Collections/
+.. _writing your own plugin:
+    http://beets.readthedocs.org/page/dev/plugins.html
+.. _HTML5 Audio:
+    http://www.w3.org/TR/html-markup/audio.html
+.. _albums that are missing tracks:
+    http://beets.readthedocs.org/page/plugins/missing.html
+.. _duplicate tracks and albums:
+    http://beets.readthedocs.org/page/plugins/duplicates.html
+.. _Transcode audio:
+    http://beets.readthedocs.org/page/plugins/convert.html
+.. _Beatport: http://www.beatport.com/
+.. _Discogs: http://www.discogs.com/
+.. _acoustic fingerprints:
+    http://beets.readthedocs.org/page/plugins/chroma.html
+.. _ReplayGain: http://beets.readthedocs.org/page/plugins/replaygain.html
+.. _tempos: http://beets.readthedocs.org/page/plugins/echonest.html
+.. _genres: http://beets.readthedocs.org/page/plugins/lastgenre.html
+.. _album art: http://beets.readthedocs.org/page/plugins/fetchart.html
+.. _lyrics: http://beets.readthedocs.org/page/plugins/lyrics.html
+.. _MusicBrainz: http://musicbrainz.org/
+
+Read More
+---------
+
+Learn more about beets at `its Web site`_. Follow `@b33ts`_ on Twitter for
+news and updates.
+
+You can install beets by typing ``pip install beets``. Then check out the
+`Getting Started`_ guide.
+
+.. _its Web site: http://beets.radbox.org/
+.. _Getting Started: http://beets.readthedocs.org/page/guides/main.html
+.. _@b33ts: http://twitter.com/b33ts/
+
+Authors
+-------
+
+Beets is by `Adrian Sampson`_ with a supporting cast of thousands. For help,
+please contact the `mailing list`_.
+
+.. _mailing list: https://groups.google.com/forum/#!forum/beets-users
+.. _Adrian Sampson: http://homes.cs.washington.edu/~asampson/

lib/beets/__init__.py

@@ -1,5 +1,5 @@
 # This file is part of beets.
-# Copyright 2013, Adrian Sampson.
+# Copyright 2014, Adrian Sampson.
 #
 # Permission is hereby granted, free of charge, to any person obtaining
 # a copy of this software and associated documentation files (the
@@ -12,13 +12,14 @@
 # The above copyright notice and this permission notice shall be
 # included in all copies or substantial portions of the Software.
 
-# This particular version has been slightly modified to work with headphones
+# This particular version has been slightly modified to work with Headphones
 # https://github.com/rembo10/headphones
-import os
 
-__version__ = '1.3.4'
+__version__ = '1.3.10-headphones'
 __author__ = 'Adrian Sampson <adrian@radbox.org>'
 
+import os
+
 import beets.library
 from beets.util import confit
 

lib/beets/autotag/__init__.py

@@ -14,135 +14,21 @@
 
 """Facilities for automatically determining files' correct metadata.
 """
-import os
 import logging
-import re
 
-from beets import library, mediafile, config
-from beets.util import sorted_walk, ancestry, displayable_path
+from beets import config
 
 # Parts of external interface.
-from .hooks import AlbumInfo, TrackInfo, AlbumMatch, TrackMatch
-from .match import tag_item, tag_album
-from .match import recommendation
+from .hooks import AlbumInfo, TrackInfo, AlbumMatch, TrackMatch  # noqa
+from .match import tag_item, tag_album  # noqa
+from .match import Recommendation  # noqa
 
 # Global logger.
 log = logging.getLogger('beets')
 
-# Constants for directory walker.
-MULTIDISC_MARKERS = (r'dis[ck]', r'cd')
-MULTIDISC_PAT_FMT = r'^(.*%s[\W_]*)\d'
-
 
 # Additional utilities for the main interface.
 
-def albums_in_dir(path):
-    """Recursively searches the given directory and returns an iterable
-    of (paths, items) where paths is a list of directories and items is
-    a list of Items that is probably an album. Specifically, any folder
-    containing any media files is an album.
-    """
-    collapse_pat = collapse_paths = collapse_items = None
-
-    for root, dirs, files in sorted_walk(path,
-                                         ignore=config['ignore'].as_str_seq(),
-                                         logger=log):
-        # Get a list of items in the directory.
-        items = []
-        for filename in files:
-            try:
-                i = library.Item.from_path(os.path.join(root, filename))
-            except mediafile.FileTypeError:
-                pass
-            except mediafile.UnreadableFileError:
-                log.warn(u'unreadable file: {0}'.format(
-                    displayable_path(filename))
-                )
-            else:
-                items.append(i)
-
-        # If we're currently collapsing the constituent directories in a
-        # multi-disc album, check whether we should continue collapsing
-        # and add the current directory. If so, just add the directory
-        # and move on to the next directory. If not, stop collapsing.
-        if collapse_paths:
-            if (not collapse_pat and collapse_paths[0] in ancestry(root)) or \
-                    (collapse_pat and
-                     collapse_pat.match(os.path.basename(root))):
-                # Still collapsing.
-                collapse_paths.append(root)
-                collapse_items += items
-                continue
-            else:
-                # Collapse finished. Yield the collapsed directory and
-                # proceed to process the current one.
-                if collapse_items:
-                    yield collapse_paths, collapse_items
-                collapse_pat = collapse_paths = collapse_items = None
-
-        # Check whether this directory looks like the *first* directory
-        # in a multi-disc sequence. There are two indicators: the file
-        # is named like part of a multi-disc sequence (e.g., "Title Disc
-        # 1") or it contains no items but only directories that are
-        # named in this way.
-        start_collapsing = False
-        for marker in MULTIDISC_MARKERS:
-            marker_pat = re.compile(MULTIDISC_PAT_FMT % marker, re.I)
-            match = marker_pat.match(os.path.basename(root))
-
-            # Is this directory the root of a nested multi-disc album?
-            if dirs and not items:
-                # Check whether all subdirectories have the same prefix.
-                start_collapsing = True
-                subdir_pat = None
-                for subdir in dirs:
-                    # The first directory dictates the pattern for
-                    # the remaining directories.
-                    if not subdir_pat:
-                        match = marker_pat.match(subdir)
-                        if match:
-                            subdir_pat = re.compile(r'^%s\d' %
-                                re.escape(match.group(1)), re.I)
-                        else:
-                            start_collapsing = False
-                            break
-
-                    # Subsequent directories must match the pattern.
-                    elif not subdir_pat.match(subdir):
-                        start_collapsing = False
-                        break
-
-                # If all subdirectories match, don't check other
-                # markers.
-                if start_collapsing:
-                    break
-
-            # Is this directory the first in a flattened multi-disc album?
-            elif match:
-                start_collapsing = True
-                # Set the current pattern to match directories with the same
-                # prefix as this one, followed by a digit.
-                collapse_pat = re.compile(r'^%s\d' %
-                    re.escape(match.group(1)), re.I)
-                break
-
-        # If either of the above heuristics indicated that this is the
-        # beginning of a multi-disc album, initialize the collapsed
-        # directory and item lists and check the next directory.
-        if start_collapsing:
-            # Start collapsing; continue to the next iteration.
-            collapse_paths = [root]
-            collapse_items = items
-            continue
-
-        # If it's nonempty, yield it.
-        if items:
-            yield [root], items
-
-    # Clear out any unfinished collapse.
-    if collapse_paths and collapse_items:
-        yield collapse_paths, collapse_items
-
 def apply_item_metadata(item, track_info):
     """Set an item's metadata from its matched TrackInfo object.
     """
@@ -156,6 +42,7 @@ def apply_item_metadata(item, track_info):
     # At the moment, the other metadata is left intact (including album
     # and track number). Perhaps these should be emptied?
 
+
 def apply_metadata(album_info, mapping):
     """Set the items' metadata to match an AlbumInfo object using a
     mapping from Items to TrackInfo objects.
@@ -171,8 +58,8 @@ def apply_metadata(album_info, mapping):
 
         # Artist sort and credit names.
         item.artist_sort = track_info.artist_sort or album_info.artist_sort
-        item.artist_credit = track_info.artist_credit or \
-                             album_info.artist_credit
+        item.artist_credit = (track_info.artist_credit or
+                              album_info.artist_credit)
         item.albumartist_sort = album_info.artist_sort
         item.albumartist_credit = album_info.artist_credit
 
@@ -235,7 +122,6 @@ def apply_metadata(album_info, mapping):
                       'language',
                       'country',
                       'albumstatus',
-                      'media',
                       'albumdisambig'):
             value = getattr(album_info, field)
             if value is not None:
@@ -243,5 +129,8 @@ def apply_metadata(album_info, mapping):
         if track_info.disctitle is not None:
             item.disctitle = track_info.disctitle
 
+        if track_info.media is not None:
+            item.media = track_info.media
+
         # Headphones seal of approval
-        item.comments = 'tagged by headphones/beets'
+        item.comments = 'tagged by headphones/beets'

lib/beets/autotag/hooks.py

@@ -116,6 +116,7 @@ class AlbumInfo(object):
             for track in self.tracks:
                 track.decode(codec)
 
+
 class TrackInfo(object):
     """Describes a canonical track present on a release. Appears as part
     of an AlbumInfo's ``tracks`` list. Consists of these data members:
@@ -126,6 +127,7 @@ class TrackInfo(object):
     - ``artist_id``
     - ``length``: float: duration of the track in seconds
     - ``index``: position on the entire release
+    - ``media``: delivery mechanism (Vinyl, etc.)
     - ``medium``: the disc number this track appears on in the album
     - ``medium_index``: the track's position on the disc
     - ``medium_total``: the number of tracks on the item's disc
@@ -140,13 +142,15 @@ class TrackInfo(object):
     def __init__(self, title, track_id, artist=None, artist_id=None,
                  length=None, index=None, medium=None, medium_index=None,
                  medium_total=None, artist_sort=None, disctitle=None,
-                 artist_credit=None, data_source=None, data_url=None):
+                 artist_credit=None, data_source=None, data_url=None,
+                 media=None):
         self.title = title
         self.track_id = track_id
         self.artist = artist
         self.artist_id = artist_id
         self.length = length
         self.index = index
+        self.media = media
         self.medium = medium
         self.medium_index = medium_index
         self.medium_total = medium_total
@@ -162,7 +166,7 @@ class TrackInfo(object):
         to Unicode.
         """
         for fld in ['title', 'artist', 'medium', 'artist_sort', 'disctitle',
-                    'artist_credit']:
+                    'artist_credit', 'media']:
             value = getattr(self, fld)
             if isinstance(value, str):
                 setattr(self, fld, value.decode(codec, 'ignore'))
@@ -187,6 +191,7 @@ SD_REPLACE = [
     (r'&', 'and'),
 ]
 
+
 def _string_dist_basic(str1, str2):
     """Basic edit distance between two strings, ignoring
     non-alphanumeric characters and case. Comparisons are based on a
@@ -201,13 +206,16 @@ def _string_dist_basic(str1, str2):
         return 0.0
     return levenshtein(str1, str2) / float(max(len(str1), len(str2)))
 
+
 def string_dist(str1, str2):
     """Gives an "intuitive" edit distance between two strings. This is
     an edit distance, normalized by the string length, with a number of
     tweaks that reflect intuition about text.
     """
-    if str1 is None and str2 is None: return 0.0
-    if str1 is None or str2 is None:  return 1.0
+    if str1 is None and str2 is None:
+        return 0.0
+    if str1 is None or str2 is None:
+        return 1.0
 
     str1 = str1.lower()
     str2 = str2.lower()
@@ -217,9 +225,9 @@ def string_dist(str1, str2):
     # "something, the".
     for word in SD_END_WORDS:
         if str1.endswith(', %s' % word):
-            str1 = '%s %s' % (word, str1[:-len(word)-2])
+            str1 = '%s %s' % (word, str1[:-len(word) - 2])
         if str2.endswith(', %s' % word):
-            str2 = '%s %s' % (word, str2[:-len(word)-2])
+            str2 = '%s %s' % (word, str2[:-len(word) - 2])
 
     # Perform a couple of basic normalizing substitutions.
     for pat, repl in SD_REPLACE:
@@ -256,6 +264,23 @@ def string_dist(str1, str2):
 
     return base_dist + penalty
 
+
+class LazyClassProperty(object):
+    """A decorator implementing a read-only property that is *lazy* in
+    the sense that the getter is only invoked once. Subsequent accesses
+    through *any* instance use the cached result.
+    """
+    def __init__(self, getter):
+        self.getter = getter
+        self.computed = False
+
+    def __get__(self, obj, owner):
+        if not self.computed:
+            self.value = self.getter(owner)
+            self.computed = True
+        return self.value
+
+
 class Distance(object):
     """Keeps track of multiple distance penalties. Provides a single
     weighted distance for all penalties as well as a weighted distance
@@ -264,11 +289,15 @@ class Distance(object):
     def __init__(self):
         self._penalties = {}
 
+    @LazyClassProperty
+    def _weights(cls):
+        """A dictionary from keys to floating-point weights.
+        """
         weights_view = config['match']['distance_weights']
-        self._weights = {}
+        weights = {}
         for key in weights_view.keys():
-            self._weights[key] = weights_view[key].as_number()
-
+            weights[key] = weights_view[key].as_number()
+        return weights
 
     # Access the components and their aggregates.
 
@@ -313,8 +342,7 @@ class Distance(object):
         # Convert distance into a negative float we can sort items in
         # ascending order (for keys, when the penalty is equal) and
         # still get the items with the biggest distance first.
-        return sorted(list_, key=lambda (key, dist): (0-dist, key))
-
+        return sorted(list_, key=lambda (key, dist): (0 - dist, key))
 
     # Behave like a float.
 
@@ -323,13 +351,13 @@ class Distance(object):
 
     def __float__(self):
         return self.distance
+
     def __sub__(self, other):
         return self.distance - other
 
     def __rsub__(self, other):
         return other - self.distance
 
-
     # Behave like a dict.
 
     def __getitem__(self, key):
@@ -355,11 +383,11 @@ class Distance(object):
         """
         if not isinstance(dist, Distance):
             raise ValueError(
-                    '`dist` must be a Distance object. It is: %r' % dist)
+                '`dist` must be a Distance object, not {0}'.format(type(dist))
+            )
         for key, penalties in dist._penalties.iteritems():
             self._penalties.setdefault(key, []).extend(penalties)
 
-
     # Adding components.
 
     def _eq(self, value1, value2):
@@ -379,7 +407,8 @@ class Distance(object):
         """
         if not 0.0 <= dist <= 1.0:
             raise ValueError(
-                    '`dist` must be between 0.0 and 1.0. It is: %r' % dist)
+                '`dist` must be between 0.0 and 1.0, not {0}'.format(dist)
+            )
         self._penalties.setdefault(key, []).append(dist)
 
     def add_equality(self, key, value, options):
@@ -476,6 +505,7 @@ def album_for_mbid(release_id):
     except mb.MusicBrainzAPIError as exc:
         exc.log(log)
 
+
 def track_for_mbid(recording_id):
     """Get a TrackInfo object for a MusicBrainz recording ID. Return None
     if the ID is not found.
@@ -485,18 +515,21 @@ def track_for_mbid(recording_id):
     except mb.MusicBrainzAPIError as exc:
         exc.log(log)
 
+
 def albums_for_id(album_id):
     """Get a list of albums for an ID."""
     candidates = [album_for_mbid(album_id)]
     candidates.extend(plugins.album_for_id(album_id))
     return filter(None, candidates)
 
+
 def tracks_for_id(track_id):
     """Get a list of tracks for an ID."""
     candidates = [track_for_mbid(track_id)]
     candidates.extend(plugins.track_for_id(track_id))
     return filter(None, candidates)
 
+
 def album_candidates(items, artist, album, va_likely):
     """Search for album matches. ``items`` is a list of Item objects
     that make up the album. ``artist`` and ``album`` are the respective
@@ -525,6 +558,7 @@ def album_candidates(items, artist, album, va_likely):
 
     return out
 
+
 def item_candidates(item, artist, title):
     """Search for item matches. ``item`` is the Item to be matched.
     ``artist`` and ``title`` are strings and either reflect the item or

lib/beets/autotag/match.py

@@ -25,11 +25,8 @@ from munkres import Munkres
 from beets import plugins
 from beets import config
 from beets.util import plurality
-from beets.util.enumeration import enum
 from beets.autotag import hooks
-
-# Recommendation enumeration.
-recommendation = enum('none', 'low', 'medium', 'strong', name='recommendation')
+from beets.util.enumeration import OrderedEnum
 
 # Artist signals that indicate "various artists". These are used at the
 # album level to determine whether a given release is likely a VA
@@ -41,6 +38,18 @@ VA_ARTISTS = (u'', u'various artists', u'various', u'va', u'unknown')
 log = logging.getLogger('beets')
 
 
+# Recommendation enumeration.
+
+class Recommendation(OrderedEnum):
+    """Indicates a qualitative suggestion to the user about what should
+    be done with a given match.
+    """
+    none = 0
+    low = 1
+    medium = 2
+    strong = 3
+
+
 # Primary matching functionality.
 
 def current_metadata(items):
@@ -56,10 +65,10 @@ def current_metadata(items):
     fields = ['artist', 'album', 'albumartist', 'year', 'disctotal',
               'mb_albumid', 'label', 'catalognum', 'country', 'media',
               'albumdisambig']
-    for key in fields:
-        values = [getattr(item, key) for item in items if item]
-        likelies[key], freq = plurality(values)
-        consensus[key] = (freq == len(values))
+    for field in fields:
+        values = [item[field] for item in items if item]
+        likelies[field], freq = plurality(values)
+        consensus[field] = (freq == len(values))
 
     # If there's an album artist consensus, use this for the artist.
     if consensus['albumartist'] and likelies['albumartist']:
@@ -67,6 +76,7 @@ def current_metadata(items):
 
     return likelies, consensus
 
+
 def assign_items(items, tracks):
     """Given a list of Items and a list of TrackInfo objects, find the
     best mapping between them. Returns a mapping from Items to TrackInfo
@@ -93,12 +103,14 @@ def assign_items(items, tracks):
     extra_tracks.sort(key=lambda t: (t.index, t.title))
     return mapping, extra_items, extra_tracks
 
+
 def track_index_changed(item, track_info):
     """Returns True if the item and track info index is different. Tolerates
     per disc and per release numbering.
     """
     return item.track not in (track_info.medium_index, track_info.index)
 
+
 def track_distance(item, track_info, incl_artist=False):
     """Determines the significance of a track metadata change. Returns a
     Distance object. `incl_artist` indicates that a distance component should
@@ -109,7 +121,7 @@ def track_distance(item, track_info, incl_artist=False):
     # Length.
     if track_info.length:
         diff = abs(item.length - track_info.length) - \
-               config['match']['track_length_grace'].as_number()
+            config['match']['track_length_grace'].as_number()
         dist.add_ratio('track_length', diff,
                        config['match']['track_length_max'].as_number())
 
@@ -134,6 +146,7 @@ def track_distance(item, track_info, incl_artist=False):
 
     return dist
 
+
 def distance(items, album_info, mapping):
     """Determines how "significant" an album metadata change would be.
     Returns a Distance object. `album_info` is an AlbumInfo object
@@ -239,6 +252,7 @@ def distance(items, album_info, mapping):
 
     return dist
 
+
 def match_by_id(items):
     """If the items are tagged with a MusicBrainz album ID, returns an
     AlbumInfo object for the corresponding album. Otherwise, returns
@@ -247,16 +261,17 @@ def match_by_id(items):
     # Is there a consensus on the MB album ID?
     albumids = [item.mb_albumid for item in items if item.mb_albumid]
     if not albumids:
-        log.debug('No album IDs found.')
+        log.debug(u'No album IDs found.')
         return None
 
     # If all album IDs are equal, look up the album.
-    if bool(reduce(lambda x,y: x if x==y else (), albumids)):
+    if bool(reduce(lambda x, y: x if x == y else (), albumids)):
         albumid = albumids[0]
-        log.debug('Searching for discovered album ID: ' + albumid)
+        log.debug(u'Searching for discovered album ID: {0}'.format(albumid))
         return hooks.album_for_mbid(albumid)
     else:
-        log.debug('No album ID consensus.')
+        log.debug(u'No album ID consensus.')
+
 
 def _recommendation(results):
     """Given a sorted list of AlbumMatch or TrackMatch objects, return a
@@ -268,26 +283,26 @@ def _recommendation(results):
     """
     if not results:
         # No candidates: no recommendation.
-        return recommendation.none
+        return Recommendation.none
 
     # Basic distance thresholding.
     min_dist = results[0].distance
     if min_dist < config['match']['strong_rec_thresh'].as_number():
         # Strong recommendation level.
-        rec = recommendation.strong
+        rec = Recommendation.strong
     elif min_dist <= config['match']['medium_rec_thresh'].as_number():
         # Medium recommendation level.
-        rec = recommendation.medium
+        rec = Recommendation.medium
     elif len(results) == 1:
         # Only a single candidate.
-        rec = recommendation.low
+        rec = Recommendation.low
     elif results[1].distance - min_dist >= \
             config['match']['rec_gap_thresh'].as_number():
         # Gap between first two candidates is large.
-        rec = recommendation.low
+        rec = Recommendation.low
     else:
         # No conclusion. Return immediately. Can't be downgraded any further.
-        return recommendation.none
+        return Recommendation.none
 
     # Downgrade to the max rec if it is lower than the current rec for an
     # applied penalty.
@@ -299,28 +314,40 @@ def _recommendation(results):
     for key in keys:
         if key in max_rec_view.keys():
             max_rec = max_rec_view[key].as_choice({
-                'strong': recommendation.strong,
-                'medium': recommendation.medium,
-                'low': recommendation.low,
-                'none': recommendation.none,
+                'strong': Recommendation.strong,
+                'medium': Recommendation.medium,
+                'low': Recommendation.low,
+                'none': Recommendation.none,
             })
             rec = min(rec, max_rec)
 
     return rec
 
+
 def _add_candidate(items, results, info):
     """Given a candidate AlbumInfo object, attempt to add the candidate
     to the output dictionary of AlbumMatch objects. This involves
     checking the track count, ordering the items, checking for
     duplicates, and calculating the distance.
     """
-    log.debug('Candidate: %s - %s' % (info.artist, info.album))
+    log.debug(u'Candidate: {0} - {1}'.format(info.artist, info.album))
+
+    # Discard albums with zero tracks.
+    if not info.tracks:
+        log.debug('No tracks.')
+        return
 
     # Don't duplicate.
     if info.album_id in results:
-        log.debug('Duplicate.')
+        log.debug(u'Duplicate.')
         return
 
+    # Discard matches without required tags.
+    for req_tag in config['match']['required'].as_str_seq():
+        if getattr(info, req_tag) is None:
+            log.debug(u'Ignored. Missing required tag: {0}'.format(req_tag))
+            return
+
     # Find mapping between the items and the track info.
     mapping, extra_items, extra_tracks = assign_items(items, info.tracks)
 
@@ -331,30 +358,36 @@ def _add_candidate(items, results, info):
     penalties = [key for _, key in dist]
     for penalty in config['match']['ignored'].as_str_seq():
         if penalty in penalties:
-            log.debug('Ignored. Penalty: %s' % penalty)
+            log.debug(u'Ignored. Penalty: {0}'.format(penalty))
             return
 
-    log.debug('Success. Distance: %f' % dist)
+    log.debug(u'Success. Distance: {0}'.format(dist))
     results[info.album_id] = hooks.AlbumMatch(dist, info, mapping,
                                               extra_items, extra_tracks)
 
+
 def tag_album(items, search_artist=None, search_album=None,
               search_id=None):
-    """Bundles together the functionality used to infer tags for a
-    set of items comprised by an album. Returns everything relevant:
-        - The current artist.
-        - The current album.
-        - A list of AlbumMatch objects. The candidates are sorted by
-          distance (i.e., best match first).
-        - A recommendation.
-    If search_artist and search_album or search_id are provided, then
-    they are used as search terms in place of the current metadata.
+    """Return a tuple of a artist name, an album name, a list of
+    `AlbumMatch` candidates from the metadata backend, and a
+    `Recommendation`.
+
+    The artist and album are the most common values of these fields
+    among `items`.
+
+    The `AlbumMatch` objects are generated by searching the metadata
+    backends. By default, the metadata of the items is used for the
+    search. This can be customized by setting the parameters. The
+    `mapping` field of the album has the matched `items` as keys.
+
+    The recommendation is calculated from the match qualitiy of the
+    candidates.
     """
     # Get current metadata.
     likelies, consensus = current_metadata(items)
     cur_artist = likelies['artist']
     cur_album = likelies['album']
-    log.debug('Tagging %s - %s' % (cur_artist, cur_album))
+    log.debug(u'Tagging {0} - {1}'.format(cur_artist, cur_album))
 
     # The output result (distance, AlbumInfo) tuples (keyed by MB album
     # ID).
@@ -362,7 +395,7 @@ def tag_album(items, search_artist=None, search_album=None,
 
     # Search by explicit ID.
     if search_id is not None:
-        log.debug('Searching for album ID: ' + search_id)
+        log.debug(u'Searching for album ID: {0}'.format(search_id))
         search_cands = hooks.albums_for_id(search_id)
 
     # Use existing metadata or text search.
@@ -372,32 +405,33 @@ def tag_album(items, search_artist=None, search_album=None,
         if id_info:
             _add_candidate(items, candidates, id_info)
             rec = _recommendation(candidates.values())
-            log.debug('Album ID match recommendation is ' + str(rec))
+            log.debug(u'Album ID match recommendation is {0}'.format(str(rec)))
             if candidates and not config['import']['timid']:
                 # If we have a very good MBID match, return immediately.
                 # Otherwise, this match will compete against metadata-based
                 # matches.
-                if rec == recommendation.strong:
-                    log.debug('ID match.')
+                if rec == Recommendation.strong:
+                    log.debug(u'ID match.')
                     return cur_artist, cur_album, candidates.values(), rec
 
         # Search terms.
         if not (search_artist and search_album):
             # No explicit search terms -- use current metadata.
             search_artist, search_album = cur_artist, cur_album
-        log.debug(u'Search terms: %s - %s' % (search_artist, search_album))
+        log.debug(u'Search terms: {0} - {1}'.format(search_artist,
+                                                    search_album))
 
         # Is this album likely to be a "various artist" release?
         va_likely = ((not consensus['artist']) or
-                    (search_artist.lower() in VA_ARTISTS) or
-                    any(item.comp for item in items))
-        log.debug(u'Album might be VA: %s' % str(va_likely))
+                     (search_artist.lower() in VA_ARTISTS) or
+                     any(item.comp for item in items))
+        log.debug(u'Album might be VA: {0}'.format(str(va_likely)))
 
         # Get the results from the data sources.
         search_cands = hooks.album_candidates(items, search_artist,
                                               search_album, va_likely)
 
-    log.debug(u'Evaluating %i candidates.' % len(search_cands))
+    log.debug(u'Evaluating {0} candidates.'.format(len(search_cands)))
     for info in search_cands:
         _add_candidate(items, candidates, info)
 
@@ -406,6 +440,7 @@ def tag_album(items, search_artist=None, search_album=None,
     rec = _recommendation(candidates)
     return cur_artist, cur_album, candidates, rec
 
+
 def tag_item(item, search_artist=None, search_title=None,
              search_id=None):
     """Attempts to find metadata for a single track. Returns a
@@ -421,15 +456,15 @@ def tag_item(item, search_artist=None, search_title=None,
     # First, try matching by MusicBrainz ID.
     trackid = search_id or item.mb_trackid
     if trackid:
-        log.debug('Searching for track ID: ' + trackid)
+        log.debug(u'Searching for track ID: {0}'.format(trackid))
         for track_info in hooks.tracks_for_id(trackid):
             dist = track_distance(item, track_info, incl_artist=True)
             candidates[track_info.track_id] = \
-                    hooks.TrackMatch(dist, track_info)
+                hooks.TrackMatch(dist, track_info)
             # If this is a good match, then don't keep searching.
             rec = _recommendation(candidates.values())
-            if rec == recommendation.strong and not config['import']['timid']:
-                log.debug('Track ID match.')
+            if rec == Recommendation.strong and not config['import']['timid']:
+                log.debug(u'Track ID match.')
                 return candidates.values(), rec
 
     # If we're searching by ID, don't proceed.
@@ -437,12 +472,13 @@ def tag_item(item, search_artist=None, search_title=None,
         if candidates:
             return candidates.values(), rec
         else:
-            return [], recommendation.none
+            return [], Recommendation.none
 
     # Search terms.
     if not (search_artist and search_title):
         search_artist, search_title = item.artist, item.title
-    log.debug(u'Item search terms: %s - %s' % (search_artist, search_title))
+    log.debug(u'Item search terms: {0} - {1}'.format(search_artist,
+                                                     search_title))
 
     # Get and evaluate candidate metadata.
     for track_info in hooks.item_candidates(item, search_artist, search_title):
@@ -450,7 +486,7 @@ def tag_item(item, search_artist=None, search_title=None,
         candidates[track_info.track_id] = hooks.TrackMatch(dist, track_info)
 
     # Sort by distance and return with recommendation.
-    log.debug('Found %i candidates.' % len(candidates))
+    log.debug(u'Found {0} candidates.'.format(len(candidates)))
     candidates = sorted(candidates.itervalues())
     rec = _recommendation(candidates)
     return candidates, rec

lib/beets/autotag/mb.py

@@ -32,6 +32,7 @@ BASE_URL = 'http://musicbrainz.org/'
 musicbrainzngs.set_useragent('beets', beets.__version__,
                              'http://beets.radbox.org/')
 
+
 class MusicBrainzAPIError(util.HumanReadableException):
     """An error while talking to MusicBrainz. The `query` field is the
     parameter to the action and may have any type.
@@ -41,7 +42,7 @@ class MusicBrainzAPIError(util.HumanReadableException):
         super(MusicBrainzAPIError, self).__init__(reason, verb, tb)
 
     def get_message(self):
-        return u'"{0}" in {1} with query {2}'.format(
+        return u'{0} in {1} with query {2}'.format(
             self._reasonstr(), self.verb, repr(self.query)
         )
 
@@ -51,12 +52,15 @@ RELEASE_INCLUDES = ['artists', 'media', 'recordings', 'release-groups',
                     'labels', 'artist-credits', 'aliases']
 TRACK_INCLUDES = ['artists', 'aliases']
 
+
 def track_url(trackid):
     return urljoin(BASE_URL, 'recording/' + trackid)
 
+
 def album_url(albumid):
     return urljoin(BASE_URL, 'release/' + albumid)
 
+
 def configure():
     """Set up the python-musicbrainz-ngs module according to settings
     from the beets configuration. This should be called at startup.
@@ -67,6 +71,7 @@ def configure():
         config['musicbrainz']['ratelimit'].get(int),
     )
 
+
 def _preferred_alias(aliases):
     """Given an list of alias structures for an artist credit, select
     and return the user's preferred alias alias or None if no matching
@@ -81,13 +86,15 @@ def _preferred_alias(aliases):
     # Search configured locales in order.
     for locale in config['import']['languages'].as_str_seq():
         # Find matching primary aliases for this locale.
-        matches = [a for a in aliases if a['locale'] == locale and 'primary' in a]
+        matches = [a for a in aliases
+                   if a['locale'] == locale and 'primary' in a]
         # Skip to the next locale if we have no matches
         if not matches:
             continue
 
         return matches[0]
 
+
 def _flatten_artist_credit(credit):
     """Given a list representing an ``artist-credit`` block, flatten the
     data into a triple of joined artist name strings: canonical, sort, and
@@ -133,6 +140,7 @@ def _flatten_artist_credit(credit):
         ''.join(artist_credit_parts),
     )
 
+
 def track_info(recording, index=None, medium=None, medium_index=None,
                medium_total=None):
     """Translates a MusicBrainz recording result dictionary into a beets
@@ -167,6 +175,7 @@ def track_info(recording, index=None, medium=None, medium_index=None,
     info.decode()
     return info
 
+
 def _set_date_str(info, date_str, original=False):
     """Given a (possibly partial) YYYY-MM-DD string and an AlbumInfo
     object, set the object's release date fields appropriately. If
@@ -186,6 +195,7 @@ def _set_date_str(info, date_str, original=False):
                     key = 'original_' + key
                 setattr(info, key, date_num)
 
+
 def album_info(release):
     """Takes a MusicBrainz release result dictionary and returns a beets
     AlbumInfo object containing the interesting data about that release.
@@ -199,6 +209,7 @@ def album_info(release):
     index = 0
     for medium in release['medium-list']:
         disctitle = medium.get('title')
+        format = medium.get('format')
         for track in medium['track-list']:
             # Basic information from the recording.
             index += 1
@@ -210,6 +221,7 @@ def album_info(release):
                 len(medium['track-list']),
             )
             ti.disctitle = disctitle
+            ti.media = format
 
             # Prefer track data, where present, over recording data.
             if track.get('title'):
@@ -288,6 +300,7 @@ def album_info(release):
     info.decode()
     return info
 
+
 def match_album(artist, album, tracks=None, limit=SEARCH_LIMIT):
     """Searches for a single album ("release" in MusicBrainz parlance)
     and returns an iterator over AlbumInfo objects. May raise a
@@ -297,9 +310,9 @@ def match_album(artist, album, tracks=None, limit=SEARCH_LIMIT):
     optionally, a number of tracks on the album.
     """
     # Build search criteria.
-    criteria = {'release': album.lower()}
+    criteria = {'release': album.lower().strip()}
     if artist is not None:
-        criteria['artist'] = artist.lower()
+        criteria['artist'] = artist.lower().strip()
     else:
         # Various Artists search.
         criteria['arid'] = VARIOUS_ARTISTS_ID
@@ -322,13 +335,14 @@ def match_album(artist, album, tracks=None, limit=SEARCH_LIMIT):
         if albuminfo is not None:
             yield albuminfo
 
+
 def match_track(artist, title, limit=SEARCH_LIMIT):
     """Searches for a single track and returns an iterable of TrackInfo
     objects. May raise a MusicBrainzAPIError.
     """
     criteria = {
-        'artist': artist.lower(),
-        'recording': title.lower(),
+        'artist': artist.lower().strip(),
+        'recording': title.lower().strip(),
     }
 
     if not any(criteria.itervalues()):
@@ -342,6 +356,7 @@ def match_track(artist, title, limit=SEARCH_LIMIT):
     for recording in res['recording-list']:
         yield track_info(recording)
 
+
 def _parse_id(s):
     """Search for a MusicBrainz ID in the given string and return it. If
     no ID can be found, return None.
@@ -351,38 +366,40 @@ def _parse_id(s):
     if match:
         return match.group()
 
-def album_for_id(albumid):
+
+def album_for_id(releaseid):
     """Fetches an album by its MusicBrainz ID and returns an AlbumInfo
     object or None if the album is not found. May raise a
     MusicBrainzAPIError.
     """
-    albumid = _parse_id(albumid)
+    albumid = _parse_id(releaseid)
     if not albumid:
-        log.error('Invalid MBID.')
+        log.debug(u'Invalid MBID ({0}).'.format(releaseid))
         return
     try:
         res = musicbrainzngs.get_release_by_id(albumid,
                                                RELEASE_INCLUDES)
     except musicbrainzngs.ResponseError:
-        log.debug('Album ID match failed.')
+        log.debug(u'Album ID match failed.')
         return None
     except musicbrainzngs.MusicBrainzError as exc:
         raise MusicBrainzAPIError(exc, 'get release by ID', albumid,
                                   traceback.format_exc())
     return album_info(res['release'])
 
-def track_for_id(trackid):
+
+def track_for_id(releaseid):
     """Fetches a track by its MusicBrainz ID. Returns a TrackInfo object
     or None if no track is found. May raise a MusicBrainzAPIError.
     """
-    trackid = _parse_id(trackid)
+    trackid = _parse_id(releaseid)
     if not trackid:
-        log.error('Invalid MBID.')
+        log.debug(u'Invalid MBID ({0}).'.format(releaseid))
         return
     try:
         res = musicbrainzngs.get_recording_by_id(trackid, TRACK_INCLUDES)
     except musicbrainzngs.ResponseError:
-        log.debug('Track ID match failed.')
+        log.debug(u'Track ID match failed.')
         return None
     except musicbrainzngs.MusicBrainzError as exc:
         raise MusicBrainzAPIError(exc, 'get recording by ID', trackid,

lib/beets/config_default.yaml

@@ -5,6 +5,7 @@ import:
     write: yes
     copy: yes
     move: no
+    link: no
     delete: no
     resume: ask
     incremental: no
@@ -20,6 +21,7 @@ import:
     detail: no
     flat: no
     group_albums: no
+    pretend: false
 
 clutter: ["Thumbs.DB", ".DS_Store"]
 ignore: [".*", "*~", "System Volume Information"]
@@ -32,6 +34,7 @@ replace:
     '\s+$': ''
     '^\s+': ''
 path_sep_replace: _
+asciify_paths: false
 art_filename: cover
 max_filename_length: 0
 
@@ -54,6 +57,9 @@ list_format_item: $artist - $album - $title
 list_format_album: $albumartist - $album
 time_format: '%Y-%m-%d %H:%M:%S'
 
+sort_album: albumartist+ album+
+sort_item: artist+ album+ disc+ track+
+
 paths:
     default: $albumartist/$album%aunique{}/$track $title
     singleton: Non-Album/$artist/$title
@@ -98,5 +104,6 @@ match:
         media: []
         original_year: no
     ignored: []
+    required: []
     track_length_grace: 10
     track_length_max: 30

lib/beets/dbcore/__init__.py

@@ -18,3 +18,8 @@ Library.
 from .db import Model, Database
 from .query import Query, FieldQuery, MatchQuery, AndQuery, OrQuery
 from .types import Type
+from .queryparse import query_from_strings
+from .queryparse import sort_from_strings
+from .queryparse import parse_sorted_query
+
+# flake8: noqa

lib/beets/dbcore/db.py

@@ -20,16 +20,62 @@ from collections import defaultdict
 import threading
 import sqlite3
 import contextlib
+import collections
 
 import beets
 from beets.util.functemplate import Template
-from .query import MatchQuery
+from beets.dbcore import types
+from .query import MatchQuery, NullSort, TrueQuery
 
 
+class FormattedMapping(collections.Mapping):
+    """A `dict`-like formatted view of a model.
+
+    The accessor `mapping[key]` returns the formated version of
+    `model[key]` as a unicode string.
+
+    If `for_path` is true, all path separators in the formatted values
+    are replaced.
+    """
+
+    def __init__(self, model, for_path=False):
+        self.for_path = for_path
+        self.model = model
+        self.model_keys = model.keys(True)
+
+    def __getitem__(self, key):
+        if key in self.model_keys:
+            return self._get_formatted(self.model, key)
+        else:
+            raise KeyError(key)
+
+    def __iter__(self):
+        return iter(self.model_keys)
+
+    def __len__(self):
+        return len(self.model_keys)
+
+    def get(self, key, default=None):
+        if default is None:
+            default = self.model._type(key).format(None)
+        return super(FormattedMapping, self).get(key, default)
+
+    def _get_formatted(self, model, key):
+        value = model._type(key).format(model.get(key))
+        if isinstance(value, bytes):
+            value = value.decode('utf8', 'ignore')
+
+        if self.for_path:
+            sep_repl = beets.config['path_sep_replace'].get(unicode)
+            for sep in (os.path.sep, os.path.altsep):
+                if sep:
+                    value = value.replace(sep, sep_repl)
+
+        return value
+
 
 # Abstract base for model classes.
 
-
 class Model(object):
     """An abstract object representing an object in the database. Model
     objects act like dictionaries (i.e., the allow subscript access like
@@ -66,12 +112,7 @@ class Model(object):
 
     _fields = {}
     """A mapping indicating available "fixed" fields on this type. The
-    keys are field names and the values are Type objects.
-    """
-
-    _bytes_keys = ()
-    """Keys whose values should be stored as raw bytes blobs rather than
-    strings.
+    keys are field names and the values are `Type` objects.
     """
 
     _search_fields = ()
@@ -79,6 +120,21 @@ class Model(object):
     terms.
     """
 
+    _types = {}
+    """Optional Types for non-fixed (i.e., flexible and computed) fields.
+    """
+
+    _sorts = {}
+    """Optional named sort criteria. The keys are strings and the values
+    are subclasses of `Sort`.
+    """
+
+    _always_dirty = False
+    """By default, fields only become "dirty" when their value actually
+    changes. Enabling this flag marks fields as dirty even when the new
+    value is the same as the old value (e.g., `o.f = o.f`).
+    """
+
     @classmethod
     def _getters(cls):
         """Return a mapping from field names to getter functions.
@@ -94,7 +150,6 @@ class Model(object):
         # As above: we could consider caching this result.
         raise NotImplementedError()
 
-
     # Basic operation.
 
     def __init__(self, db=None, **values):
@@ -110,6 +165,20 @@ class Model(object):
         self.update(values)
         self.clear_dirty()
 
+    @classmethod
+    def _awaken(cls, db=None, fixed_values={}, flex_values={}):
+        """Create an object with values drawn from the database.
+
+        This is a performance optimization: the checks involved with
+        ordinary construction are bypassed.
+        """
+        obj = cls(db)
+        for key, value in fixed_values.iteritems():
+            obj._values_fixed[key] = cls._type(key).from_sql(value)
+        for key, value in flex_values.iteritems():
+            obj._values_flex[key] = cls._type(key).from_sql(value)
+        return obj
+
     def __repr__(self):
         return '{0}({1})'.format(
             type(self).__name__,
@@ -132,9 +201,17 @@ class Model(object):
         if need_id and not self.id:
             raise ValueError('{0} has no id'.format(type(self).__name__))
 
-
     # Essential field accessors.
 
+    @classmethod
+    def _type(self, key):
+        """Get the type of a field, a `Type` instance.
+
+        If the field has no explicit type, it is given the base `Type`,
+        which does no conversion.
+        """
+        return self._fields.get(key) or self._types.get(key) or types.DEFAULT
+
     def __getitem__(self, key):
         """Get the value for a field. Raise a KeyError if the field is
         not available.
@@ -152,11 +229,19 @@ class Model(object):
     def __setitem__(self, key, value):
         """Assign the value for a field.
         """
-        source = self._values_fixed if key in self._fields \
-                 else self._values_flex
+        # Choose where to place the value.
+        if key in self._fields:
+            source = self._values_fixed
+        else:
+            source = self._values_flex
+
+        # If the field has a type, filter the value.
+        value = self._type(key).normalize(value)
+
+        # Assign value and possibly mark as dirty.
         old_value = source.get(key)
         source[key] = value
-        if old_value != value:
+        if self._always_dirty or old_value != value:
             self._dirty.add(key)
 
     def __delitem__(self, key):
@@ -183,7 +268,6 @@ class Model(object):
         else:
             return base_keys
 
-
     # Act like a dictionary.
 
     def update(self, values):
@@ -219,7 +303,6 @@ class Model(object):
         """
         return iter(self.keys())
 
-
     # Convenient attribute access.
 
     def __getattr__(self, key):
@@ -243,7 +326,6 @@ class Model(object):
         else:
             del self[key]
 
-
     # Database interaction (CRUD methods).
 
     def store(self):
@@ -252,19 +334,15 @@ class Model(object):
         self._check_db()
 
         # Build assignments for query.
-        assignments = ''
+        assignments = []
         subvars = []
         for key in self._fields:
             if key != 'id' and key in self._dirty:
                 self._dirty.remove(key)
-                assignments += key + '=?,'
-                value = self[key]
-                # Wrap path strings in buffers so they get stored
-                # "in the raw".
-                if key in self._bytes_keys and isinstance(value, str):
-                    value = buffer(value)
+                assignments.append(key + '=?')
+                value = self._type(key).to_sql(self[key])
                 subvars.append(value)
-        assignments = assignments[:-1]  # Knock off last ,
+        assignments = ','.join(assignments)
 
         with self._db.transaction() as tx:
             # Main table update.
@@ -302,6 +380,8 @@ class Model(object):
         self._check_db()
         stored_obj = self._db._get(type(self), self.id)
         assert stored_obj is not None, "object {0} not in DB".format(self.id)
+        self._values_fixed = {}
+        self._values_flex = {}
         self.update(dict(stored_obj))
         self.clear_dirty()
 
@@ -344,76 +424,26 @@ class Model(object):
                     self._dirty.add(key)
             self.store()
 
-
     # Formatting and templating.
 
-    @classmethod
-    def _format(cls, key, value, for_path=False):
-        """Format a value as the given field for this model.
-        """
-        # Format the value as a string according to its type, if any.
-        if key in cls._fields:
-            value = cls._fields[key].format(value)
-            # Formatting must result in a string. To deal with
-            # Python2isms, implicitly convert ASCII strings.
-            assert isinstance(value, basestring), \
-                    u'field formatter must produce strings'
-            if isinstance(value, bytes):
-                value = value.decode('utf8', 'ignore')
+    _formatter = FormattedMapping
 
-        elif not isinstance(value, unicode):
-            # Fallback formatter. Convert to unicode at all cost.
-            if value is None:
-                value = u''
-            elif isinstance(value, basestring):
-                if isinstance(value, bytes):
-                    value = value.decode('utf8', 'ignore')
-            else:
-                value = unicode(value)
-
-        if for_path:
-            sep_repl = beets.config['path_sep_replace'].get(unicode)
-            for sep in (os.path.sep, os.path.altsep):
-                if sep:
-                    value = value.replace(sep, sep_repl)
-
-        return value
-
-    def _get_formatted(self, key, for_path=False):
-        """Get a field value formatted as a string (`unicode` object)
-        for display to the user. If `for_path` is true, then the value
-        will be sanitized for inclusion in a pathname (i.e., path
-        separators will be removed from the value).
-        """
-        return self._format(key, self.get(key), for_path)
-
-    def _formatted_mapping(self, for_path=False):
+    def formatted(self, for_path=False):
         """Get a mapping containing all values on this object formatted
-        as human-readable strings.
+        as human-readable unicode strings.
         """
-        # In the future, this could be made "lazy" to avoid computing
-        # fields unnecessarily.
-        out = {}
-        for key in self.keys(True):
-            out[key] = self._get_formatted(key, for_path)
-        return out
+        return self._formatter(self, for_path)
 
     def evaluate_template(self, template, for_path=False):
         """Evaluate a template (a string or a `Template` object) using
         the object's fields. If `for_path` is true, then no new path
         separators will be added to the template.
         """
-        # Build value mapping.
-        mapping = self._formatted_mapping(for_path)
-
-        # Get template functions.
-        funcs = self._template_funcs()
-
         # Perform substitution.
         if isinstance(template, basestring):
             template = Template(template)
-        return template.substitute(mapping, funcs)
-
+        return template.substitute(self.formatted(for_path),
+                                   self._template_funcs())
 
     # Parsing.
 
@@ -424,63 +454,117 @@ class Model(object):
         if not isinstance(string, basestring):
             raise TypeError("_parse() argument must be a string")
 
-        typ = cls._fields.get(key)
-        if typ:
-            return typ.parse(string)
-        else:
-            # Fall back to unparsed string.
-            return string
-
+        return cls._type(key).parse(string)
 
 
 # Database controller and supporting interfaces.
 
-
 class Results(object):
     """An item query result set. Iterating over the collection lazily
     constructs LibModel objects that reflect database rows.
     """
-    def __init__(self, model_class, rows, db, query=None):
+    def __init__(self, model_class, rows, db, query=None, sort=None):
         """Create a result set that will construct objects of type
-        `model_class`, which should be a subclass of `LibModel`, out of
-        the query result mapping in `rows`. The new objects are
-        associated with the database `db`. If `query` is provided, it is
-        used as a predicate to filter the results for a "slow query" that
-        cannot be evaluated by the database directly.
+        `model_class`.
+
+        `model_class` is a subclass of `LibModel` that will be
+        constructed. `rows` is a query result: a list of mappings. The
+        new objects will be associated with the database `db`.
+
+        If `query` is provided, it is used as a predicate to filter the
+        results for a "slow query" that cannot be evaluated by the
+        database directly. If `sort` is provided, it is used to sort the
+        full list of results before returning. This means it is a "slow
+        sort" and all objects must be built before returning the first
+        one.
         """
         self.model_class = model_class
         self.rows = rows
         self.db = db
         self.query = query
+        self.sort = sort
+
+        # We keep a queue of rows we haven't yet consumed for
+        # materialization. We preserve the original total number of
+        # rows.
+        self._rows = rows
+        self._row_count = len(rows)
+
+        # The materialized objects corresponding to rows that have been
+        # consumed.
+        self._objects = []
+
+    def _get_objects(self):
+        """Construct and generate Model objects for they query. The
+        objects are returned in the order emitted from the database; no
+        slow sort is applied.
+
+        For performance, this generator caches materialized objects to
+        avoid constructing them more than once. This way, iterating over
+        a `Results` object a second time should be much faster than the
+        first.
+        """
+        index = 0  # Position in the materialized objects.
+        while index < len(self._objects) or self._rows:
+            # Are there previously-materialized objects to produce?
+            if index < len(self._objects):
+                yield self._objects[index]
+                index += 1
+
+            # Otherwise, we consume another row, materialize its object
+            # and produce it.
+            else:
+                while self._rows:
+                    row = self._rows.pop(0)
+                    obj = self._make_model(row)
+                    # If there is a slow-query predicate, ensurer that the
+                    # object passes it.
+                    if not self.query or self.query.match(obj):
+                        self._objects.append(obj)
+                        index += 1
+                        yield obj
+                        break
 
     def __iter__(self):
-        """Construct Python objects for all rows that pass the query
-        predicate.
+        """Construct and generate Model objects for all matching
+        objects, in sorted order.
         """
-        for row in self.rows:
-            # Get the flexible attributes for the object.
-            with self.db.transaction() as tx:
-                flex_rows = tx.query(
-                    'SELECT * FROM {0} WHERE entity_id=?'.format(
-                        self.model_class._flex_table
-                    ),
-                    (row['id'],)
-                )
-            values = dict(row)
-            values.update(
-                dict((row['key'], row['value']) for row in flex_rows)
+        if self.sort:
+            # Slow sort. Must build the full list first.
+            objects = self.sort.sort(list(self._get_objects()))
+            return iter(objects)
+
+        else:
+            # Objects are pre-sorted (i.e., by the database).
+            return self._get_objects()
+
+    def _make_model(self, row):
+        # Get the flexible attributes for the object.
+        with self.db.transaction() as tx:
+            flex_rows = tx.query(
+                'SELECT * FROM {0} WHERE entity_id=?'.format(
+                    self.model_class._flex_table
+                ),
+                (row['id'],)
             )
 
-            # Construct the Python object and yield it if it passes the
-            # predicate.
-            obj = self.model_class(self.db, **values)
-            if not self.query or self.query.match(obj):
-                yield obj
+        cols = dict(row)
+        values = dict((k, v) for (k, v) in cols.items()
+                      if not k[:4] == 'flex')
+        flex_values = dict((row['key'], row['value']) for row in flex_rows)
+
+        # Construct the Python object
+        obj = self.model_class._awaken(self.db, values, flex_values)
+        return obj
 
     def __len__(self):
         """Get the number of matching objects.
         """
-        if self.query:
+        if not self._rows:
+            # Fully materialized. Just count the objects.
+            return len(self._objects)
+
+        elif self.query:
             # A slow query. Fall back to testing every object.
             count = 0
             for obj in self:
@@ -489,7 +573,7 @@ class Results(object):
 
         else:
             # A fast query. Just count the rows.
-            return len(self.rows)
+            return self._row_count
 
     def __nonzero__(self):
         """Does this result contain any objects?
@@ -500,6 +584,11 @@ class Results(object):
         """Get the nth item in this result set. This is inefficient: all
         items up to n are materialized and thrown away.
         """
+        if not self._rows and not self.sort:
+            # Fully materialized and already in order. Just look up the
+            # object.
+            return self._objects[n]
+
         it = iter(self)
         try:
             for i in range(n):
@@ -604,7 +693,6 @@ class Database(object):
             self._make_table(model_cls._table, model_cls._fields)
             self._make_attribute_table(model_cls._flex_table)
 
-
     # Primitive access control: connections and transactions.
 
     def _connection(self):
@@ -644,7 +732,6 @@ class Database(object):
         """
         return Transaction(self)
 
-
     # Schema setup and migration.
 
     def _make_table(self, table, fields):
@@ -698,27 +785,33 @@ class Database(object):
                     ON {0} (entity_id);
                 """.format(flex_table))
 
-
     # Querying.
 
-    def _fetch(self, model_cls, query, order_by=None):
+    def _fetch(self, model_cls, query=None, sort=None):
         """Fetch the objects of type `model_cls` matching the given
         query. The query may be given as a string, string sequence, a
-        Query object, or None (to fetch everything). If provided,
-        `order_by` is a SQLite ORDER BY clause for sorting.
+        Query object, or None (to fetch everything). `sort` is an
+        `Sort` object.
         """
+        query = query or TrueQuery()  # A null query.
+        sort = sort or NullSort()  # Unsorted.
         where, subvals = query.clause()
+        order_by = sort.order_clause()
 
-        sql = "SELECT * FROM {0} WHERE {1}".format(
+        sql = ("SELECT * FROM {0} WHERE {1} {2}").format(
             model_cls._table,
             where or '1',
+            "ORDER BY {0}".format(order_by) if order_by else '',
         )
-        if order_by:
-            sql += " ORDER BY {0}".format(order_by)
+
         with self.transaction() as tx:
             rows = tx.query(sql, subvals)
 
-        return Results(model_cls, rows, self, None if where else query)
+        return Results(
+            model_cls, rows, self,
+            None if where else query,  # Slow query component.
+            sort if sort.is_slow() else None,  # Slow sort component.
+        )
 
     def _get(self, model_cls, id):
         """Get a Model object by its id or None if the id does not

lib/beets/dbcore/query.py

@@ -15,6 +15,7 @@
 """The Query type hierarchy for DBCore.
 """
 import re
+from operator import attrgetter
 from beets import util
 from datetime import datetime, timedelta
 
@@ -82,6 +83,23 @@ class MatchQuery(FieldQuery):
         return pattern == value
 
 
+class NoneQuery(FieldQuery):
+
+    def __init__(self, field, fast=True):
+        self.field = field
+        self.fast = fast
+
+    def col_clause(self):
+        return self.field + " IS NULL", ()
+
+    @classmethod
+    def match(self, item):
+        try:
+            return item[self.field] is None
+        except KeyError:
+            return True
+
+
 class StringFieldQuery(FieldQuery):
     """A FieldQuery that converts values to strings before matching
     them.
@@ -104,8 +122,11 @@ class StringFieldQuery(FieldQuery):
 class SubstringQuery(StringFieldQuery):
     """A query that matches a substring in a specific item field."""
     def col_clause(self):
-        search = '%' + (self.pattern.replace('\\','\\\\').replace('%','\\%')
-                            .replace('_','\\_')) + '%'
+        pattern = (self.pattern
+                       .replace('\\', '\\\\')
+                       .replace('%', '\\%')
+                       .replace('_', '\\_'))
+        search = '%' + pattern + '%'
         clause = self.field + " like ? escape '\\'"
         subvals = [search]
         return clause, subvals
@@ -200,7 +221,9 @@ class NumericQuery(FieldQuery):
             self.rangemax = self._convert(parts[1])
 
     def match(self, item):
-        value = getattr(item, self.field)
+        if self.field not in item:
+            return False
+        value = item[self.field]
         if isinstance(value, basestring):
             value = self._convert(value)
 
@@ -236,12 +259,16 @@ class CollectionQuery(Query):
         self.subqueries = subqueries
 
     # Act like a sequence.
+
     def __len__(self):
         return len(self.subqueries)
+
     def __getitem__(self, key):
         return self.subqueries[key]
+
     def __iter__(self):
         return iter(self.subqueries)
+
     def __contains__(self, item):
         return item in self.subqueries
 
@@ -334,10 +361,8 @@ class FalseQuery(Query):
         return False
 
 
-
 # Time/date queries.
 
-
 def _to_epoch_time(date):
     """Convert a `datetime` object to an integer number of seconds since
     the (local) Unix epoch.
@@ -393,10 +418,14 @@ class Period(object):
             return None
         ordinal = string.count('-')
         if ordinal >= len(cls.date_formats):
-            raise ValueError('date is not in one of the formats '
-                             + ', '.join(cls.date_formats))
+            # Too many components.
+            return None
         date_format = cls.date_formats[ordinal]
-        date = datetime.strptime(string, date_format)
+        try:
+            date = datetime.strptime(string, date_format)
+        except ValueError:
+            # Parsing failed.
+            return None
         precision = cls.precisions[ordinal]
         return cls(date, precision)
 
@@ -492,3 +521,134 @@ class DateQuery(FieldQuery):
             # Match any date.
             clause = '1'
         return clause, subvals
+
+
+# Sorting.
+
+class Sort(object):
+    """An abstract class representing a sort operation for a query into
+    the item database.
+    """
+
+    def order_clause(self):
+        """Generates a SQL fragment to be used in a ORDER BY clause, or
+        None if no fragment is used (i.e., this is a slow sort).
+        """
+        return None
+
+    def sort(self, items):
+        """Sort the list of objects and return a list.
+        """
+        return sorted(items)
+
+    def is_slow(self):
+        """Indicate whether this query is *slow*, meaning that it cannot
+        be executed in SQL and must be executed in Python.
+        """
+        return False
+
+
+class MultipleSort(Sort):
+    """Sort that encapsulates multiple sub-sorts.
+    """
+
+    def __init__(self, sorts=None):
+        self.sorts = sorts or []
+
+    def add_sort(self, sort):
+        self.sorts.append(sort)
+
+    def _sql_sorts(self):
+        """Return the list of sub-sorts for which we can be (at least
+        partially) fast.
+
+        A contiguous suffix of fast (SQL-capable) sub-sorts are
+        executable in SQL. The remaining, even if they are fast
+        independently, must be executed slowly.
+        """
+        sql_sorts = []
+        for sort in reversed(self.sorts):
+            if not sort.order_clause() is None:
+                sql_sorts.append(sort)
+            else:
+                break
+        sql_sorts.reverse()
+        return sql_sorts
+
+    def order_clause(self):
+        order_strings = []
+        for sort in self._sql_sorts():
+            order = sort.order_clause()
+            order_strings.append(order)
+
+        return ", ".join(order_strings)
+
+    def is_slow(self):
+        for sort in self.sorts:
+            if sort.is_slow():
+                return True
+        return False
+
+    def sort(self, items):
+        slow_sorts = []
+        switch_slow = False
+        for sort in reversed(self.sorts):
+            if switch_slow:
+                slow_sorts.append(sort)
+            elif sort.order_clause() is None:
+                switch_slow = True
+                slow_sorts.append(sort)
+            else:
+                pass
+
+        for sort in slow_sorts:
+            items = sort.sort(items)
+        return items
+
+    def __repr__(self):
+        return u'MultipleSort({0})'.format(repr(self.sorts))
+
+
+class FieldSort(Sort):
+    """An abstract sort criterion that orders by a specific field (of
+    any kind).
+    """
+    def __init__(self, field, ascending=True):
+        self.field = field
+        self.ascending = ascending
+
+    def sort(self, objs):
+        # TODO: Conversion and null-detection here. In Python 3,
+        # comparisons with None fail. We should also support flexible
+        # attributes with different types without falling over.
+        return sorted(objs, key=attrgetter(self.field),
+                      reverse=not self.ascending)
+
+    def __repr__(self):
+        return u'<{0}: {1}{2}>'.format(
+            type(self).__name__,
+            self.field,
+            '+' if self.ascending else '-',
+        )
+
+
+class FixedFieldSort(FieldSort):
+    """Sort object to sort on a fixed field.
+    """
+    def order_clause(self):
+        order = "ASC" if self.ascending else "DESC"
+        return "{0} {1}".format(self.field, order)
+
+
+class SlowFieldSort(FieldSort):
+    """A sort criterion by some model field other than a fixed field:
+    i.e., a computed or flexible field.
+    """
+    def is_slow(self):
+        return True
+
+
+class NullSort(Sort):
+    """No sorting. Leave results unsorted."""
+    def sort(items):
+        return items

lib/beets/dbcore/queryparse.py

@@ -0,0 +1,180 @@
+# This file is part of beets.
+# Copyright 2014, Adrian Sampson.
+#
+# Permission is hereby granted, free of charge, to any person obtaining
+# a copy of this software and associated documentation files (the
+# "Software"), to deal in the Software without restriction, including
+# without limitation the rights to use, copy, modify, merge, publish,
+# distribute, sublicense, and/or sell copies of the Software, and to
+# permit persons to whom the Software is furnished to do so, subject to
+# the following conditions:
+#
+# The above copyright notice and this permission notice shall be
+# included in all copies or substantial portions of the Software.
+
+"""Parsing of strings into DBCore queries.
+"""
+import re
+import itertools
+from . import query
+
+
+PARSE_QUERY_PART_REGEX = re.compile(
+    # Non-capturing optional segment for the keyword.
+    r'(?:'
+    r'(\S+?)'    # The field key.
+    r'(?<!\\):'  # Unescaped :
+    r')?'
+
+    r'(.*)',         # The term itself.
+
+    re.I  # Case-insensitive.
+)
+
+
+def parse_query_part(part, query_classes={}, prefixes={},
+                     default_class=query.SubstringQuery):
+    """Take a query in the form of a key/value pair separated by a
+    colon and return a tuple of `(key, value, cls)`. `key` may be None,
+    indicating that any field may be matched. `cls` is a subclass of
+    `FieldQuery`.
+
+    The optional `query_classes` parameter maps field names to default
+    query types; `default_class` is the fallback. `prefixes` is a map
+    from query prefix markers and query types. Prefix-indicated queries
+    take precedence over type-based queries.
+
+    To determine the query class, two factors are used: prefixes and
+    field types. For example, the colon prefix denotes a regular
+    expression query and a type map might provide a special kind of
+    query for numeric values. If neither a prefix nor a specific query
+    class is available, `default_class` is used.
+
+    For instance,
+    'stapler' -> (None, 'stapler', SubstringQuery)
+    'color:red' -> ('color', 'red', SubstringQuery)
+    ':^Quiet' -> (None, '^Quiet', RegexpQuery)
+    'color::b..e' -> ('color', 'b..e', RegexpQuery)
+
+    Prefixes may be "escaped" with a backslash to disable the keying
+    behavior.
+    """
+    part = part.strip()
+    match = PARSE_QUERY_PART_REGEX.match(part)
+
+    assert match  # Regex should always match.
+    key = match.group(1)
+    term = match.group(2).replace('\:', ':')
+
+    # Match the search term against the list of prefixes.
+    for pre, query_class in prefixes.items():
+        if term.startswith(pre):
+            return key, term[len(pre):], query_class
+
+    # No matching prefix: use type-based or fallback/default query.
+    query_class = query_classes.get(key, default_class)
+    return key, term, query_class
+
+
+def construct_query_part(model_cls, prefixes, query_part):
+    """Create a query from a single query component, `query_part`, for
+    querying instances of `model_cls`. Return a `Query` instance.
+    """
+    # Shortcut for empty query parts.
+    if not query_part:
+        return query.TrueQuery()
+
+    # Get the query classes for each possible field.
+    query_classes = {}
+    for k, t in itertools.chain(model_cls._fields.items(),
+                                model_cls._types.items()):
+        query_classes[k] = t.query
+
+    # Parse the string.
+    key, pattern, query_class = \
+        parse_query_part(query_part, query_classes, prefixes)
+
+    # No key specified.
+    if key is None:
+        if issubclass(query_class, query.FieldQuery):
+            # The query type matches a specific field, but none was
+            # specified. So we use a version of the query that matches
+            # any field.
+            return query.AnyFieldQuery(pattern, model_cls._search_fields,
+                                       query_class)
+        else:
+            # Other query type.
+            return query_class(pattern)
+
+    key = key.lower()
+    return query_class(key.lower(), pattern, key in model_cls._fields)
+
+
+def query_from_strings(query_cls, model_cls, prefixes, query_parts):
+    """Creates a collection query of type `query_cls` from a list of
+    strings in the format used by parse_query_part. `model_cls`
+    determines how queries are constructed from strings.
+    """
+    subqueries = []
+    for part in query_parts:
+        subqueries.append(construct_query_part(model_cls, prefixes, part))
+    if not subqueries:  # No terms in query.
+        subqueries = [query.TrueQuery()]
+    return query_cls(subqueries)
+
+
+def construct_sort_part(model_cls, part):
+    """Create a `Sort` from a single string criterion.
+
+    `model_cls` is the `Model` being queried. `part` is a single string
+    ending in ``+`` or ``-`` indicating the sort.
+    """
+    assert part, "part must be a field name and + or -"
+    field = part[:-1]
+    assert field, "field is missing"
+    direction = part[-1]
+    assert direction in ('+', '-'), "part must end with + or -"
+    is_ascending = direction == '+'
+
+    if field in model_cls._sorts:
+        sort = model_cls._sorts[field](model_cls, is_ascending)
+    elif field in model_cls._fields:
+        sort = query.FixedFieldSort(field, is_ascending)
+    else:
+        # Flexible or computed.
+        sort = query.SlowFieldSort(field, is_ascending)
+    return sort
+
+
+def sort_from_strings(model_cls, sort_parts):
+    """Create a `Sort` from a list of sort criteria (strings).
+    """
+    if not sort_parts:
+        return query.NullSort()
+    else:
+        sort = query.MultipleSort()
+        for part in sort_parts:
+            sort.add_sort(construct_sort_part(model_cls, part))
+        return sort
+
+
+def parse_sorted_query(model_cls, parts, prefixes={},
+                       query_cls=query.AndQuery):
+    """Given a list of strings, create the `Query` and `Sort` that they
+    represent.
+    """
+    # Separate query token and sort token.
+    query_parts = []
+    sort_parts = []
+    for part in parts:
+        if part.endswith((u'+', u'-')) and u':' not in part:
+            sort_parts.append(part)
+        else:
+            query_parts.append(part)
+
+    # Parse each.
+    q = query_from_strings(
+        query_cls, model_cls, prefixes, query_parts
+    )
+    s = sort_from_strings(model_cls, sort_parts)
+    return q, s

lib/beets/dbcore/types.py

@@ -18,55 +18,111 @@ from . import query
 from beets.util import str2bool
 
 
-
 # Abstract base.
 
-
 class Type(object):
     """An object encapsulating the type of a model field. Includes
-    information about how to store the value in the database, query,
-    format, and parse a given field.
+    information about how to store, query, format, and parse a given
+    field.
     """
 
-    sql = None
+    sql = u'TEXT'
     """The SQLite column type for the value.
     """
 
-    query = None
+    query = query.SubstringQuery
     """The `Query` subclass to be used when querying the field.
     """
 
+    model_type = unicode
+    """The Python type that is used to represent the value in the model.
+
+    The model is guaranteed to return a value of this type if the field
+    is accessed.  To this end, the constructor is used by the `normalize`
+    and `from_sql` methods and the `default` property.
+    """
+
+    @property
+    def null(self):
+        """The value to be exposed when the underlying value is None.
+        """
+        return self.model_type()
+
     def format(self, value):
         """Given a value of this type, produce a Unicode string
         representing the value. This is used in template evaluation.
         """
-        raise NotImplementedError()
+        if value is None:
+            value = self.null
+        # `self.null` might be `None`
+        if value is None:
+            value = u''
+        if isinstance(value, bytes):
+            value = value.decode('utf8', 'ignore')
+
+        return unicode(value)
 
     def parse(self, string):
         """Parse a (possibly human-written) string and return the
         indicated value of this type.
         """
-        raise NotImplementedError()
+        try:
+            return self.model_type(string)
+        except ValueError:
+            return self.null
 
+    def normalize(self, value):
+        """Given a value that will be assigned into a field of this
+        type, normalize the value to have the appropriate type. This
+        base implementation only reinterprets `None`.
+        """
+        if value is None:
+            return self.null
+        else:
+            # TODO This should eventually be replaced by
+            # `self.model_type(value)`
+            return value
+
+    def from_sql(self, sql_value):
+        """Receives the value stored in the SQL backend and return the
+        value to be stored in the model.
+
+        For fixed fields the type of `value` is determined by the column
+        type affinity given in the `sql` property and the SQL to Python
+        mapping of the database adapter. For more information see:
+        http://www.sqlite.org/datatype3.html
+        https://docs.python.org/2/library/sqlite3.html#sqlite-and-python-types
+
+        Flexible fields have the type afinity `TEXT`. This means the
+        `sql_value` is either a `buffer` or a `unicode` object` and the
+        method must handle these in addition.
+        """
+        if isinstance(sql_value, buffer):
+            sql_value = bytes(sql_value).decode('utf8', 'ignore')
+        if isinstance(sql_value, unicode):
+            return self.parse(sql_value)
+        else:
+            return self.normalize(sql_value)
+
+    def to_sql(self, model_value):
+        """Convert a value as stored in the model object to a value used
+        by the database adapter.
+        """
+        return model_value
 
 
 # Reusable types.
 
+class Default(Type):
+    null = None
+
 
 class Integer(Type):
     """A basic integer type.
     """
     sql = u'INTEGER'
     query = query.NumericQuery
-
-    def format(self, value):
-        return unicode(value or 0)
-
-    def parse(self, string):
-        try:
-            return int(string)
-        except ValueError:
-            return 0
+    model_type = int
 
 
 class PaddedInt(Integer):
@@ -93,9 +149,14 @@ class ScaledInt(Integer):
 
 
 class Id(Integer):
-    """An integer used as the row key for a SQLite table.
+    """An integer used as the row id or a foreign key in a SQLite table.
+    This type is nullable: None values are not translated to zero.
     """
-    sql = u'INTEGER PRIMARY KEY'
+    null = None
+
+    def __init__(self, primary=True):
+        if primary:
+            self.sql = u'INTEGER PRIMARY KEY'
 
 
 class Float(Type):
@@ -103,15 +164,16 @@ class Float(Type):
     """
     sql = u'REAL'
     query = query.NumericQuery
+    model_type = float
 
     def format(self, value):
         return u'{0:.1f}'.format(value or 0.0)
 
-    def parse(self, string):
-        try:
-            return float(string)
-        except ValueError:
-            return 0.0
+
+class NullFloat(Float):
+    """Same as `Float`, but does not normalize `None` to `0.0`.
+    """
+    null = None
 
 
 class String(Type):
@@ -120,21 +182,27 @@ class String(Type):
     sql = u'TEXT'
     query = query.SubstringQuery
 
-    def format(self, value):
-        return unicode(value) if value else u''
-
-    def parse(self, string):
-        return string
-
 
 class Boolean(Type):
     """A boolean type.
     """
     sql = u'INTEGER'
     query = query.BooleanQuery
+    model_type = bool
 
     def format(self, value):
         return unicode(bool(value))
 
     def parse(self, string):
         return str2bool(string)
+
+
+# Shared instances of common types.
+DEFAULT = Default()
+INTEGER = Integer()
+PRIMARY_ID = Id(True)
+FOREIGN_ID = Id(False)
+FLOAT = Float()
+NULL_FLOAT = NullFloat()
+STRING = String()
+BOOLEAN = Boolean()

lib/beets/plugins.py

@@ -16,8 +16,11 @@
 
 import logging
 import traceback
+import inspect
+import re
 from collections import defaultdict
 
+
 import beets
 from beets import mediafile
 
@@ -30,6 +33,14 @@ LASTFM_KEY = '2dc3914abf35f0d9c92d97d8f8e42b43'
 log = logging.getLogger('beets')
 
 
+class PluginConflictException(Exception):
+    """Indicates that the services provided by one plugin conflict with
+    those of another.
+
+    For example two plugins may define different types for flexible fields.
+    """
+
+
 # Managing the plugins themselves.
 
 class BeetsPlugin(object):
@@ -40,7 +51,6 @@ class BeetsPlugin(object):
     def __init__(self, name=None):
         """Perform one-time plugin setup.
         """
-        _add_media_fields(self.item_fields())
         self.import_stages = []
         self.name = name or self.__module__.split('.')[-1]
         self.config = beets.config[self.name]
@@ -86,14 +96,6 @@ class BeetsPlugin(object):
         """
         return ()
 
-    def item_fields(self):
-        """Returns field descriptors to be added to the MediaFile class,
-        in the form of a dictionary whose keys are field names and whose
-        values are descriptor (e.g., MediaField) instances. The Library
-        database schema is not (currently) extended.
-        """
-        return {}
-
     def album_for_id(self, album_id):
         """Return an AlbumInfo object or None if no matching release was
         found.
@@ -106,6 +108,20 @@ class BeetsPlugin(object):
         """
         return None
 
+    def add_media_field(self, name, descriptor):
+        """Add a field that is synchronized between media files and items.
+
+        When a media field is added ``item.write()`` will set the name
+        property of the item's MediaFile to ``item[name]`` and save the
+        changes. Similarly ``item.read()`` will set ``item[name]`` to
+        the value of the name property of the media file.
+
+        ``descriptor`` must be an instance of ``mediafile.MediaField``.
+        """
+        # Defer impor to prevent circular dependency
+        from beets import library
+        mediafile.MediaFile.add_field(name, descriptor)
+        library.Item._media_fields.add(name)
 
     listeners = None
 
@@ -130,7 +146,7 @@ class BeetsPlugin(object):
 
             >>> @MyPlugin.listen("imported")
             >>> def importListener(**kwargs):
-            >>>     pass
+            ...     pass
         """
         def helper(func):
             if cls.listeners is None:
@@ -170,7 +186,10 @@ class BeetsPlugin(object):
             return func
         return helper
 
+
 _classes = set()
+
+
 def load_plugins(names=()):
     """Imports the modules for a sequence of plugin names. Each name
     must be the name of a Python module under the "beetsplug" namespace
@@ -185,7 +204,7 @@ def load_plugins(names=()):
             except ImportError as exc:
                 # Again, this is hacky:
                 if exc.args[0].endswith(' ' + name):
-                    log.warn('** plugin %s not found' % name)
+                    log.warn(u'** plugin {0} not found'.format(name))
                 else:
                     raise
             else:
@@ -195,10 +214,13 @@ def load_plugins(names=()):
                         _classes.add(obj)
 
         except:
-            log.warn('** error loading plugin %s' % name)
+            log.warn(u'** error loading plugin {0}'.format(name))
             log.warn(traceback.format_exc())
 
+
 _instances = {}
+
+
 def find_plugins():
     """Returns a list of BeetsPlugin subclass instances from all
     currently loaded beets plugins. Loads the default plugin set
@@ -224,6 +246,7 @@ def commands():
         out += plugin.commands()
     return out
 
+
 def queries():
     """Returns a dict mapping prefix strings to Query subclasses all loaded
     plugins.
@@ -233,6 +256,24 @@ def queries():
         out.update(plugin.queries())
     return out
 
+
+def types(model_cls):
+    # Gives us `item_types` and `album_types`
+    attr_name = '{0}_types'.format(model_cls.__name__.lower())
+    types = {}
+    for plugin in find_plugins():
+        plugin_types = getattr(plugin, attr_name, {})
+        for field in plugin_types:
+            if field in types and plugin_types[field] != types[field]:
+                raise PluginConflictException(
+                    u'Plugin {0} defines flexible field {1} '
+                    'which has already been defined with '
+                    'another type.'.format(plugin.name, field)
+                )
+        types.update(plugin_types)
+    return types
+
+
 def track_distance(item, info):
     """Gets the track distance calculated by all loaded plugins.
     Returns a Distance object.
@@ -243,6 +284,7 @@ def track_distance(item, info):
         dist.update(plugin.track_distance(item, info))
     return dist
 
+
 def album_distance(items, album_info, mapping):
     """Returns the album distance calculated by plugins."""
     from beets.autotag.hooks import Distance
@@ -251,6 +293,7 @@ def album_distance(items, album_info, mapping):
         dist.update(plugin.album_distance(items, album_info, mapping))
     return dist
 
+
 def candidates(items, artist, album, va_likely):
     """Gets MusicBrainz candidates for an album from each plugin.
     """
@@ -259,6 +302,7 @@ def candidates(items, artist, album, va_likely):
         out.extend(plugin.candidates(items, artist, album, va_likely))
     return out
 
+
 def item_candidates(item, artist, title):
     """Gets MusicBrainz candidates for an item from the plugins.
     """
@@ -267,6 +311,7 @@ def item_candidates(item, artist, title):
         out.extend(plugin.item_candidates(item, artist, title))
     return out
 
+
 def album_for_id(album_id):
     """Get AlbumInfo objects for a given ID string.
     """
@@ -277,6 +322,7 @@ def album_for_id(album_id):
             out.append(res)
     return out
 
+
 def track_for_id(track_id):
     """Get TrackInfo objects for a given ID string.
     """
@@ -287,6 +333,7 @@ def track_for_id(track_id):
             out.append(res)
     return out
 
+
 def template_funcs():
     """Get all the template functions declared by plugins as a
     dictionary.
@@ -297,12 +344,6 @@ def template_funcs():
             funcs.update(plugin.template_funcs)
     return funcs
 
-def _add_media_fields(fields):
-    """Adds a {name: descriptor} dictionary of fields to the MediaFile
-    class. Called during the plugin initialization.
-    """
-    for key, value in fields.iteritems():
-        setattr(mediafile.MediaFile, key, value)
 
 def import_stages():
     """Get a list of import stage functions defined by plugins."""
@@ -325,6 +366,7 @@ def item_field_getters():
             funcs.update(plugin.template_fields)
     return funcs
 
+
 def album_field_getters():
     """As above, for album fields.
     """
@@ -348,6 +390,7 @@ def event_handlers():
                 all_handlers[event] += handlers
     return all_handlers
 
+
 def send(event, **arguments):
     """Sends an event to all assigned event listeners. Event is the
     name of  the event to send, all other named arguments go to the
@@ -355,5 +398,38 @@ def send(event, **arguments):
 
     Returns a list of return values from the handlers.
     """
-    log.debug('Sending event: %s' % event)
-    return [handler(**arguments) for handler in event_handlers()[event]]
+    log.debug(u'Sending event: {0}'.format(event))
+    for handler in event_handlers()[event]:
+        # Don't break legacy plugins if we want to pass more arguments
+        argspec = inspect.getargspec(handler).args
+        args = dict((k, v) for k, v in arguments.items() if k in argspec)
+        handler(**args)
+
+
+def feat_tokens(for_artist=True):
+    """Return a regular expression that matches phrases like "featuring"
+    that separate a main artist or a song title from secondary artists.
+    The `for_artist` option determines whether the regex should be
+    suitable for matching artist fields (the default) or title fields.
+    """
+    feat_words = ['ft', 'featuring', 'feat', 'feat.', 'ft.']
+    if for_artist:
+        feat_words += ['with', 'vs', 'and', 'con', '&']
+    return '(?<=\s)(?:{0})(?=\s)'.format(
+        '|'.join(re.escape(x) for x in feat_words)
+    )
+
+
+def sanitize_choices(choices, choices_all):
+    """Clean up a stringlist configuration attribute: keep only choices
+    elements present in choices_all, remove duplicate elements, expand '*'
+    wildcard while keeping original stringlist order.
+    """
+    seen = set()
+    others = [x for x in choices_all if x not in choices]
+    res = []
+    for s in choices:
+        if s in list(choices_all) + ['*']:
+            if not (s in seen or seen.add(s)):
+                res.extend(list(others) if s == '*' else [s])
+    return res

lib/beets/ui/__init__.py

@@ -29,6 +29,7 @@ import errno
 import re
 import struct
 import traceback
+import os.path
 
 from beets import library
 from beets import plugins
@@ -38,9 +39,7 @@ from beets import config
 from beets.util import confit
 from beets.autotag import mb
 
-
 # On Windows platforms, use colorama to support "ANSI" terminal colors.
-
 if sys.platform == 'win32':
     try:
         import colorama
@@ -50,8 +49,10 @@ if sys.platform == 'win32':
         colorama.init()
 
 
-
-# Constants.
+log = logging.getLogger('beets')
+if not log.handlers:
+    log.addHandler(logging.StreamHandler())
+log.propagate = False  # Don't propagate to root handler.
 
 
 PF_KEY_QUERIES = {
@@ -59,19 +60,15 @@ PF_KEY_QUERIES = {
     'singleton': 'singleton:true',
 }
 
-# UI exception. Commands should throw this in order to display
-# nonrecoverable errors to the user.
+
 class UserError(Exception):
-    pass
-
-# Main logger.
-log = logging.getLogger('beets')
-
+    """UI exception. Commands should throw this in order to display
+    nonrecoverable errors to the user.
+    """
 
 
 # Utilities.
 
-
 def _encoding():
     """Tries to guess the encoding used by the terminal."""
     # Configured override?
@@ -170,7 +167,7 @@ def input_options(options, require=False, prompt=None, fallback_prompt=None,
             # Infer a letter.
             for letter in option:
                 if not letter.isalpha():
-                    continue # Don't use punctuation.
+                    continue  # Don't use punctuation.
                 if letter not in letters:
                     found_letter = letter
                     break
@@ -181,9 +178,10 @@ def input_options(options, require=False, prompt=None, fallback_prompt=None,
         index = option.index(found_letter)
 
         # Mark the option's shortcut letter for display.
-        if not require and ((default is None and not numrange and first) or
-                (isinstance(default, basestring) and
-                 found_letter.lower() == default.lower())):
+        if not require and (
+            (default is None and not numrange and first) or
+            (isinstance(default, basestring) and
+             found_letter.lower() == default.lower())):
             # The first option is the default; mark it.
             show_letter = '[%s]' % found_letter.upper()
             is_default = True
@@ -352,11 +350,13 @@ def human_seconds_short(interval):
 # http://dev.pocoo.org/hg/pygments-main/file/b2deea5b5030/pygments/console.py
 # (pygments is by Tim Hatch, Armin Ronacher, et al.)
 COLOR_ESCAPE = "\x1b["
-DARK_COLORS  = ["black", "darkred", "darkgreen", "brown", "darkblue",
-                "purple", "teal", "lightgray"]
+DARK_COLORS = ["black", "darkred", "darkgreen", "brown", "darkblue",
+               "purple", "teal", "lightgray"]
 LIGHT_COLORS = ["darkgray", "red", "green", "yellow", "blue",
                 "fuchsia", "turquoise", "white"]
 RESET_COLOR = COLOR_ESCAPE + "39;49;00m"
+
+
 def _colorize(color, text):
     """Returns a string that prints the given text in the given color
     in a terminal that is ANSI color-aware. The color must be something
@@ -441,30 +441,6 @@ def colordiff(a, b, highlight='red'):
         return unicode(a), unicode(b)
 
 
-def color_diff_suffix(a, b, highlight='red'):
-    """Colorize the differing suffix between two strings."""
-    a, b = unicode(a), unicode(b)
-    if not config['color']:
-        return a, b
-
-    # Fast path.
-    if a == b:
-        return a, b
-
-    # Find the longest common prefix.
-    first_diff = None
-    for i in range(min(len(a), len(b))):
-        if a[i] != b[i]:
-            first_diff = i
-            break
-    else:
-        first_diff = min(len(a), len(b))
-
-    # Colorize from the first difference on.
-    return a[:first_diff] + colorize(highlight, a[first_diff:]), \
-           b[:first_diff] + colorize(highlight, b[first_diff:])
-
-
 def get_path_formats(subview=None):
     """Get the configuration's path formats as a list of query/template
     pairs.
@@ -494,21 +470,6 @@ def get_replacements():
     return replacements
 
 
-def get_plugin_paths():
-    """Get the list of search paths for plugins from the config file.
-    The value for "pluginpath" may be a single string or a list of
-    strings.
-    """
-    pluginpaths = config['pluginpath'].get()
-    if isinstance(pluginpaths, basestring):
-        pluginpaths = [pluginpaths]
-    if not isinstance(pluginpaths, list):
-        raise confit.ConfigTypeError(
-            u'pluginpath must be string or a list of strings'
-        )
-    return map(util.normpath, pluginpaths)
-
-
 def _pick_format(album, fmt=None):
     """Pick a format string for printing Album or Item objects,
     falling back to config options and defaults.
@@ -558,6 +519,8 @@ def term_width():
 
 
 FLOAT_EPSILON = 0.01
+
+
 def _field_diff(field, old, new):
     """Given two Model objects, format their values for `field` and
     highlight changes among them. Return a human-readable string. If the
@@ -574,13 +537,13 @@ def _field_diff(field, old, new):
         return None
 
     # Get formatted values for output.
-    oldstr = old._get_formatted(field)
-    newstr = new._get_formatted(field)
+    oldstr = old.formatted().get(field, u'')
+    newstr = new.formatted().get(field, u'')
 
     # For strings, highlight changes. For others, colorize the whole
     # thing.
     if isinstance(oldval, basestring):
-        oldstr, newstr = colordiff(oldval, newval)
+        oldstr, newstr = colordiff(oldval, newstr)
     else:
         oldstr, newstr = colorize('red', oldstr), colorize('red', newstr)
 
@@ -613,9 +576,12 @@ def show_model_changes(new, old=None, fields=None, always=False):
 
     # New fields.
     for field in set(new) - set(old):
+        if fields and field not in fields:
+            continue
+
         changes.append(u'  {0}: {1}'.format(
             field,
-            colorize('red', new._get_formatted(field))
+            colorize('red', new.formatted()[field])
         ))
 
     # Print changes.
@@ -627,10 +593,8 @@ def show_model_changes(new, old=None, fields=None, always=False):
     return bool(changes)
 
 
-
 # Subcommand parsing infrastructure.
-
-
+#
 # This is a fairly generic subcommand parser for optparse. It is
 # maintained externally here:
 # http://gist.github.com/462717
@@ -653,46 +617,56 @@ class Subcommand(object):
         self.aliases = aliases
         self.help = help
         self.hide = hide
+        self._root_parser = None
+
+    def print_help(self):
+        self.parser.print_help()
+
+    def parse_args(self, args):
+        return self.parser.parse_args(args)
+
+    @property
+    def root_parser(self):
+        return self._root_parser
+
+    @root_parser.setter
+    def root_parser(self, root_parser):
+        self._root_parser = root_parser
+        self.parser.prog = '{0} {1}'.format(root_parser.get_prog_name(),
+                                            self.name)
+
 
 class SubcommandsOptionParser(optparse.OptionParser):
     """A variant of OptionParser that parses subcommands and their
     arguments.
     """
-    # A singleton command used to give help on other subcommands.
-    _HelpSubcommand = Subcommand('help', optparse.OptionParser(),
-        help='give detailed help on a specific sub-command',
-        aliases=('?',))
 
     def __init__(self, *args, **kwargs):
         """Create a new subcommand-aware option parser. All of the
         options to OptionParser.__init__ are supported in addition
         to subcommands, a sequence of Subcommand objects.
         """
-        # The subcommand array, with the help command included.
-        self.subcommands = list(kwargs.pop('subcommands', []))
-        self.subcommands.append(self._HelpSubcommand)
-
         # A more helpful default usage.
         if 'usage' not in kwargs:
             kwargs['usage'] = """
   %prog COMMAND [ARGS...]
   %prog help COMMAND"""
+        kwargs['add_help_option'] = False
 
         # Super constructor.
         optparse.OptionParser.__init__(self, *args, **kwargs)
 
-        # Adjust the help-visible name of each subcommand.
-        for subcommand in self.subcommands:
-            subcommand.parser.prog = '%s %s' % \
-                    (self.get_prog_name(), subcommand.name)
-
         # Our root parser needs to stop on the first unrecognized argument.
         self.disable_interspersed_args()
 
-    def add_subcommand(self, cmd):
+        self.subcommands = []
+
+    def add_subcommand(self, *cmds):
         """Adds a Subcommand object to the parser's list of commands.
         """
-        self.subcommands.append(cmd)
+        for cmd in cmds:
+            cmd.root_parser = self
+            self.subcommands.append(cmd)
 
     # Add the list of subcommands to the help message.
     def format_help(self, formatter=None):
@@ -711,6 +685,7 @@ class SubcommandsOptionParser(optparse.OptionParser):
         disp_names = []
         help_position = 0
         subcommands = [c for c in self.subcommands if not c.hide]
+        subcommands.sort(key=lambda c: c.name)
         for subcommand in subcommands:
             name = subcommand.name
             if subcommand.aliases:
@@ -756,52 +731,40 @@ class SubcommandsOptionParser(optparse.OptionParser):
                 return subcommand
         return None
 
-    def parse_args(self, a=None, v=None):
-        """Like OptionParser.parse_args, but returns these four items:
-        - options: the options passed to the root parser
-        - subcommand: the Subcommand object that was invoked
-        - suboptions: the options passed to the subcommand parser
-        - subargs: the positional arguments passed to the subcommand
+    def parse_global_options(self, args):
+        """Parse options up to the subcommand argument. Returns a tuple
+        of the options object and the remaining arguments.
         """
-        options, args = optparse.OptionParser.parse_args(self, a, v)
-        subcommand, suboptions, subargs = self._parse_sub(args)
-        return options, subcommand, suboptions, subargs
+        options, subargs = self.parse_args(args)
 
-    def _parse_sub(self, args):
-        """Given the `args` left unused by a typical OptionParser
-        `parse_args`, return the invoked subcommand, the subcommand
-        options, and the subcommand arguments.
+        # Force the help command
+        if options.help:
+            subargs = ['help']
+        elif options.version:
+            subargs = ['version']
+        return options, subargs
+
+    def parse_subcommand(self, args):
+        """Given the `args` left unused by a `parse_global_options`,
+        return the invoked subcommand, the subcommand options, and the
+        subcommand arguments.
         """
+        # Help is default command
         if not args:
-            # No command given.
-            self.print_help()
-            self.exit()
-        else:
-            cmdname = args.pop(0)
-            subcommand = self._subcommand_for_name(cmdname)
-            if not subcommand:
-                self.error('unknown command ' + cmdname)
+            args = ['help']
 
-        suboptions, subargs = subcommand.parser.parse_args(args)
-
-        if subcommand is self._HelpSubcommand:
-            if subargs:
-                # particular
-                cmdname = subargs[0]
-                helpcommand = self._subcommand_for_name(cmdname)
-                if not helpcommand:
-                    self.error('no command named {0}'.format(cmdname))
-                helpcommand.parser.print_help()
-                self.exit()
-            else:
-                # general
-                self.print_help()
-                self.exit()
+        cmdname = args.pop(0)
+        subcommand = self._subcommand_for_name(cmdname)
+        if not subcommand:
+            raise UserError("unknown command '{0}'".format(cmdname))
 
+        suboptions, subargs = subcommand.parse_args(args)
         return subcommand, suboptions, subargs
 
 
 optparse.Option.ALWAYS_TYPED_ACTIONS += ('callback',)
+
+
 def vararg_callback(option, opt_str, value, parser):
     """Callback for an option with variable arguments.
     Manually collect arguments right of a callback-action
@@ -838,53 +801,54 @@ def vararg_callback(option, opt_str, value, parser):
     setattr(parser.values, option.dest, value)
 
 
-
 # The main entry point and bootstrapping.
 
-
-def _load_plugins():
+def _load_plugins(config):
     """Load the plugins specified in the configuration.
     """
-    # Add plugin paths.
+    paths = config['pluginpath'].get(confit.StrSeq(split=False))
+    paths = map(util.normpath, paths)
+
     import beetsplug
-    beetsplug.__path__ = get_plugin_paths() + beetsplug.__path__
-
+    beetsplug.__path__ = paths + beetsplug.__path__
     # For backwards compatibility.
-    sys.path += get_plugin_paths()
+    sys.path += paths
 
-    # Load requested plugins.
     plugins.load_plugins(config['plugins'].as_str_seq())
     plugins.send("pluginload")
+    return plugins
 
 
-def _configure(args):
-    """Parse the command line, load configuration files (including
-    loading any indicated plugins), and return the invoked subcomand,
-    the subcommand options, and the subcommand arguments.
+def _setup(options, lib=None):
+    """Prepare and global state and updates it with command line options.
+
+    Returns a list of subcommands, a list of plugins, and a library instance.
     """
-    # Temporary: Migrate from 1.0-style configuration.
-    from beets.ui import migrate
-    migrate.automigrate()
+    # Configure the MusicBrainz API.
+    mb.configure()
+
+    config = _configure(options)
+
+    plugins = _load_plugins(config)
 
     # Get the default subcommands.
     from beets.ui.commands import default_commands
 
-    # Construct the root parser.
-    commands = list(default_commands)
-    commands.append(migrate.migrate_cmd)  # Temporary.
-    parser = SubcommandsOptionParser(subcommands=commands)
-    parser.add_option('-l', '--library', dest='library',
-                      help='library database file to use')
-    parser.add_option('-d', '--directory', dest='directory',
-                      help="destination music directory")
-    parser.add_option('-v', '--verbose', dest='verbose', action='store_true',
-                      help='print debugging information')
-    parser.add_option('-c', '--config', dest='config',
-                      help='path to configuration file')
+    subcommands = list(default_commands)
+    subcommands.extend(plugins.commands())
 
-    # Parse the command-line!
-    options, args = optparse.OptionParser.parse_args(parser, args)
+    if lib is None:
+        lib = _open_library(config)
+        plugins.send("library_opened", lib=lib)
+    library.Item._types = plugins.types(library.Item)
+    library.Album._types = plugins.types(library.Album)
 
+    return subcommands, plugins, lib
+
+
+def _configure(options):
+    """Amend the global configuration object with command line options.
+    """
     # Add any additional config files specified with --config. This
     # special handling lets specified plugins get loaded before we
     # finish parsing the command line.
@@ -894,22 +858,28 @@ def _configure(args):
         config.set_file(config_path)
     config.set_args(options)
 
-    # Now add the plugin commands to the parser.
-    _load_plugins()
-    for cmd in plugins.commands():
-        parser.add_subcommand(cmd)
+    # Configure the logger.
+    if config['verbose'].get(bool):
+        log.setLevel(logging.DEBUG)
+    else:
+        log.setLevel(logging.INFO)
 
-    # Parse the remainder of the command line with loaded plugins.
-    return parser._parse_sub(args)
+    config_path = config.user_config_path()
+    if os.path.isfile(config_path):
+        log.debug(u'user configuration: {0}'.format(
+            util.displayable_path(config_path)))
+    else:
+        log.debug(u'no user configuration found at {0}'.format(
+            util.displayable_path(config_path)))
+
+    log.debug(u'data directory: {0}'
+              .format(util.displayable_path(config.config_dir())))
+    return config
 
 
-def _raw_main(args):
-    """A helper function for `main` without top-level exception
-    handling.
+def _open_library(config):
+    """Create a new library instance from the configuration.
     """
-    subcommand, suboptions, subargs = _configure(args)
-
-    # Open library file.
     dbpath = config['library'].as_filename()
     try:
         lib = library.Library(
@@ -918,32 +888,52 @@ def _raw_main(args):
             get_path_formats(),
             get_replacements(),
         )
-    except sqlite3.OperationalError:
+        lib.get_item(0)  # Test database connection.
+    except (sqlite3.OperationalError, sqlite3.DatabaseError):
+        log.debug(traceback.format_exc())
         raise UserError(u"database file {0} could not be opened".format(
             util.displayable_path(dbpath)
         ))
-    plugins.send("library_opened", lib=lib)
+    log.debug(u'library database: {0}\n'
+              u'library directory: {1}'
+              .format(util.displayable_path(lib.path),
+                      util.displayable_path(lib.directory)))
+    return lib
 
-    # Configure the logger.
-    if config['verbose'].get(bool):
-        log.setLevel(logging.DEBUG)
-    else:
-        log.setLevel(logging.INFO)
-    log.debug(u'data directory: {0}\n'
-              u'library database: {1}\n'
-              u'library directory: {2}'
-        .format(
-            util.displayable_path(config.config_dir()),
-            util.displayable_path(lib.path),
-            util.displayable_path(lib.directory),
-        )
-    )
 
-    # Configure the MusicBrainz API.
-    mb.configure()
+def _raw_main(args, lib=None):
+    """A helper function for `main` without top-level exception
+    handling.
+    """
+    parser = SubcommandsOptionParser()
+    parser.add_option('-l', '--library', dest='library',
+                      help='library database file to use')
+    parser.add_option('-d', '--directory', dest='directory',
+                      help="destination music directory")
+    parser.add_option('-v', '--verbose', dest='verbose', action='store_true',
+                      help='print debugging information')
+    parser.add_option('-c', '--config', dest='config',
+                      help='path to configuration file')
+    parser.add_option('-h', '--help', dest='help', action='store_true',
+                      help='how this help message and exit')
+    parser.add_option('--version', dest='version', action='store_true',
+                      help=optparse.SUPPRESS_HELP)
 
-    # Invoke the subcommand.
+    options, subargs = parser.parse_global_options(args)
+
+    # Special case for the `config --edit` command: bypass _setup so
+    # that an invalid configuration does not prevent the editor from
+    # starting.
+    if subargs[0] == 'config' and ('-e' in subargs or '--edit' in subargs):
+        from beets.ui.commands import config_edit
+        return config_edit()
+
+    subcommands, plugins, lib = _setup(options, lib)
+    parser.add_subcommand(*subcommands)
+
+    subcommand, suboptions, subargs = parser.parse_subcommand(subargs)
     subcommand.func(lib, suboptions, subargs)
+
     plugins.send('cli_exit', lib=lib)
 
 

lib/beets/ui/migrate.py

@@ -1,401 +0,0 @@
-# This file is part of beets.
-# Copyright 2013, Adrian Sampson.
-#
-# Permission is hereby granted, free of charge, to any person obtaining
-# a copy of this software and associated documentation files (the
-# "Software"), to deal in the Software without restriction, including
-# without limitation the rights to use, copy, modify, merge, publish,
-# distribute, sublicense, and/or sell copies of the Software, and to
-# permit persons to whom the Software is furnished to do so, subject to
-# the following conditions:
-#
-# The above copyright notice and this permission notice shall be
-# included in all copies or substantial portions of the Software.
-
-"""Conversion from legacy (pre-1.1) configuration to Confit/YAML
-configuration.
-"""
-import os
-import ConfigParser
-import codecs
-import yaml
-import logging
-import time
-import itertools
-import re
-
-import beets
-from beets import util
-from beets import ui
-from beets.util import confit
-
-CONFIG_PATH_VAR = 'BEETSCONFIG'
-DEFAULT_CONFIG_FILENAME_UNIX = '.beetsconfig'
-DEFAULT_CONFIG_FILENAME_WINDOWS = 'beetsconfig.ini'
-DEFAULT_LIBRARY_FILENAME_UNIX = '.beetsmusic.blb'
-DEFAULT_LIBRARY_FILENAME_WINDOWS = 'beetsmusic.blb'
-WINDOWS_BASEDIR = os.environ.get('APPDATA') or '~'
-
-OLD_CONFIG_SUFFIX = '.old'
-PLUGIN_NAMES = {
-    'rdm': 'random',
-    'fuzzy_search': 'fuzzy',
-}
-AUTO_KEYS = ('automatic', 'autofetch', 'autoembed', 'autoscrub')
-IMPORTFEEDS_PREFIX = 'feeds_'
-CONFIG_MIGRATED_MESSAGE = u"""
-You appear to be upgrading from beets 1.0 (or earlier) to 1.1. Your
-configuration file has been migrated automatically to:
-{newconfig}
-Edit this file to configure beets. You might want to remove your
-old-style ".beetsconfig" file now. See the documentation for more
-details on the new configuration system:
-http://beets.readthedocs.org/page/reference/config.html
-""".strip()
-DB_MIGRATED_MESSAGE = u'Your database file has also been copied to:\n{newdb}'
-YAML_COMMENT = '# Automatically migrated from legacy .beetsconfig.\n\n'
-
-log = logging.getLogger('beets')
-
-# An itertools recipe.
-def grouper(n, iterable):
-    args = [iter(iterable)] * n
-    return itertools.izip_longest(*args)
-
-def _displace(fn):
-    """Move a file aside using a timestamp suffix so a new file can be
-    put in its place.
-    """
-    util.move(
-        fn,
-        u'{0}.old.{1}'.format(fn, int(time.time())),
-        True
-    )
-
-def default_paths():
-    """Produces the appropriate default config and library database
-    paths for the current system. On Unix, this is always in ~. On
-    Windows, tries ~ first and then $APPDATA for the config and library
-    files (for backwards compatibility).
-    """
-    windows = os.path.__name__ == 'ntpath'
-    if windows:
-        windata = os.environ.get('APPDATA') or '~'
-
-    # Shorthand for joining paths.
-    def exp(*vals):
-        return os.path.expanduser(os.path.join(*vals))
-
-    config = exp('~', DEFAULT_CONFIG_FILENAME_UNIX)
-    if windows and not os.path.exists(config):
-        config = exp(windata, DEFAULT_CONFIG_FILENAME_WINDOWS)
-
-    libpath = exp('~', DEFAULT_LIBRARY_FILENAME_UNIX)
-    if windows and not os.path.exists(libpath):
-        libpath = exp(windata, DEFAULT_LIBRARY_FILENAME_WINDOWS)
-
-    return config, libpath
-
-def get_config():
-    """Using the same logic as beets 1.0, locate and read the
-    .beetsconfig file. Return a ConfigParser instance or None if no
-    config is found.
-    """
-    default_config, default_libpath = default_paths()
-    if CONFIG_PATH_VAR in os.environ:
-        configpath = os.path.expanduser(os.environ[CONFIG_PATH_VAR])
-    else:
-        configpath = default_config
-
-    config = ConfigParser.SafeConfigParser()
-    if os.path.exists(util.syspath(configpath)):
-        with codecs.open(configpath, 'r', encoding='utf-8') as f:
-            config.readfp(f)
-        return config, configpath
-    else:
-        return None, configpath
-
-def flatten_config(config):
-    """Given a ConfigParser, flatten the values into a dict-of-dicts
-    representation where each section gets its own dictionary of values.
-    """
-    out = confit.OrderedDict()
-    for section in config.sections():
-        sec_dict = out[section] = confit.OrderedDict()
-        for option in config.options(section):
-            sec_dict[option] = config.get(section, option, True)
-    return out
-
-def transform_value(value):
-    """Given a string read as the value of a config option, return a
-    massaged version of that value (possibly with a different type).
-    """
-    # Booleans.
-    if value.lower() in ('false', 'no', 'off'):
-        return False
-    elif value.lower() in ('true', 'yes', 'on'):
-        return True
-
-    # Integers.
-    try:
-        return int(value)
-    except ValueError:
-        pass
-
-    # Floats.
-    try:
-        return float(value)
-    except ValueError:
-        pass
-
-    return value
-
-def transform_data(data):
-    """Given a dict-of-dicts representation of legacy config data, tweak
-    the data into a new form. This new form is suitable for dumping as
-    YAML.
-    """
-    out = confit.OrderedDict()
-
-    for section, pairs in data.items():
-        if section == 'beets':
-            # The "main" section. In the new config system, these values
-            # are in the "root": no section at all.
-            for key, value in pairs.items():
-                value = transform_value(value)
-
-                if key.startswith('import_'):
-                    # Importer config is now under an "import:" key.
-                    if 'import' not in out:
-                        out['import'] = confit.OrderedDict()
-                    out['import'][key[7:]] = value
-
-                elif key == 'plugins':
-                    # Renamed plugins.
-                    plugins = value.split()
-                    new_plugins = [PLUGIN_NAMES.get(p, p) for p in plugins]
-                    out['plugins'] = ' '.join(new_plugins)
-
-                elif key == 'replace':
-                    # YAMLy representation for character replacements.
-                    replacements = confit.OrderedDict()
-                    for pat, repl in grouper(2, value.split()):
-                        if repl == '<strip>':
-                            repl = ''
-                        replacements[pat] = repl
-                    out['replace'] = replacements
-
-                elif key == 'pluginpath':
-                    # Used to be a colon-separated string. Now a list.
-                    out['pluginpath'] = value.split(':')
-
-                else:
-                    out[key] = value
-
-        elif pairs:
-            # Other sections (plugins, etc).
-            sec_out = out[section] = confit.OrderedDict()
-            for key, value in pairs.items():
-
-                # Standardized "auto" option.
-                if key in AUTO_KEYS:
-                    key = 'auto'
-
-                # Unnecessary : hack in queries.
-                if section == 'paths':
-                    key = key.replace('_', ':')
-
-                # Changed option names for importfeeds plugin.
-                if section == 'importfeeds':
-                    if key.startswith(IMPORTFEEDS_PREFIX):
-                        key = key[len(IMPORTFEEDS_PREFIX):]
-
-                sec_out[key] = transform_value(value)
-
-    return out
-
-class Dumper(yaml.SafeDumper):
-    """A PyYAML Dumper that represents OrderedDicts as ordinary mappings
-    (in order, of course).
-    """
-    # From http://pyyaml.org/attachment/ticket/161/use_ordered_dict.py
-    def represent_mapping(self, tag, mapping, flow_style=None):
-        value = []
-        node = yaml.MappingNode(tag, value, flow_style=flow_style)
-        if self.alias_key is not None:
-            self.represented_objects[self.alias_key] = node
-        best_style = True
-        if hasattr(mapping, 'items'):
-            mapping = list(mapping.items())
-        for item_key, item_value in mapping:
-            node_key = self.represent_data(item_key)
-            node_value = self.represent_data(item_value)
-            if not (isinstance(node_key, yaml.ScalarNode) and \
-                    not node_key.style):
-                best_style = False
-            if not (isinstance(node_value, yaml.ScalarNode) and \
-                    not node_value.style):
-                best_style = False
-            value.append((node_key, node_value))
-        if flow_style is None:
-            if self.default_flow_style is not None:
-                node.flow_style = self.default_flow_style
-            else:
-                node.flow_style = best_style
-        return node
-Dumper.add_representer(confit.OrderedDict, Dumper.represent_dict)
-
-def migrate_config(replace=False):
-    """Migrate a legacy beetsconfig file to a new-style config.yaml file
-    in an appropriate place. If `replace` is enabled, then any existing
-    config.yaml will be moved aside. Otherwise, the process is aborted
-    when the file exists.
-    """
-
-    # Load legacy configuration data, if any.
-    config, configpath = get_config()
-    if not config:
-        log.debug(u'no config file found at {0}'.format(
-            util.displayable_path(configpath)
-            ))
-        return
-
-    # Get the new configuration file path and possibly move it out of
-    # the way.
-    destfn = os.path.join(beets.config.config_dir(), confit.CONFIG_FILENAME)
-    if os.path.exists(destfn):
-        if replace:
-            log.debug(u'moving old config aside: {0}'.format(
-                util.displayable_path(destfn)
-            ))
-            _displace(destfn)
-        else:
-            # File exists and we won't replace it. We're done.
-            return
-
-    log.debug(u'migrating config file {0}'.format(
-        util.displayable_path(configpath)
-    ))
-
-    # Convert the configuration to a data structure ready to be dumped
-    # as the new Confit file.
-    data = transform_data(flatten_config(config))
-
-    # Encode result as YAML.
-    yaml_out = yaml.dump(
-        data,
-        Dumper=Dumper,
-        default_flow_style=False,
-        indent=4,
-        width=1000,
-    )
-    # A ridiculous little hack to add some whitespace between "sections"
-    # in the YAML output. I hope this doesn't break any YAML syntax.
-    yaml_out = re.sub(r'(\n\w+:\n    [^-\s])', '\n\\1', yaml_out)
-    yaml_out = YAML_COMMENT + yaml_out
-
-    # Write the data to the new config destination.
-    log.debug(u'writing migrated config to {0}'.format(
-        util.displayable_path(destfn)
-    ))
-    with open(destfn, 'w') as f:
-        f.write(yaml_out)
-    return destfn
-
-def migrate_db(replace=False):
-    """Copy the beets library database file to the new location (e.g.,
-    from ~/.beetsmusic.blb to ~/.config/beets/library.db).
-    """
-    _, srcfn = default_paths()
-    destfn = beets.config['library'].as_filename()
-
-    if not os.path.exists(srcfn) or srcfn == destfn:
-        # Old DB does not exist or we're configured to point to the same
-        # database. Do nothing.
-        return
-
-    if os.path.exists(destfn):
-        if replace:
-            log.debug(u'moving old database aside: {0}'.format(
-                util.displayable_path(destfn)
-            ))
-            _displace(destfn)
-        else:
-            return
-
-    log.debug(u'copying database from {0} to {1}'.format(
-        util.displayable_path(srcfn), util.displayable_path(destfn)
-    ))
-    util.copy(srcfn, destfn)
-    return destfn
-
-def migrate_state(replace=False):
-    """Copy the beets runtime state file from the old path (i.e.,
-    ~/.beetsstate) to the new path (i.e., ~/.config/beets/state.pickle).
-    """
-    srcfn = os.path.expanduser(os.path.join('~', '.beetsstate'))
-    if not os.path.exists(srcfn):
-        return
-
-    destfn = beets.config['statefile'].as_filename()
-    if os.path.exists(destfn):
-        if replace:
-            _displace(destfn)
-        else:
-            return
-
-    log.debug(u'copying state file from {0} to {1}'.format(
-        util.displayable_path(srcfn), util.displayable_path(destfn)
-    ))
-    util.copy(srcfn, destfn)
-    return destfn
-
-
-# Automatic migration when beets starts.
-
-def automigrate():
-    """Migrate the configuration, database, and state files. If any
-    migration occurs, print out a notice with some helpful next steps.
-    """
-    config_fn = migrate_config()
-    db_fn = migrate_db()
-    migrate_state()
-
-    if config_fn:
-        ui.print_(ui.colorize('fuchsia', u'MIGRATED CONFIGURATION'))
-
-        ui.print_(CONFIG_MIGRATED_MESSAGE.format(
-            newconfig=util.displayable_path(config_fn))
-        )
-        if db_fn:
-            ui.print_(DB_MIGRATED_MESSAGE.format(
-                newdb=util.displayable_path(db_fn)
-            ))
-
-        ui.input_(ui.colorize('fuchsia', u'Press ENTER to continue:'))
-        ui.print_()
-
-
-# CLI command for explicit migration.
-
-migrate_cmd = ui.Subcommand('migrate', help='convert legacy config')
-def migrate_func(lib, opts, args):
-    """Explicit command for migrating files. Existing files in each
-    destination are moved aside.
-    """
-    config_fn = migrate_config(replace=True)
-    if config_fn:
-        log.info(u'Migrated configuration to: {0}'.format(
-            util.displayable_path(config_fn)
-        ))
-    db_fn = migrate_db(replace=True)
-    if db_fn:
-        log.info(u'Migrated library database to: {0}'.format(
-            util.displayable_path(db_fn)
-        ))
-    state_fn = migrate_state(replace=True)
-    if state_fn:
-        log.info(u'Migrated state file to: {0}'.format(
-            util.displayable_path(state_fn)
-        ))
-migrate_cmd.func = migrate_func

lib/beets/util/__init__.py

@@ -23,10 +23,13 @@ import fnmatch
 from collections import defaultdict
 import traceback
 import subprocess
+import platform
+
 
 MAX_FILENAME_LENGTH = 200
 WINDOWS_MAGIC_PREFIX = u'\\\\?\\'
 
+
 class HumanReadableException(Exception):
     """An Exception that can include a human-readable error message to
     be logged without a traceback. Can preserve a traceback for
@@ -82,6 +85,7 @@ class HumanReadableException(Exception):
             logger.debug(self.tb)
         logger.error(u'{0}: {1}'.format(self.error_kind, self.args[0]))
 
+
 class FilesystemError(HumanReadableException):
     """An error that occurred while performing a filesystem manipulation
     via a function in this module. The `paths` field is a sequence of
@@ -111,6 +115,7 @@ class FilesystemError(HumanReadableException):
 
         return u'{0} {1}'.format(self._reasonstr(), clause)
 
+
 def normpath(path):
     """Provide the canonical form of the path suitable for storing in
     the database.
@@ -119,6 +124,7 @@ def normpath(path):
     path = os.path.normpath(os.path.abspath(os.path.expanduser(path)))
     return bytestring_path(path)
 
+
 def ancestry(path):
     """Return a list consisting of path's parent directory, its
     grandparent, and so on. For instance:
@@ -137,10 +143,12 @@ def ancestry(path):
             break
         last_path = path
 
-        if path: # don't yield ''
+        if path:
+            # don't yield ''
             out.insert(0, path)
     return out
 
+
 def sorted_walk(path, ignore=(), logger=None):
     """Like `os.walk`, but yields things in case-insensitive sorted,
     breadth-first order.  Directory and file names matching any glob
@@ -192,6 +200,7 @@ def sorted_walk(path, ignore=(), logger=None):
         for res in sorted_walk(cur, ignore, logger):
             yield res
 
+
 def mkdirall(path):
     """Make all the enclosing directories of path (like mkdir -p on the
     parent).
@@ -204,6 +213,7 @@ def mkdirall(path):
                 raise FilesystemError(exc, 'create', (ancestor,),
                                       traceback.format_exc())
 
+
 def fnmatch_all(names, patterns):
     """Determine whether all strings in `names` match at least one of
     the `patterns`, which should be shell glob expressions.
@@ -218,6 +228,7 @@ def fnmatch_all(names, patterns):
             return False
     return True
 
+
 def prune_dirs(path, root=None, clutter=('.DS_Store', 'Thumbs.db')):
     """If path is an empty directory, then remove it. Recursively remove
     path's ancestry up to root (which is never removed) where there are
@@ -236,7 +247,7 @@ def prune_dirs(path, root=None, clutter=('.DS_Store', 'Thumbs.db')):
         ancestors = []
     elif root in ancestors:
         # Only remove directories below the root.
-        ancestors = ancestors[ancestors.index(root)+1:]
+        ancestors = ancestors[ancestors.index(root) + 1:]
     else:
         # Remove nothing.
         return
@@ -258,6 +269,7 @@ def prune_dirs(path, root=None, clutter=('.DS_Store', 'Thumbs.db')):
         else:
             break
 
+
 def components(path):
     """Return a list of the path components in path. For instance:
 
@@ -281,6 +293,7 @@ def components(path):
 
     return comps
 
+
 def _fsencoding():
     """Get the system's filesystem encoding. On Windows, this is always
     UTF-8 (not MBCS).
@@ -295,6 +308,7 @@ def _fsencoding():
         encoding = 'utf8'
     return encoding
 
+
 def bytestring_path(path):
     """Given a path, which is either a str or a unicode, returns a str
     path (ensuring that we never deal with Unicode pathnames).
@@ -315,6 +329,7 @@ def bytestring_path(path):
     except (UnicodeError, LookupError):
         return path.encode('utf8')
 
+
 def displayable_path(path, separator=u'; '):
     """Attempts to decode a bytestring path to a unicode object for the
     purpose of displaying it to the user. If the `path` argument is a
@@ -333,6 +348,7 @@ def displayable_path(path, separator=u'; '):
     except (UnicodeError, LookupError):
         return path.decode('utf8', 'ignore')
 
+
 def syspath(path, prefix=True):
     """Convert a path for use by the operating system. In particular,
     paths on Windows must receive a magic prefix and must be converted
@@ -356,16 +372,22 @@ def syspath(path, prefix=True):
             encoding = sys.getfilesystemencoding() or sys.getdefaultencoding()
             path = path.decode(encoding, 'replace')
 
-    # Add the magic prefix if it isn't already there
+    # Add the magic prefix if it isn't already there.
+    # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247.aspx
     if prefix and not path.startswith(WINDOWS_MAGIC_PREFIX):
+        if path.startswith(u'\\\\'):
+            # UNC path. Final path should look like \\?\UNC\...
+            path = u'UNC' + path[1:]
         path = WINDOWS_MAGIC_PREFIX + path
 
     return path
 
+
 def samefile(p1, p2):
     """Safer equality for paths."""
     return shutil._samefile(syspath(p1), syspath(p2))
 
+
 def remove(path, soft=True):
     """Remove the file. If `soft`, then no error will be raised if the
     file does not exist.
@@ -378,6 +400,7 @@ def remove(path, soft=True):
     except (OSError, IOError) as exc:
         raise FilesystemError(exc, 'delete', (path,), traceback.format_exc())
 
+
 def copy(path, dest, replace=False):
     """Copy a plain file. Permissions are not copied. If `dest` already
     exists, raises a FilesystemError unless `replace` is True. Has no
@@ -396,6 +419,7 @@ def copy(path, dest, replace=False):
         raise FilesystemError(exc, 'copy', (path, dest),
                               traceback.format_exc())
 
+
 def move(path, dest, replace=False):
     """Rename a file. `dest` may not be a directory. If `dest` already
     exists, raises an OSError unless `replace` is True. Has no effect if
@@ -424,6 +448,27 @@ def move(path, dest, replace=False):
             raise FilesystemError(exc, 'move', (path, dest),
                                   traceback.format_exc())
 
+
+def link(path, dest, replace=False):
+    """Create a symbolic link from path to `dest`. Raises an OSError if
+    `dest` already exists, unless `replace` is True. Does nothing if
+    `path` == `dest`."""
+    if (samefile(path, dest)):
+        return
+
+    path = syspath(path)
+    dest = syspath(dest)
+    if os.path.exists(dest) and not replace:
+        raise FilesystemError('file exists', 'rename', (path, dest),
+                              traceback.format_exc())
+    try:
+        os.symlink(path, dest)
+    except OSError:
+        raise FilesystemError('Operating system does not support symbolic '
+                              'links.', 'link', (path, dest),
+                              traceback.format_exc())
+
+
 def unique_path(path):
     """Returns a version of ``path`` that does not exist on the
     filesystem. Specifically, if ``path` itself already exists, then
@@ -457,6 +502,8 @@ CHAR_REPLACE = [
     (re.compile(ur'\.$'), u'_'),  # Trailing dots.
     (re.compile(ur'\s+$'), u''),  # Trailing whitespace.
 ]
+
+
 def sanitize_path(path, replacements=None):
     """Takes a path (as a Unicode string) and makes sure that it is
     legal. Returns a new path. Only works with fragments; won't work
@@ -477,6 +524,7 @@ def sanitize_path(path, replacements=None):
         comps[i] = comp
     return os.path.join(*comps)
 
+
 def truncate_path(path, length=MAX_FILENAME_LENGTH):
     """Given a bytestring path or a Unicode path fragment, truncate the
     components to a legal length. In the last component, the extension
@@ -493,6 +541,7 @@ def truncate_path(path, length=MAX_FILENAME_LENGTH):
 
     return os.path.join(*out)
 
+
 def str2bool(value):
     """Returns a boolean reflecting a human-entered string."""
     if value.lower() in ('yes', '1', 'true', 't', 'y'):
@@ -500,6 +549,7 @@ def str2bool(value):
     else:
         return False
 
+
 def as_string(value):
     """Convert a value to a Unicode object for matching with a query.
     None becomes the empty string. Bytestrings are silently decoded.
@@ -513,6 +563,7 @@ def as_string(value):
     else:
         return unicode(value)
 
+
 def levenshtein(s1, s2):
     """A nice DP edit distance implementation from Wikibooks:
     http://en.wikibooks.org/wiki/Algorithm_implementation/Strings/
@@ -535,6 +586,7 @@ def levenshtein(s1, s2):
 
     return previous_row[-1]
 
+
 def plurality(objs):
     """Given a sequence of comparable objects, returns the object that
     is most common in the set and the frequency of that object. The
@@ -558,6 +610,7 @@ def plurality(objs):
 
     return res, max_freq
 
+
 def cpu_count():
     """Return the number of hardware thread contexts (cores or SMT
     threads) in the system.
@@ -571,7 +624,7 @@ def cpu_count():
             num = 0
     elif sys.platform == 'darwin':
         try:
-            num = int(os.popen('sysctl -n hw.ncpu').read())
+            num = int(command_output(['sysctl', '-n', 'hw.ncpu']))
         except ValueError:
             num = 0
     else:
@@ -584,23 +637,38 @@ def cpu_count():
     else:
         return 1
 
-def command_output(cmd):
-    """Wraps the `subprocess` module to invoke a command (given as a
-    list of arguments starting with the command name) and collect
-    stdout. The stderr stream is ignored. May raise
-    `subprocess.CalledProcessError` or an `OSError`.
+
+def command_output(cmd, shell=False):
+    """Runs the command and returns its output after it has exited.
+
+    ``cmd`` is a list of arguments starting with the command names.  If
+    ``shell`` is true, ``cmd`` is assumed to be a string and passed to a
+    shell to execute.
+
+    If the process exits with a non-zero return code
+    ``subprocess.CalledProcessError`` is raised. May also raise
+    ``OSError``.
 
     This replaces `subprocess.check_output`, which isn't available in
     Python 2.6 and which can have problems if lots of output is sent to
     stderr.
     """
-    with open(os.devnull, 'w') as devnull:
-        proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=devnull)
-        stdout, _ = proc.communicate()
+    proc = subprocess.Popen(
+        cmd,
+        stdout=subprocess.PIPE,
+        stderr=subprocess.PIPE,
+        close_fds=platform.system() != 'Windows',
+        shell=shell
+    )
+    stdout, stderr = proc.communicate()
     if proc.returncode:
-        raise subprocess.CalledProcessError(proc.returncode, cmd)
+        raise subprocess.CalledProcessError(
+            returncode=proc.returncode,
+            cmd=' '.join(cmd),
+        )
     return stdout
 
+
 def max_filename_length(path, limit=MAX_FILENAME_LENGTH):
     """Attempt to determine the maximum filename length for the
     filesystem containing `path`. If the value is greater than `limit`,

lib/beets/util/artresizer.py

@@ -1,5 +1,5 @@
 # This file is part of beets.
-# Copyright 2013, Fabrice Laporte
+# Copyright 2014, Fabrice Laporte
 #
 # Permission is hereby granted, free of charge, to any person obtaining
 # a copy of this software and associated documentation files (the
@@ -18,6 +18,7 @@ public resizing proxy if neither is available.
 import urllib
 import subprocess
 import os
+import re
 from tempfile import NamedTemporaryFile
 import logging
 from beets import util
@@ -37,7 +38,7 @@ def resize_url(url, maxwidth):
     maxwidth (preserving aspect ratio).
     """
     return '{0}?{1}'.format(PROXY_URL, urllib.urlencode({
-        'url': url.replace('http://',''),
+        'url': url.replace('http://', ''),
         'w': str(maxwidth),
     }))
 
@@ -76,7 +77,7 @@ def pil_resize(maxwidth, path_in, path_out=None):
 
 def im_resize(maxwidth, path_in, path_out=None):
     """Resize using ImageMagick's ``convert`` tool.
-    tool. Return the output path of resized image.
+    Return the output path of resized image.
     """
     path_out = path_out or temp_file_for(path_in)
     log.debug(u'artresizer: ImageMagick resizing {0} to {1}'.format(
@@ -132,8 +133,9 @@ class ArtResizer(object):
         """Create a resizer object for the given method or, if none is
         specified, with an inferred method.
         """
-        self.method = method or self._guess_method()
+        self.method = self._check_method(method)
         log.debug(u"artresizer: method is {0}".format(self.method))
+        self.can_compare = self._can_compare()
 
     def resize(self, maxwidth, path_in, path_out=None):
         """Manipulate an image file according to the method, returning a
@@ -141,7 +143,7 @@ class ArtResizer(object):
         temporary file. For WEBPROXY, returns `path_in` unmodified.
         """
         if self.local:
-            func = BACKEND_FUNCS[self.method]
+            func = BACKEND_FUNCS[self.method[0]]
             return func(maxwidth, path_in, path_out)
         else:
             return path_in
@@ -159,30 +161,51 @@ class ArtResizer(object):
     @property
     def local(self):
         """A boolean indicating whether the resizing method is performed
-        locally (i.e., PIL or IMAGEMAGICK).
+        locally (i.e., PIL or ImageMagick).
         """
-        return self.method in BACKEND_FUNCS
+        return self.method[0] in BACKEND_FUNCS
+
+    def _can_compare(self):
+        """A boolean indicating whether image comparison is available"""
+
+        return self.method[0] == IMAGEMAGICK and self.method[1] > (6, 8, 7)
 
     @staticmethod
-    def _guess_method():
-        """Determine which resizing method to use. Returns PIL,
-        IMAGEMAGICK, or WEBPROXY depending on available dependencies.
+    def _check_method(method=None):
+        """A tuple indicating whether current method is available and its
+        version. If no method is given, it returns a supported one.
         """
-        # Try importing PIL.
-        try:
-            __import__('PIL', fromlist=['Image'])
-            return PIL
-        except ImportError:
-            pass
+        # Guess available method
+        if not method:
+            for m in [IMAGEMAGICK, PIL]:
+                _, version = ArtResizer._check_method(m)
+                if version:
+                    return (m, version)
+            return (WEBPROXY, (0))
 
-        # Try invoking ImageMagick's "convert".
-        try:
-            out = util.command_output(['convert', '--version'])
-            if 'imagemagick' in out.lower():
-                # system32/convert.exe may be interfering
-                return IMAGEMAGICK
-        except (subprocess.CalledProcessError, OSError):
-            pass
+        if method == IMAGEMAGICK:
 
-        # Fall back to Web proxy method.
-        return WEBPROXY
+            # Try invoking ImageMagick's "convert".
+            try:
+                out = util.command_output(['identify', '--version'])
+
+                if 'imagemagick' in out.lower():
+                    pattern = r".+ (\d+)\.(\d+)\.(\d+).*"
+                    match = re.search(pattern, out)
+                    if match:
+                        return (IMAGEMAGICK,
+                                (int(match.group(1)),
+                                 int(match.group(2)),
+                                 int(match.group(3))))
+                    return (IMAGEMAGICK, (0))
+
+            except (subprocess.CalledProcessError, OSError):
+                return (IMAGEMAGICK, None)
+
+        if method == PIL:
+            # Try importing PIL.
+            try:
+                __import__('PIL', fromlist=['Image'])
+                return (PIL, (0))
+            except ImportError:
+                return (PIL, None)

lib/beets/util/bluelet.py

@@ -38,6 +38,7 @@ class Event(object):
     """
     pass
 
+
 class WaitableEvent(Event):
     """A waitable event is one encapsulating an action that can be
     waited for using a select() call. That is, it's an event with an
@@ -57,21 +58,25 @@ class WaitableEvent(Event):
         """
         pass
 
+
 class ValueEvent(Event):
     """An event that does nothing but return a fixed value."""
     def __init__(self, value):
         self.value = value
 
+
 class ExceptionEvent(Event):
     """Raise an exception at the yield point. Used internally."""
     def __init__(self, exc_info):
         self.exc_info = exc_info
 
+
 class SpawnEvent(Event):
     """Add a new coroutine thread to the scheduler."""
     def __init__(self, coro):
         self.spawned = coro
 
+
 class JoinEvent(Event):
     """Suspend the thread until the specified child thread has
     completed.
@@ -79,11 +84,13 @@ class JoinEvent(Event):
     def __init__(self, child):
         self.child = child
 
+
 class KillEvent(Event):
     """Unschedule a child thread."""
     def __init__(self, child):
         self.child = child
 
+
 class DelegationEvent(Event):
     """Suspend execution of the current thread, start a new thread and,
     once the child thread finished, return control to the parent
@@ -92,6 +99,7 @@ class DelegationEvent(Event):
     def __init__(self, coro):
         self.spawned = coro
 
+
 class ReturnEvent(Event):
     """Return a value the current thread's delegator at the point of
     delegation. Ends the current (delegate) thread.
@@ -99,6 +107,7 @@ class ReturnEvent(Event):
     def __init__(self, value):
         self.value = value
 
+
 class SleepEvent(WaitableEvent):
     """Suspend the thread for a given duration.
     """
@@ -108,6 +117,7 @@ class SleepEvent(WaitableEvent):
     def time_left(self):
         return max(self.wakeup_time - time.time(), 0.0)
 
+
 class ReadEvent(WaitableEvent):
     """Reads from a file-like object."""
     def __init__(self, fd, bufsize):
@@ -120,6 +130,7 @@ class ReadEvent(WaitableEvent):
     def fire(self):
         return self.fd.read(self.bufsize)
 
+
 class WriteEvent(WaitableEvent):
     """Writes to a file-like object."""
     def __init__(self, fd, data):
@@ -192,15 +203,19 @@ def _event_select(events):
 
     return ready_events
 
+
 class ThreadException(Exception):
     def __init__(self, coro, exc_info):
         self.coro = coro
         self.exc_info = exc_info
+
     def reraise(self):
         _reraise(self.exc_info[0], self.exc_info[1], self.exc_info[2])
 
+
 SUSPENDED = Event()  # Special sentinel placeholder for suspended threads.
 
+
 class Delegated(Event):
     """Placeholder indicating that a thread has delegated execution to a
     different thread.
@@ -208,6 +223,7 @@ class Delegated(Event):
     def __init__(self, child):
         self.child = child
 
+
 def run(root_coro):
     """Schedules a coroutine, running it to completion. This
     encapsulates the Bluelet scheduler, which the root coroutine can
@@ -329,7 +345,7 @@ def run(root_coro):
                     break
 
             # Wait and fire.
-            event2coro = dict((v,k) for k,v in threads.items())
+            event2coro = dict((v, k) for k, v in threads.items())
             for event in _event_select(threads.values()):
                 # Run the IO operation, but catch socket errors.
                 try:
@@ -378,6 +394,7 @@ def run(root_coro):
 class SocketClosedError(Exception):
     pass
 
+
 class Listener(object):
     """A socket wrapper object for listening sockets.
     """
@@ -407,6 +424,7 @@ class Listener(object):
         self._closed = True
         self.sock.close()
 
+
 class Connection(object):
     """A socket wrapper object for connected sockets.
     """
@@ -468,6 +486,7 @@ class Connection(object):
                 yield ReturnEvent(line)
                 break
 
+
 class AcceptEvent(WaitableEvent):
     """An event for Listener objects (listening sockets) that suspends
     execution until the socket gets a connection.
@@ -482,6 +501,7 @@ class AcceptEvent(WaitableEvent):
         sock, addr = self.listener.sock.accept()
         return Connection(sock, addr)
 
+
 class ReceiveEvent(WaitableEvent):
     """An event for Connection objects (connected sockets) for
     asynchronously reading data.
@@ -496,6 +516,7 @@ class ReceiveEvent(WaitableEvent):
     def fire(self):
         return self.conn.sock.recv(self.bufsize)
 
+
 class SendEvent(WaitableEvent):
     """An event for Connection objects (connected sockets) for
     asynchronously writing data.
@@ -523,6 +544,7 @@ def null():
     """
     return ValueEvent(None)
 
+
 def spawn(coro):
     """Event: add another coroutine to the scheduler. Both the parent
     and child coroutines run concurrently.
@@ -531,6 +553,7 @@ def spawn(coro):
         raise ValueError('%s is not a coroutine' % str(coro))
     return SpawnEvent(coro)
 
+
 def call(coro):
     """Event: delegate to another coroutine. The current coroutine
     is resumed once the sub-coroutine finishes. If the sub-coroutine
@@ -540,12 +563,14 @@ def call(coro):
         raise ValueError('%s is not a coroutine' % str(coro))
     return DelegationEvent(coro)
 
+
 def end(value=None):
     """Event: ends the coroutine and returns a value to its
     delegator.
     """
     return ReturnEvent(value)
 
+
 def read(fd, bufsize=None):
     """Event: read from a file descriptor asynchronously."""
     if bufsize is None:
@@ -563,10 +588,12 @@ def read(fd, bufsize=None):
     else:
         return ReadEvent(fd, bufsize)
 
+
 def write(fd, data):
     """Event: write to a file descriptor asynchronously."""
     return WriteEvent(fd, data)
 
+
 def connect(host, port):
     """Event: connect to a network address and return a Connection
     object for communicating on the socket.
@@ -575,17 +602,20 @@ def connect(host, port):
     sock = socket.create_connection(addr)
     return ValueEvent(Connection(sock, addr))
 
+
 def sleep(duration):
     """Event: suspend the thread for ``duration`` seconds.
     """
     return SleepEvent(duration)
 
+
 def join(coro):
     """Suspend the thread until another, previously `spawn`ed thread
     completes.
     """
     return JoinEvent(coro)
 
+
 def kill(coro):
     """Halt the execution of a different `spawn`ed thread.
     """

lib/beets/util/confit.py

@@ -21,6 +21,8 @@ import pkgutil
 import sys
 import yaml
 import types
+import collections
+import re
 try:
     from collections import OrderedDict
 except ImportError:
@@ -47,6 +49,7 @@ BASESTRING = str if PY3 else basestring
 NUMERIC_TYPES = (int, float) if PY3 else (int, float, long)
 TYPE_TYPES = (type,) if PY3 else (type, types.ClassType)
 
+
 def iter_first(sequence):
     """Get the first element from an iterable or raise a ValueError if
     the iterator generates no values.
@@ -67,16 +70,25 @@ class ConfigError(Exception):
     """Base class for exceptions raised when querying a configuration.
     """
 
+
 class NotFoundError(ConfigError):
     """A requested value could not be found in the configuration trees.
     """
 
-class ConfigTypeError(ConfigError, TypeError):
+
+class ConfigValueError(ConfigError):
+    """The value in the configuration is illegal."""
+
+
+class ConfigTypeError(ConfigValueError):
     """The value in the configuration did not match the expected type.
     """
 
-class ConfigValueError(ConfigError, ValueError):
-    """The value in the configuration is illegal."""
+
+class ConfigTemplateError(ConfigError):
+    """Base class for exceptions raised because of an invalid template.
+    """
+
 
 class ConfigReadError(ConfigError):
     """A configuration file could not be read."""
@@ -132,6 +144,7 @@ class ConfigSource(dict):
         else:
             raise TypeError('source value must be a dict')
 
+
 class ConfigView(object):
     """A configuration "view" is a query into a program's configuration
     data. A view represents a hypothetical location in the configuration
@@ -207,6 +220,9 @@ class ConfigView(object):
         """
         self.set({key: value})
 
+    def __contains__(self, key):
+        return self[key].exists()
+
     def set_args(self, namespace):
         """Overlay parsed command-line arguments, generated by a library
         like argparse or optparse, onto this view's value.
@@ -310,98 +326,6 @@ class ConfigView(object):
 
     # Validation and conversion.
 
-    def get(self, typ=None):
-        """Returns the canonical value for the view, checked against the
-        passed-in type. If the value is not an instance of the given
-        type, a ConfigTypeError is raised. May also raise a
-        NotFoundError.
-        """
-        value, _ = self.first()
-
-        if typ is not None:
-            if not isinstance(typ, TYPE_TYPES):
-                raise TypeError('argument to get() must be a type')
-
-            if not isinstance(value, typ):
-                raise ConfigTypeError(
-                    "{0} must be of type {1}, not {2}".format(
-                        self.name, typ.__name__, type(value).__name__
-                    )
-                )
-
-        return value
-
-    def as_filename(self):
-        """Get a string as a normalized as an absolute, tilde-free path.
-
-        Relative paths are relative to the configuration directory (see
-        the `config_dir` method) if they come from a file. Otherwise,
-        they are relative to the current working directory. This helps
-        attain the expected behavior when using command-line options.
-        """
-        path, source = self.first()
-        if not isinstance(path, BASESTRING):
-            raise ConfigTypeError('{0} must be a filename, not {1}'.format(
-                self.name, type(path).__name__
-            ))
-        path = os.path.expanduser(STRING(path))
-
-        if not os.path.isabs(path) and source.filename:
-            # From defaults: relative to the app's directory.
-            path = os.path.join(self.root().config_dir(), path)
-
-        return os.path.abspath(path)
-
-    def as_choice(self, choices):
-        """Ensure that the value is among a collection of choices and
-        return it. If `choices` is a dictionary, then return the
-        corresponding value rather than the value itself (the key).
-        """
-        value = self.get()
-
-        if value not in choices:
-            raise ConfigValueError(
-                '{0} must be one of {1}, not {2}'.format(
-                    self.name, repr(list(choices)), repr(value)
-                )
-            )
-
-        if isinstance(choices, dict):
-            return choices[value]
-        else:
-            return value
-
-    def as_number(self):
-        """Ensure that a value is of numeric type."""
-        value = self.get()
-        if isinstance(value, NUMERIC_TYPES):
-            return value
-        raise ConfigTypeError(
-            '{0} must be numeric, not {1}'.format(
-                self.name, type(value).__name__
-            )
-        )
-
-    def as_str_seq(self):
-        """Get the value as a list of strings. The underlying configured
-        value can be a sequence or a single string. In the latter case,
-        the string is treated as a white-space separated list of words.
-        """
-        value = self.get()
-        if isinstance(value, bytes):
-            value = value.decode('utf8', 'ignore')
-
-        if isinstance(value, STRING):
-            return value.split()
-        else:
-            try:
-                return list(value)
-            except TypeError:
-                raise ConfigTypeError(
-                    '{0} must be a whitespace-separated string or '
-                    'a list'.format(self.name)
-                )
-
     def flatten(self):
         """Create a hierarchy of OrderedDicts containing the data from
         this view, recursively reifying all views to get their
@@ -415,6 +339,35 @@ class ConfigView(object):
                 od[key] = view.get()
         return od
 
+    def get(self, template=None):
+        """Retrieve the value for this view according to the template.
+
+        The `template` against which the values are checked can be
+        anything convertible to a `Template` using `as_template`. This
+        means you can pass in a default integer or string value, for
+        example, or a type to just check that something matches the type
+        you expect.
+
+        May raise a `ConfigValueError` (or its subclass,
+        `ConfigTypeError`) or a `NotFoundError` when the configuration
+        doesn't satisfy the template.
+        """
+        return as_template(template).value(self, template)
+
+    # Old validation methods (deprecated).
+
+    def as_filename(self):
+        return self.get(Filename())
+
+    def as_choice(self, choices):
+        return self.get(Choice(choices))
+
+    def as_number(self):
+        return self.get(Number())
+
+    def as_str_seq(self):
+        return self.get(StrSeq())
+
 
 class RootView(ConfigView):
     """The base of a view hierarchy. This view keeps track of the
@@ -518,6 +471,7 @@ def _package_path(name):
 
     return os.path.dirname(os.path.abspath(filepath))
 
+
 def config_dirs():
     """Return a platform-specific list of candidates for user
     configuration directories on the system.
@@ -606,10 +560,12 @@ class Loader(yaml.SafeLoader):
         plain = super(Loader, self).check_plain()
         return plain or self.peek() == '%'
 
+
 Loader.add_constructor('tag:yaml.org,2002:str', Loader._construct_unicode)
 Loader.add_constructor('tag:yaml.org,2002:map', Loader.construct_yaml_map)
 Loader.add_constructor('tag:yaml.org,2002:omap', Loader.construct_yaml_map)
 
+
 def load_yaml(filename):
     """Read a YAML document from a file. If the file cannot be read or
     parsed, a ConfigReadError is raised.
@@ -679,11 +635,13 @@ class Dumper(yaml.SafeDumper):
         """
         return self.represent_scalar('tag:yaml.org,2002:null', '')
 
+
 Dumper.add_representer(OrderedDict, Dumper.represent_dict)
 Dumper.add_representer(bool, Dumper.represent_bool)
 Dumper.add_representer(type(None), Dumper.represent_none)
 Dumper.add_representer(list, Dumper.represent_list)
 
+
 def restore_yaml_comments(data, default_data):
     """Scan default_data for comments (we include empty lines in our
     definition of comments) and place them before the same keys in data.
@@ -898,3 +856,426 @@ class LazyConfig(Configuration):
         del self.sources[:]
         self._lazy_suffix = []
         self._lazy_prefix = []
+
+
+# "Validated" configuration views: experimental!
+
+
+REQUIRED = object()
+"""A sentinel indicating that there is no default value and an exception
+should be raised when the value is missing.
+"""
+
+
+class Template(object):
+    """A value template for configuration fields.
+
+    The template works like a type and instructs Confit about how to
+    interpret a deserialized YAML value. This includes type conversions,
+    providing a default value, and validating for errors. For example, a
+    filepath type might expand tildes and check that the file exists.
+    """
+    def __init__(self, default=REQUIRED):
+        """Create a template with a given default value.
+
+        If `default` is the sentinel `REQUIRED` (as it is by default),
+        then an error will be raised when a value is missing. Otherwise,
+        missing values will instead return `default`.
+        """
+        self.default = default
+
+    def __call__(self, view):
+        """Invoking a template on a view gets the view's value according
+        to the template.
+        """
+        return self.value(view, self)
+
+    def value(self, view, template=None):
+        """Get the value for a `ConfigView`.
+
+        May raise a `NotFoundError` if the value is missing (and the
+        template requires it) or a `ConfigValueError` for invalid values.
+        """
+        if view.exists():
+            value, _ = view.first()
+            return self.convert(value, view)
+        elif self.default is REQUIRED:
+            # Missing required value. This is an error.
+            raise NotFoundError("{0} not found".format(view.name))
+        else:
+            # Missing value, but not required.
+            return self.default
+
+    def convert(self, value, view):
+        """Convert the YAML-deserialized value to a value of the desired
+        type.
+
+        Subclasses should override this to provide useful conversions.
+        May raise a `ConfigValueError` when the configuration is wrong.
+        """
+        # Default implementation does no conversion.
+        return value
+
+    def fail(self, message, view, type_error=False):
+        """Raise an exception indicating that a value cannot be
+        accepted.
+
+        `type_error` indicates whether the error is due to a type
+        mismatch rather than a malformed value. In this case, a more
+        specific exception is raised.
+        """
+        exc_class = ConfigTypeError if type_error else ConfigValueError
+        raise exc_class(
+            '{0}: {1}'.format(view.name, message)
+        )
+
+    def __repr__(self):
+        return '{0}({1})'.format(
+            type(self).__name__,
+            '' if self.default is REQUIRED else repr(self.default),
+        )
+
+
+class Integer(Template):
+    """An integer configuration value template.
+    """
+    def convert(self, value, view):
+        """Check that the value is an integer. Floats are rounded.
+        """
+        if isinstance(value, int):
+            return value
+        elif isinstance(value, float):
+            return int(value)
+        else:
+            self.fail('must be a number', view, True)
+
+
+class Number(Template):
+    """A numeric type: either an integer or a floating-point number.
+    """
+    def convert(self, value, view):
+        """Check that the value is an int or a float.
+        """
+        if isinstance(value, NUMERIC_TYPES):
+            return value
+        else:
+            self.fail(
+                'must be numeric, not {0}'.format(type(value).__name__),
+                view,
+                True
+            )
+
+
+class MappingTemplate(Template):
+    """A template that uses a dictionary to specify other types for the
+    values for a set of keys and produce a validated `AttrDict`.
+    """
+    def __init__(self, mapping):
+        """Create a template according to a dict (mapping). The
+        mapping's values should themselves either be Types or
+        convertible to Types.
+        """
+        subtemplates = {}
+        for key, typ in mapping.items():
+            subtemplates[key] = as_template(typ)
+        self.subtemplates = subtemplates
+
+    def value(self, view, template=None):
+        """Get a dict with the same keys as the template and values
+        validated according to the value types.
+        """
+        out = AttrDict()
+        for key, typ in self.subtemplates.items():
+            out[key] = typ.value(view[key], self)
+        return out
+
+    def __repr__(self):
+        return 'MappingTemplate({0})'.format(repr(self.subtemplates))
+
+
+class String(Template):
+    """A string configuration value template.
+    """
+    def __init__(self, default=REQUIRED, pattern=None):
+        """Create a template with the added optional `pattern` argument,
+        a regular expression string that the value should match.
+        """
+        super(String, self).__init__(default)
+        self.pattern = pattern
+        if pattern:
+            self.regex = re.compile(pattern)
+
+    def convert(self, value, view):
+        """Check that the value is a string and matches the pattern.
+        """
+        if isinstance(value, BASESTRING):
+            if self.pattern and not self.regex.match(value):
+                self.fail(
+                    "must match the pattern {0}".format(self.pattern),
+                    view
+                )
+            return value
+        else:
+            self.fail('must be a string', view, True)
+
+
+class Choice(Template):
+    """A template that permits values from a sequence of choices.
+    """
+    def __init__(self, choices):
+        """Create a template that validates any of the values from the
+        iterable `choices`.
+
+        If `choices` is a map, then the corresponding value is emitted.
+        Otherwise, the value itself is emitted.
+        """
+        self.choices = choices
+
+    def convert(self, value, view):
+        """Ensure that the value is among the choices (and remap if the
+        choices are a mapping).
+        """
+        if value not in self.choices:
+            self.fail(
+                'must be one of {0}, not {1}'.format(
+                    repr(list(self.choices)), repr(value)
+                ),
+                view
+            )
+
+        if isinstance(self.choices, collections.Mapping):
+            return self.choices[value]
+        else:
+            return value
+
+    def __repr__(self):
+        return 'Choice({0!r})'.format(self.choices)
+
+
+class StrSeq(Template):
+    """A template for values that are lists of strings.
+
+    Validates both actual YAML string lists and single strings. Strings
+    can optionally be split on whitespace.
+    """
+    def __init__(self, split=True):
+        """Create a new template.
+
+        `split` indicates whether, when the underlying value is a single
+        string, it should be split on whitespace. Otherwise, the
+        resulting value is a list containing a single string.
+        """
+        super(StrSeq, self).__init__()
+        self.split = split
+
+    def convert(self, value, view):
+        if isinstance(value, bytes):
+            value = value.decode('utf8', 'ignore')
+
+        if isinstance(value, STRING):
+            if self.split:
+                return value.split()
+            else:
+                return [value]
+
+        try:
+            value = list(value)
+        except TypeError:
+            self.fail('must be a whitespace-separated string or a list',
+                      view, True)
+
+        def convert(x):
+            if isinstance(x, unicode):
+                return x
+            elif isinstance(x, BASESTRING):
+                return x.decode('utf8', 'ignore')
+            else:
+                self.fail('must be a list of strings', view, True)
+        return map(convert, value)
+
+
+class Filename(Template):
+    """A template that validates strings as filenames.
+
+    Filenames are returned as absolute, tilde-free paths.
+
+    Relative paths are relative to the template's `cwd` argument
+    when it is specified, then the configuration directory (see
+    the `config_dir` method) if they come from a file. Otherwise,
+    they are relative to the current working directory. This helps
+    attain the expected behavior when using command-line options.
+    """
+    def __init__(self, default=REQUIRED, cwd=None, relative_to=None,
+                 in_app_dir=False):
+        """ `relative_to` is the name of a sibling value that is
+        being validated at the same time.
+
+        `in_app_dir` indicates whether the path should be resolved
+        inside the application's config directory (even when the setting
+        does not come from a file).
+        """
+        super(Filename, self).__init__(default)
+        self.cwd = cwd
+        self.relative_to = relative_to
+        self.in_app_dir = in_app_dir
+
+    def __repr__(self):
+        args = []
+
+        if self.default is not REQUIRED:
+            args.append(repr(self.default))
+
+        if self.cwd is not None:
+            args.append('cwd=' + repr(self.cwd))
+
+        if self.relative_to is not None:
+            args.append('relative_to=' + repr(self.relative_to))
+
+        if self.in_app_dir:
+            args.append('in_app_dir=True')
+
+        return 'Filename({0})'.format(', '.join(args))
+
+    def resolve_relative_to(self, view, template):
+        if not isinstance(template, (collections.Mapping, MappingTemplate)):
+            # disallow config.get(Filename(relative_to='foo'))
+            raise ConfigTemplateError(
+                'relative_to may only be used when getting multiple values.'
+            )
+
+        elif self.relative_to == view.key:
+            raise ConfigTemplateError(
+                '{0} is relative to itself'.format(view.name)
+            )
+
+        elif self.relative_to not in view.parent.keys():
+            # self.relative_to is not in the config
+            self.fail(
+                (
+                    'needs sibling value "{0}" to expand relative path'
+                ).format(self.relative_to),
+                view
+            )
+
+        old_template = {}
+        old_template.update(template.subtemplates)
+
+        # save time by skipping MappingTemplate's init loop
+        next_template = MappingTemplate({})
+        next_relative = self.relative_to
+
+        # gather all the needed templates and nothing else
+        while next_relative is not None:
+            try:
+                # pop to avoid infinite loop because of recursive
+                # relative paths
+                rel_to_template = old_template.pop(next_relative)
+            except KeyError:
+                if next_relative in template.subtemplates:
+                    # we encountered this config key previously
+                    raise ConfigTemplateError((
+                        '{0} and {1} are recursively relative'
+                    ).format(view.name, self.relative_to))
+                else:
+                    raise ConfigTemplateError((
+                        'missing template for {0}, needed to expand {1}\'s' +
+                        'relative path'
+                    ).format(self.relative_to, view.name))
+
+            next_template.subtemplates[next_relative] = rel_to_template
+            next_relative = rel_to_template.relative_to
+
+        return view.parent.get(next_template)[self.relative_to]
+
+    def value(self, view, template=None):
+        path, source = view.first()
+        if not isinstance(path, BASESTRING):
+            self.fail(
+                'must be a filename, not {0}'.format(type(path).__name__),
+                view,
+                True
+            )
+        path = os.path.expanduser(STRING(path))
+
+        if not os.path.isabs(path):
+            if self.cwd is not None:
+                # relative to the template's argument
+                path = os.path.join(self.cwd, path)
+
+            elif self.relative_to is not None:
+                path = os.path.join(
+                    self.resolve_relative_to(view, template),
+                    path,
+                )
+
+            elif source.filename or self.in_app_dir:
+                # From defaults: relative to the app's directory.
+                path = os.path.join(view.root().config_dir(), path)
+
+        return os.path.abspath(path)
+
+
+class TypeTemplate(Template):
+    """A simple template that checks that a value is an instance of a
+    desired Python type.
+    """
+    def __init__(self, typ, default=REQUIRED):
+        """Create a template that checks that the value is an instance
+        of `typ`.
+        """
+        super(TypeTemplate, self).__init__(default)
+        self.typ = typ
+
+    def convert(self, value, view):
+        if not isinstance(value, self.typ):
+            self.fail(
+                'must be a {0}, not {1}'.format(
+                    self.typ.__name__,
+                    type(value).__name__,
+                ),
+                view,
+                True
+            )
+        return value
+
+
+class AttrDict(dict):
+    """A `dict` subclass that can be accessed via attributes (dot
+    notation) for convenience.
+    """
+    def __getattr__(self, key):
+        if key in self:
+            return self[key]
+        else:
+            raise AttributeError(key)
+
+
+def as_template(value):
+    """Convert a simple "shorthand" Python value to a `Template`.
+    """
+    if isinstance(value, Template):
+        # If it's already a Template, pass it through.
+        return value
+    elif isinstance(value, collections.Mapping):
+        # Dictionaries work as templates.
+        return MappingTemplate(value)
+    elif value is int:
+        return Integer()
+    elif isinstance(value, int):
+        return Integer(value)
+    elif isinstance(value, type) and issubclass(value, BASESTRING):
+        return String()
+    elif isinstance(value, BASESTRING):
+        return String(value)
+    elif value is float:
+        return Number()
+    elif value is None:
+        return Template()
+    elif value is dict:
+        return TypeTemplate(collections.Mapping)
+    elif value is list:
+        return TypeTemplate(collections.Sequence)
+    elif isinstance(value, type):
+        return TypeTemplate(value)
+    else:
+        raise ValueError('cannot convert to template: {0!r}'.format(value))

lib/beets/util/enumeration.py

@@ -12,167 +12,29 @@
 # The above copyright notice and this permission notice shall be
 # included in all copies or substantial portions of the Software.
 
-"""A metaclass for enumerated types that really are types.
+from enum import Enum
 
-You can create enumerations with `enum(values, [name])` and they work
-how you would expect them to.
 
-    >>> from enumeration import enum
-    >>> Direction = enum('north east south west', name='Direction')
-    >>> Direction.west
-    Direction.west
-    >>> Direction.west == Direction.west
-    True
-    >>> Direction.west == Direction.east
-    False
-    >>> isinstance(Direction.west, Direction)
-    True
-    >>> Direction[3]
-    Direction.west
-    >>> Direction['west']
-    Direction.west
-    >>> Direction.west.name
-    'west'
-    >>> Direction.north < Direction.west
-    True
-
-Enumerations are classes; their instances represent the possible values
-of the enumeration. Because Python classes must have names, you may
-provide a `name` parameter to `enum`; if you don't, a meaningless one
-will be chosen for you.
-"""
-import random
-
-class Enumeration(type):
-    """A metaclass whose classes are enumerations.
-
-    The `values` attribute of the class is used to populate the
-    enumeration. Values may either be a list of enumerated names or a
-    string containing a space-separated list of names. When the class
-    is created, it is instantiated for each name value in `values`.
-    Each such instance is the name of the enumerated item as the sole
-    argument.
-
-    The `Enumerated` class is a good choice for a superclass.
+class OrderedEnum(Enum):
     """
-
-    def __init__(cls, name, bases, dic):
-        super(Enumeration, cls).__init__(name, bases, dic)
-
-        if 'values' not in dic:
-            # Do nothing if no values are provided (i.e., with
-            # Enumerated itself).
-            return
-
-        # May be called with a single string, in which case we split on
-        # whitespace for convenience.
-        values = dic['values']
-        if isinstance(values, basestring):
-            values = values.split()
-
-        # Create the Enumerated instances for each value. We have to use
-        # super's __setattr__ here because we disallow setattr below.
-        super(Enumeration, cls).__setattr__('_items_dict', {})
-        super(Enumeration, cls).__setattr__('_items_list', [])
-        for value in values:
-            item = cls(value, len(cls._items_list))
-            cls._items_dict[value] = item
-            cls._items_list.append(item)
-
-    def __getattr__(cls, key):
-        try:
-            return cls._items_dict[key]
-        except KeyError:
-            raise AttributeError("enumeration '" + cls.__name__ +
-                                 "' has no item '" + key + "'")
-
-    def __setattr__(cls, key, val):
-        raise TypeError("enumerations do not support attribute assignment")
-
-    def __getitem__(cls, key):
-        if isinstance(key, int):
-            return cls._items_list[key]
-        else:
-            return getattr(cls, key)
-
-    def __len__(cls):
-        return len(cls._items_list)
-
-    def __iter__(cls):
-        return iter(cls._items_list)
-
-    def __nonzero__(cls):
-        # Ensures that __len__ doesn't get called before __init__ by
-        # pydoc.
-        return True
-
-class Enumerated(object):
-    """An item in an enumeration.
-
-    Contains instance methods inherited by enumerated objects. The
-    metaclass is preset to `Enumeration` for your convenience.
-
-    Instance attributes:
-    name -- The name of the item.
-    index -- The index of the item in its enumeration.
-
-        >>> from enumeration import Enumerated
-        >>> class Garment(Enumerated):
-        ...     values = 'hat glove belt poncho lederhosen suspenders'
-        ...     def wear(self):
-        ...         print('now wearing a ' + self.name)
-        ...
-        >>> Garment.poncho.wear()
-        now wearing a poncho
+    An Enum subclass that allows comparison of members.
     """
+    def __ge__(self, other):
+        if self.__class__ is other.__class__:
+            return self.value >= other.value
+        return NotImplemented
 
-    __metaclass__ = Enumeration
+    def __gt__(self, other):
+        if self.__class__ is other.__class__:
+            return self.value > other.value
+        return NotImplemented
 
-    def __init__(self, name, index):
-        self.name = name
-        self.index = index
+    def __le__(self, other):
+        if self.__class__ is other.__class__:
+            return self.value <= other.value
+        return NotImplemented
 
-    def __str__(self):
-        return type(self).__name__ + '.' + self.name
-
-    def __repr__(self):
-        return str(self)
-
-    def __cmp__(self, other):
-        if type(self) is type(other):
-            # Note that we're assuming that the items are direct
-            # instances of the same Enumeration (i.e., no fancy
-            # subclassing), which is probably okay.
-            return cmp(self.index, other.index)
-        else:
-            return NotImplemented
-
-def enum(*values, **kwargs):
-    """Shorthand for creating a new Enumeration class.
-
-    Call with enumeration values as a list, a space-delimited string, or
-    just an argument list. To give the class a name, pass it as the
-    `name` keyword argument. Otherwise, a name will be chosen for you.
-
-    The following are all equivalent:
-
-        enum('pinkie ring middle index thumb')
-        enum('pinkie', 'ring', 'middle', 'index', 'thumb')
-        enum(['pinkie', 'ring', 'middle', 'index', 'thumb'])
-    """
-
-    if ('name' not in kwargs) or kwargs['name'] is None:
-        # Create a probably-unique name. It doesn't really have to be
-        # unique, but getting distinct names each time helps with
-        # identification in debugging.
-        name = 'Enumeration' + hex(random.randint(0,0xfffffff))[2:].upper()
-    else:
-        name = kwargs['name']
-
-    if len(values) == 1:
-        # If there's only one value, we have a couple of alternate calling
-        # styles.
-        if isinstance(values[0], basestring) or hasattr(values[0], '__iter__'):
-            values = values[0]
-
-    return type(name, (Enumerated,), {'values': values})
+    def __lt__(self, other):
+        if self.__class__ is other.__class__:
+            return self.value < other.value
+        return NotImplemented

lib/beets/util/functemplate.py

@@ -42,6 +42,7 @@ ESCAPE_CHAR = u'$'
 VARIABLE_PREFIX = '__var_'
 FUNCTION_PREFIX = '__func_'
 
+
 class Environment(object):
     """Contains the values and functions to be substituted into a
     template.
@@ -57,10 +58,12 @@ def ex_lvalue(name):
     """A variable load expression."""
     return ast.Name(name, ast.Store())
 
+
 def ex_rvalue(name):
     """A variable store expression."""
     return ast.Name(name, ast.Load())
 
+
 def ex_literal(val):
     """An int, float, long, bool, string, or None literal with the given
     value.
@@ -75,6 +78,7 @@ def ex_literal(val):
         return ast.Str(val)
     raise TypeError('no literal for {0}'.format(type(val)))
 
+
 def ex_varassign(name, expr):
     """Assign an expression into a single variable. The expression may
     either be an `ast.expr` object or a value to be used as a literal.
@@ -83,6 +87,7 @@ def ex_varassign(name, expr):
         expr = ex_literal(expr)
     return ast.Assign([ex_lvalue(name)], expr)
 
+
 def ex_call(func, args):
     """A function-call expression with only positional parameters. The
     function may be an expression or the name of a function. Each
@@ -98,6 +103,7 @@ def ex_call(func, args):
 
     return ast.Call(func, args, [], None, None)
 
+
 def compile_func(arg_names, statements, name='_the_func', debug=False):
     """Compile a list of statements as the body of a function and return
     the resulting Python function. If `debug`, then print out the
@@ -157,6 +163,7 @@ class Symbol(object):
         expr = ex_rvalue(VARIABLE_PREFIX + self.ident.encode('utf8'))
         return [expr], set([self.ident.encode('utf8')]), set()
 
+
 class Call(object):
     """A function call in a template."""
     def __init__(self, ident, args, original):
@@ -214,6 +221,7 @@ class Call(object):
         )
         return [subexpr_call], varnames, funcnames
 
+
 class Expression(object):
     """Top-level template construct: contains a list of text blobs,
     Symbols, and Calls.
@@ -259,6 +267,7 @@ class Expression(object):
 class ParseError(Exception):
     pass
 
+
 class Parser(object):
     """Parses a template expression string. Instantiate the class with
     the template source and call ``parse_expression``. The ``pos`` field
@@ -316,13 +325,13 @@ class Parser(object):
 
             next_char = self.string[self.pos + 1]
             if char == ESCAPE_CHAR and next_char in \
-                  (SYMBOL_DELIM, FUNC_DELIM, GROUP_CLOSE, ARG_SEP):
+                    (SYMBOL_DELIM, FUNC_DELIM, GROUP_CLOSE, ARG_SEP):
                 # An escaped special character ($$, $}, etc.). Note that
                 # ${ is not an escape sequence: this is ambiguous with
                 # the start of a symbol and it's not necessary (just
                 # using { suffices in all cases).
                 text_parts.append(next_char)
-                self.pos += 2 # Skip the next character.
+                self.pos += 2  # Skip the next character.
                 continue
 
             # Shift all characters collected so far into a single string.
@@ -372,7 +381,7 @@ class Parser(object):
 
         if next_char == GROUP_OPEN:
             # A symbol like ${this}.
-            self.pos += 1 # Skip opening.
+            self.pos += 1  # Skip opening.
             closer = self.string.find(GROUP_CLOSE, self.pos)
             if closer == -1 or closer == self.pos:
                 # No closing brace found or identifier is empty.
@@ -431,7 +440,7 @@ class Parser(object):
             self.parts.append(self.string[start_pos:self.pos])
             return
 
-        self.pos += 1 # Move past closing brace.
+        self.pos += 1  # Move past closing brace.
         self.parts.append(Call(ident, args, self.string[start_pos:self.pos]))
 
     def parse_argument_list(self):
@@ -472,6 +481,7 @@ class Parser(object):
         self.pos += len(ident)
         return ident
 
+
 def _parse(template):
     """Parse a top-level template string Expression. Any extraneous text
     is considered literal text.

lib/beets/util/pipeline.py

@@ -35,13 +35,13 @@ from __future__ import print_function
 import Queue
 from threading import Thread, Lock
 import sys
-import types
 
 BUBBLE = '__PIPELINE_BUBBLE__'
 POISON = '__PIPELINE_POISON__'
 
 DEFAULT_QUEUE_SIZE = 16
 
+
 def _invalidate_queue(q, val=None, sync=True):
     """Breaks a Queue such that it never blocks, always has size 1,
     and has no maximum size. get()ing from the queue returns `val`,
@@ -50,8 +50,10 @@ def _invalidate_queue(q, val=None, sync=True):
     """
     def _qsize(len=len):
         return 1
+
     def _put(item):
         pass
+
     def _get():
         return val
 
@@ -70,6 +72,7 @@ def _invalidate_queue(q, val=None, sync=True):
         if sync:
             q.mutex.release()
 
+
 class CountedQueue(Queue.Queue):
     """A queue that keeps track of the number of threads that are
     still feeding into it. The queue is poisoned when all threads are
@@ -104,6 +107,7 @@ class CountedQueue(Queue.Queue):
 
                 # Replacement _get invalidates when no items remain.
                 _old_get = self._get
+
                 def _get():
                     out = _old_get()
                     if not self.queue:
@@ -117,18 +121,67 @@ class CountedQueue(Queue.Queue):
                     # No items. Invalidate immediately.
                     _invalidate_queue(self, POISON, False)
 
+
 class MultiMessage(object):
     """A message yielded by a pipeline stage encapsulating multiple
     values to be sent to the next stage.
     """
     def __init__(self, messages):
         self.messages = messages
+
+
 def multiple(messages):
     """Yield multiple([message, ..]) from a pipeline stage to send
     multiple values to the next pipeline stage.
     """
     return MultiMessage(messages)
 
+
+def stage(func):
+    """Decorate a function to become a simple stage.
+
+    >>> @stage
+    ... def add(n, i):
+    ...     return i + n
+    >>> pipe = Pipeline([
+    ...     iter([1, 2, 3]),
+    ...     add(2),
+    ... ])
+    >>> list(pipe.pull())
+    [3, 4, 5]
+    """
+
+    def coro(*args):
+        task = None
+        while True:
+            task = yield task
+            task = func(*(args + (task,)))
+    return coro
+
+
+def mutator_stage(func):
+    """Decorate a function that manipulates items in a coroutine to
+    become a simple stage.
+
+    >>> @mutator_stage
+    ... def setkey(key, item):
+    ...     item[key] = True
+    >>> pipe = Pipeline([
+    ...     iter([{'x': False}, {'a': False}]),
+    ...     setkey('x'),
+    ... ])
+    >>> list(pipe.pull())
+    [{'x': True}, {'a': False, 'x': True}]
+    """
+
+    def coro(*args):
+        task = None
+        while True:
+            task = yield task
+            func(*(args + (task,)))
+    return coro
+
+
 def _allmsgs(obj):
     """Returns a list of all the messages encapsulated in obj. If obj
     is a MultiMessage, returns its enclosed messages. If obj is BUBBLE,
@@ -141,6 +194,7 @@ def _allmsgs(obj):
     else:
         return [obj]
 
+
 class PipelineThread(Thread):
     """Abstract base class for pipeline-stage threads."""
     def __init__(self, all_threads):
@@ -169,6 +223,7 @@ class PipelineThread(Thread):
         for thread in self.all_threads:
             thread.abort()
 
+
 class FirstPipelineThread(PipelineThread):
     """The thread running the first stage in a parallel pipeline setup.
     The coroutine should just be a generator.
@@ -209,6 +264,7 @@ class FirstPipelineThread(PipelineThread):
         # Generator finished; shut down the pipeline.
         self.out_queue.release()
 
+
 class MiddlePipelineThread(PipelineThread):
     """A thread running any stage in the pipeline except the first or
     last.
@@ -256,6 +312,7 @@ class MiddlePipelineThread(PipelineThread):
         # Pipeline is shutting down normally.
         self.out_queue.release()
 
+
 class LastPipelineThread(PipelineThread):
     """A thread running the last stage in a pipeline. The coroutine
     should yield nothing.
@@ -291,6 +348,7 @@ class LastPipelineThread(PipelineThread):
             self.abort_all(sys.exc_info())
             return
 
+
 class Pipeline(object):
     """Represents a staged pattern of work. Each stage in the pipeline
     is a coroutine that receives messages from the previous stage and
@@ -322,7 +380,8 @@ class Pipeline(object):
         messages between the stages are stored in queues of the given
         size.
         """
-        queues = [CountedQueue(queue_size) for i in range(len(self.stages)-1)]
+        queue_count = len(self.stages) - 1
+        queues = [CountedQueue(queue_size) for i in range(queue_count)]
         threads = []
 
         # Set up first stage.
@@ -330,10 +389,10 @@ class Pipeline(object):
             threads.append(FirstPipelineThread(coro, queues[0], threads))
 
         # Middle stages.
-        for i in range(1, len(self.stages)-1):
+        for i in range(1, queue_count):
             for coro in self.stages[i]:
                 threads.append(MiddlePipelineThread(
-                    coro, queues[i-1], queues[i], threads
+                    coro, queues[i - 1], queues[i], threads
                 ))
 
         # Last stage.
@@ -408,17 +467,20 @@ if __name__ == '__main__':
             print('generating %i' % i)
             time.sleep(1)
             yield i
+
     def work():
         num = yield
         while True:
             print('processing %i' % num)
             time.sleep(2)
-            num = yield num*2
+            num = yield num * 2
+
     def consume():
         while True:
             num = yield
             time.sleep(1)
             print('received %i' % num)
+
     ts_start = time.time()
     Pipeline([produce(), work(), consume()]).run_sequential()
     ts_seq = time.time()
@@ -437,6 +499,7 @@ if __name__ == '__main__':
             print('generating %i' % i)
             time.sleep(1)
             yield i
+
     def exc_work():
         num = yield
         while True:
@@ -445,10 +508,10 @@ if __name__ == '__main__':
             if num == 3:
                 raise Exception()
             num = yield num * 2
+
     def exc_consume():
         while True:
             num = yield
-            #if num == 4:
-            #   raise Exception()
             print('received %i' % num)
+
     Pipeline([exc_produce(), exc_work(), exc_consume()]).run_parallel(1)

lib/beets/vfs.py

@@ -20,6 +20,7 @@ from beets import util
 
 Node = namedtuple('Node', ['files', 'dirs'])
 
+
 def _insert(node, path, itemid):
     """Insert an item into a virtual filesystem node."""
     if len(path) == 1:
@@ -33,6 +34,7 @@ def _insert(node, path, itemid):
             node.dirs[dirname] = Node({}, {})
         _insert(node.dirs[dirname], rest, itemid)
 
+
 def libtree(lib):
     """Generates a filesystem-like directory tree for the files
     contained in `lib`. Filesystem nodes are (files, dirs) named

lib/beetsplug/embedart.py

@@ -16,6 +16,9 @@
 import os.path
 import logging
 import imghdr
+import subprocess
+import platform
+from tempfile import NamedTemporaryFile
 
 from beets.plugins import BeetsPlugin
 from beets import mediafile
@@ -25,6 +28,7 @@ from beets.util import syspath, normpath, displayable_path
 from beets.util.artresizer import ArtResizer
 from beets import config
 
+
 log = logging.getLogger('beets')
 
 
@@ -36,12 +40,19 @@ class EmbedCoverArtPlugin(BeetsPlugin):
         self.config.add({
             'maxwidth': 0,
             'auto': True,
+            'compare_threshold': 0,
+            'ifempty': False,
         })
-        if self.config['maxwidth'].get(int) and \
-                not ArtResizer.shared.local:
+
+        if self.config['maxwidth'].get(int) and not ArtResizer.shared.local:
             self.config['maxwidth'] = 0
             log.warn(u"embedart: ImageMagick or PIL not found; "
                      u"'maxwidth' option ignored")
+        if self.config['compare_threshold'].get(int) and not \
+                ArtResizer.shared.can_compare:
+            self.config['compare_threshold'] = 0
+            log.warn(u"embedart: ImageMagick 6.8.7 or higher not installed; "
+                     u"'compare_threshold' option ignored")
 
     def commands(self):
         # Embed command.
@@ -52,12 +63,15 @@ class EmbedCoverArtPlugin(BeetsPlugin):
             '-f', '--file', metavar='PATH', help='the image file to embed'
         )
         maxwidth = config['embedart']['maxwidth'].get(int)
+        compare_threshold = config['embedart']['compare_threshold'].get(int)
+        ifempty = config['embedart']['ifempty'].get(bool)
 
         def embed_func(lib, opts, args):
             if opts.file:
                 imagepath = normpath(opts.file)
                 for item in lib.items(decargs(args)):
-                    embed_item(item, imagepath, maxwidth)
+                    embed_item(item, imagepath, maxwidth, None,
+                               compare_threshold, ifempty)
             else:
                 for album in lib.albums(decargs(args)):
                     embed_album(album, maxwidth)
@@ -72,7 +86,8 @@ class EmbedCoverArtPlugin(BeetsPlugin):
 
         def extract_func(lib, opts, args):
             outpath = normpath(opts.outpath or 'cover')
-            extract(lib, outpath, decargs(args))
+            item = lib.items(decargs(args)).get()
+            extract(outpath, item)
         extract_cmd.func = extract_func
 
         # Clear command.
@@ -91,23 +106,43 @@ def album_imported(lib, album):
     """Automatically embed art into imported albums.
     """
     if album.artpath and config['embedart']['auto']:
-        embed_album(album, config['embedart']['maxwidth'].get(int))
+        embed_album(album, config['embedart']['maxwidth'].get(int), True)
 
 
-def embed_item(item, imagepath, maxwidth=None, itempath=None):
+def embed_item(item, imagepath, maxwidth=None, itempath=None,
+               compare_threshold=0, ifempty=False, as_album=False):
     """Embed an image into the item's media file.
     """
+    if compare_threshold:
+        if not check_art_similarity(item, imagepath, compare_threshold):
+            log.warn(u'Image not similar; skipping.')
+            return
+    if ifempty:
+        art = get_art(item)
+        if not art:
+            pass
+        else:
+            log.debug(u'embedart: media file contained art already {0}'.format(
+                displayable_path(imagepath)
+            ))
+            return
+    if maxwidth and not as_album:
+        imagepath = resize_image(imagepath, maxwidth)
+
     try:
+        log.debug(u'embedart: embedding {0}'.format(
+            displayable_path(imagepath)
+        ))
         item['images'] = [_mediafile_image(imagepath, maxwidth)]
-        item.try_write(itempath)
     except IOError as exc:
         log.error(u'embedart: could not read image file: {0}'.format(exc))
-    finally:
-        # We don't want to store the image in the database
+    else:
+        # We don't want to store the image in the database.
+        item.try_write(itempath)
         del item['images']
 
 
-def embed_album(album, maxwidth=None):
+def embed_album(album, maxwidth=None, quiet=False):
     """Embed album art into all of the album's items.
     """
     imagepath = album.artpath
@@ -115,39 +150,78 @@ def embed_album(album, maxwidth=None):
         log.info(u'No album art present: {0} - {1}'.
                  format(album.albumartist, album.album))
         return
-    if not os.path.isfile(imagepath):
+    if not os.path.isfile(syspath(imagepath)):
         log.error(u'Album art not found at {0}'
-                  .format(imagepath))
+                  .format(displayable_path(imagepath)))
         return
+    if maxwidth:
+        imagepath = resize_image(imagepath, maxwidth)
 
-    log.info(u'Embedding album art into {0.albumartist} - {0.album}.'
-             .format(album))
+    log.log(
+        logging.DEBUG if quiet else logging.INFO,
+        u'Embedding album art into {0.albumartist} - {0.album}.'.format(album),
+    )
 
     for item in album.items():
-        embed_item(item, imagepath, maxwidth)
+        embed_item(item, imagepath, maxwidth, None,
+                   config['embedart']['compare_threshold'].get(int),
+                   config['embedart']['ifempty'].get(bool), as_album=True)
+
+
+def resize_image(imagepath, maxwidth):
+    """Returns path to an image resized to maxwidth.
+    """
+    log.info(u'Resizing album art to {0} pixels wide'
+             .format(maxwidth))
+    imagepath = ArtResizer.shared.resize(maxwidth, syspath(imagepath))
+    return imagepath
+
+
+def check_art_similarity(item, imagepath, compare_threshold):
+    """A boolean indicating if an image is similar to embedded item art.
+    """
+    with NamedTemporaryFile(delete=True) as f:
+        art = extract(f.name, item)
+
+        if art:
+            # Converting images to grayscale tends to minimize the weight
+            # of colors in the diff score
+            cmd = 'convert {0} {1} -colorspace gray MIFF:- | ' \
+                  'compare -metric PHASH - null:'.format(syspath(imagepath),
+                                                         syspath(art))
+
+            proc = subprocess.Popen(cmd, stdout=subprocess.PIPE,
+                                    stderr=subprocess.PIPE,
+                                    close_fds=platform.system() != 'Windows',
+                                    shell=True)
+            stdout, stderr = proc.communicate()
+            if proc.returncode:
+                if proc.returncode != 1:
+                    log.warn(u'embedart: IM phashes compare failed for {0}, \
+                   {1}'.format(displayable_path(imagepath),
+                               displayable_path(art)))
+                    return
+                phashDiff = float(stderr)
+            else:
+                phashDiff = float(stdout)
+
+            log.info(u'embedart: compare PHASH score is {0}'.format(phashDiff))
+            if phashDiff > compare_threshold:
+                return False
+
+    return True
 
 
 def _mediafile_image(image_path, maxwidth=None):
     """Return a `mediafile.Image` object for the path.
-
-    If maxwidth is set the image is resized if necessary.
     """
-    if maxwidth:
-        image_path = ArtResizer.shared.resize(maxwidth, syspath(image_path))
 
     with open(syspath(image_path), 'rb') as f:
         data = f.read()
     return mediafile.Image(data, type=mediafile.ImageType.front)
 
 
-# 'extractart' command.
-
-def extract(lib, outpath, query):
-    item = lib.items(query).get()
-    if not item:
-        log.error(u'No item matches query.')
-        return
-
+def get_art(item):
     # Extract the art.
     try:
         mf = mediafile.MediaFile(syspath(item.path))
@@ -157,7 +231,18 @@ def extract(lib, outpath, query):
         ))
         return
 
-    art = mf.art
+    return mf.art
+
+# 'extractart' command.
+
+
+def extract(outpath, item):
+    if not item:
+        log.error(u'No item matches query.')
+        return
+
+    art = get_art(item)
+
     if not art:
         log.error(u'No album art present in {0} - {1}.'
                   .format(item.artist, item.title))
@@ -170,10 +255,11 @@ def extract(lib, outpath, query):
         return
     outpath += '.' + ext
 
-    log.info(u'Extracting album art from: {0.artist} - {0.title}\n'
-             u'To: {1}'.format(item, displayable_path(outpath)))
+    log.info(u'Extracting album art from: {0.artist} - {0.title} '
+             u'to: {1}'.format(item, displayable_path(outpath)))
     with open(syspath(outpath), 'wb') as f:
         f.write(art)
+    return outpath
 
 
 # 'clearart' command.
@@ -190,5 +276,5 @@ def clear(lib, query):
                 displayable_path(item.path), exc
             ))
             continue
-        mf.art = None
+        del mf.art
         mf.save()

lib/beetsplug/fetchart.py

@@ -1,5 +1,5 @@
 # This file is part of beets.
-# Copyright 2013, Adrian Sampson.
+# Copyright 2014, Adrian Sampson.
 #
 # Permission is hereby granted, free of charge, to any person obtaining
 # a copy of this software and associated documentation files (the
@@ -22,12 +22,18 @@ from tempfile import NamedTemporaryFile
 
 import requests
 
-from beets.plugins import BeetsPlugin
-from beets.util.artresizer import ArtResizer
+from beets import plugins
 from beets import importer
 from beets import ui
 from beets import util
 from beets import config
+from beets.util.artresizer import ArtResizer
+
+try:
+    import itunes
+    HAVE_ITUNES = True
+except ImportError:
+    HAVE_ITUNES = False
 
 IMAGE_EXTENSIONS = ['png', 'jpg', 'jpeg']
 CONTENT_TYPES = ('image/jpeg',)
@@ -73,17 +79,14 @@ CAA_URL = 'http://coverartarchive.org/release/{mbid}/front-500.jpg'
 CAA_GROUP_URL = 'http://coverartarchive.org/release-group/{mbid}/front-500.jpg'
 
 
-def caa_art(release_id):
-    """Return the Cover Art Archive URL given a MusicBrainz release ID.
+def caa_art(album):
+    """Return the Cover Art Archive and Cover Art Archive release group URLs
+    using album MusicBrainz release ID and release group ID.
     """
-    return CAA_URL.format(mbid=release_id)
-
-
-def caa_group_art(release_group_id):
-    """Return the Cover Art Archive release group URL given a MusicBrainz
-    release group ID.
-    """
-    return CAA_GROUP_URL.format(mbid=release_group_id)
+    if album.mb_albumid:
+        yield CAA_URL.format(mbid=album.mb_albumid)
+    if album.mb_releasegroupid:
+        yield CAA_GROUP_URL.format(mbid=album.mb_releasegroupid)
 
 
 # Art from Amazon.
@@ -92,10 +95,12 @@ AMAZON_URL = 'http://images.amazon.com/images/P/%s.%02i.LZZZZZZZ.jpg'
 AMAZON_INDICES = (1, 2)
 
 
-def art_for_asin(asin):
-    """Generate URLs for an Amazon ID (ASIN) string."""
-    for index in AMAZON_INDICES:
-        yield AMAZON_URL % (asin, index)
+def art_for_asin(album):
+    """Generate URLs using Amazon ID (ASIN) string.
+    """
+    if album.asin:
+        for index in AMAZON_INDICES:
+            yield AMAZON_URL % (album.asin, index)
 
 
 # AlbumArt.org scraper.
@@ -104,11 +109,14 @@ AAO_URL = 'http://www.albumart.org/index_detail.php'
 AAO_PAT = r'href\s*=\s*"([^>"]*)"[^>]*title\s*=\s*"View larger image"'
 
 
-def aao_art(asin):
-    """Return art URL from AlbumArt.org given an ASIN."""
+def aao_art(album):
+    """Return art URL from AlbumArt.org using album ASIN.
+    """
+    if not album.asin:
+        return
     # Get the page from albumart.org.
     try:
-        resp = requests_session.get(AAO_URL, params={'asin': asin})
+        resp = requests_session.get(AAO_URL, params={'asin': album.asin})
         log.debug(u'fetchart: scraped art URL: {0}'.format(resp.url))
     except requests.RequestException:
         log.debug(u'fetchart: error scraping art page')
@@ -118,7 +126,7 @@ def aao_art(asin):
     m = re.search(AAO_PAT, resp.text)
     if m:
         image_url = m.group(1)
-        return image_url
+        yield image_url
     else:
         log.debug(u'fetchart: no image found on page')
 
@@ -132,6 +140,8 @@ def google_art(album):
     """Return art URL from google.org given an album title and
     interpreter.
     """
+    if not (album.albumartist and album.album):
+        return
     search_string = (album.albumartist + ',' + album.album).encode('utf-8')
     response = requests_session.get(GOOGLE_URL, params={
         'v': '1.0',
@@ -145,14 +155,39 @@ def google_art(album):
         data = results['responseData']
         dataInfo = data['results']
         for myUrl in dataInfo:
-            return myUrl['unescapedUrl']
+            yield myUrl['unescapedUrl']
     except:
         log.debug(u'fetchart: error scraping art page')
         return
 
 
+# Art from the iTunes Store.
+
+def itunes_art(album):
+    """Return art URL from iTunes Store given an album title.
+    """
+    search_string = (album.albumartist + ' ' + album.album).encode('utf-8')
+    try:
+        # Isolate bugs in the iTunes library while searching.
+        try:
+            itunes_album = itunes.search_album(search_string)[0]
+        except Exception as exc:
+            log.debug('fetchart: iTunes search failed: {0}'.format(exc))
+            return
+
+        if itunes_album.get_artwork()['100']:
+            small_url = itunes_album.get_artwork()['100']
+            big_url = small_url.replace('100x100', '1200x1200')
+            yield big_url
+        else:
+            log.debug(u'fetchart: album has no artwork in iTunes Store')
+    except IndexError:
+        log.debug(u'fetchart: album not found in iTunes Store')
+
+
 # Art from the filesystem.
 
+
 def filename_priority(filename, cover_names):
     """Sort order for image names.
 
@@ -164,7 +199,8 @@ def filename_priority(filename, cover_names):
 
 
 def art_in_path(path, cover_names, cautious):
-    """Look for album art files in a specified directory."""
+    """Look for album art files in a specified directory.
+    """
     if not os.path.isdir(path):
         return
 
@@ -195,31 +231,27 @@ def art_in_path(path, cover_names, cautious):
 
 # Try each source in turn.
 
+SOURCES_ALL = [u'coverart', u'itunes', u'amazon', u'albumart', u'google']
 
-def _source_urls(album):
+ART_FUNCS = {
+    u'coverart': caa_art,
+    u'itunes': itunes_art,
+    u'albumart': aao_art,
+    u'amazon': art_for_asin,
+    u'google': google_art,
+}
+
+
+def _source_urls(album, sources=SOURCES_ALL):
     """Generate possible source URLs for an album's art. The URLs are
     not guaranteed to work so they each need to be attempted in turn.
     This allows the main `art_for_album` function to abort iteration
     through this sequence early to avoid the cost of scraping when not
     necessary.
     """
-    # Cover Art Archive.
-    if album.mb_albumid:
-        yield caa_art(album.mb_albumid)
-    if album.mb_releasegroupid:
-        yield caa_group_art(album.mb_releasegroupid)
-
-    # Amazon and AlbumArt.org.
-    if album.asin:
-        for url in art_for_asin(album.asin):
-            yield url
-        url = aao_art(album.asin)
-        if url:
-            yield url
-
-    if config['fetchart']['google_search']:
-        url = google_art(album)
-        if url:
+    for s in sources:
+        urls = ART_FUNCS[s](album)
+        for url in urls:
             yield url
 
 
@@ -245,7 +277,8 @@ def art_for_album(album, paths, maxwidth=None, local_only=False):
     # Web art sources.
     remote_priority = config['fetchart']['remote_priority'].get(bool)
     if not local_only and (remote_priority or not out):
-        for url in _source_urls(album):
+        for url in _source_urls(album,
+                                config['fetchart']['sources'].as_str_seq()):
             if maxwidth:
                 url = ArtResizer.shared.proxy_url(maxwidth, url)
             candidate = _fetch_image(url)
@@ -286,7 +319,7 @@ def batch_fetch_art(lib, albums, force, maxwidth=None):
                                           message))
 
 
-class FetchArtPlugin(BeetsPlugin):
+class FetchArtPlugin(plugins.BeetsPlugin):
     def __init__(self):
         super(FetchArtPlugin, self).__init__()
 
@@ -297,6 +330,7 @@ class FetchArtPlugin(BeetsPlugin):
             'cautious': False,
             'google_search': False,
             'cover_names': ['cover', 'front', 'art', 'album', 'folder'],
+            'sources': SOURCES_ALL,
         })
 
         # Holds paths to downloaded images between fetching them and
@@ -309,6 +343,12 @@ class FetchArtPlugin(BeetsPlugin):
             self.import_stages = [self.fetch_art]
             self.register_listener('import_task_files', self.assign_art)
 
+        available_sources = list(SOURCES_ALL)
+        if not HAVE_ITUNES and u'itunes' in available_sources:
+            available_sources.remove(u'itunes')
+        self.config['sources'] = plugins.sanitize_choices(
+            self.config['sources'].as_str_seq(), available_sources)
+
     # Asynchronous; after music is added to the library.
     def fetch_art(self, session, task):
         """Find art for the album being imported."""

lib/beetsplug/lyrics.py

@@ -18,25 +18,26 @@ from __future__ import print_function
 
 import re
 import logging
-import urllib
+import requests
 import json
 import unicodedata
+import urllib
 import difflib
 import itertools
+from HTMLParser import HTMLParseError
 
-from beets.plugins import BeetsPlugin
-from beets import ui
-from beets import config
+from beets import plugins
+from beets import config, ui
 
 
 # Global logger.
 
 log = logging.getLogger('beets')
 
-DIV_RE = re.compile(r'<(/?)div>?')
+DIV_RE = re.compile(r'<(/?)div>?', re.I)
 COMMENT_RE = re.compile(r'<!--.*-->', re.S)
 TAG_RE = re.compile(r'<[^>]*>')
-BREAK_RE = re.compile(r'<br\s*/?>')
+BREAK_RE = re.compile(r'\n?\s*<br([\s|/][^>]*)*>\s*\n?', re.I)
 URL_CHARACTERS = {
     u'\u2018': u"'",
     u'\u2019': u"'",
@@ -60,10 +61,14 @@ def fetch_url(url):
     is unreachable.
     """
     try:
-        return urllib.urlopen(url).read()
-    except IOError as exc:
-        log.debug(u'failed to fetch: {0} ({1})'.format(url, unicode(exc)))
-        return None
+        r = requests.get(url, verify=False)
+    except requests.RequestException as exc:
+        log.debug(u'lyrics request failed: {0}'.format(exc))
+        return
+    if r.status_code == requests.codes.ok:
+        return r.text
+    else:
+        log.debug(u'failed to fetch: {0} ({1})'.format(url, r.status_code))
 
 
 def unescape(text):
@@ -79,10 +84,20 @@ def unescape(text):
     return out
 
 
-def extract_text(html, starttag):
+def extract_text_between(html, start_marker, end_marker):
+    try:
+        _, html = html.split(start_marker, 1)
+        html, _ = html.split(end_marker, 1)
+    except ValueError:
+        return u''
+    return html
+
+
+def extract_text_in(html, starttag):
     """Extract the text from a <DIV> tag in the HTML starting with
     ``starttag``. Returns None if parsing fails.
     """
+
     # Strip off the leading text before opening tag.
     try:
         _, html = html.split(starttag, 1)
@@ -101,7 +116,6 @@ def extract_text(html, starttag):
         else:  # Opening tag.
             if level == 0:
                 parts.append(html[pos:match.start()])
-
             level += 1
 
         if level == -1:
@@ -110,26 +124,7 @@ def extract_text(html, starttag):
     else:
         print('no closing tag found!')
         return
-    lyrics = ''.join(parts)
-    return strip_cruft(lyrics)
-
-
-def strip_cruft(lyrics, wscollapse=True):
-    """Clean up HTML from an extracted lyrics string. For example, <BR>
-    tags are replaced with newlines.
-    """
-    lyrics = COMMENT_RE.sub('', lyrics)
-    lyrics = unescape(lyrics)
-    if wscollapse:
-        lyrics = re.sub(r'\s+', ' ', lyrics)  # Whitespace collapse.
-    lyrics = re.sub(r'<(script).*?</\1>(?s)', '', lyrics)  # Strip script tags.
-    lyrics = BREAK_RE.sub('\n', lyrics)  # <BR> newlines.
-    lyrics = re.sub(r'\n +', '\n', lyrics)
-    lyrics = re.sub(r' +\n', '\n', lyrics)
-    lyrics = TAG_RE.sub('', lyrics)  # Strip remaining HTML tags.
-    lyrics = lyrics.replace('\r', '\n')
-    lyrics = lyrics.strip()
-    return lyrics
+    return u''.join(parts)
 
 
 def search_pairs(item):
@@ -140,7 +135,7 @@ def search_pairs(item):
 
     In addition to the artist and title obtained from the `item` the
     method tries to strip extra information like paranthesized suffixes
-    and featured artists from the strings and add them as caniddates.
+    and featured artists from the strings and add them as candidates.
     The method also tries to split multiple titles separated with `/`.
     """
 
@@ -149,7 +144,7 @@ def search_pairs(item):
     artists = [artist]
 
     # Remove any featuring artists from the artists name
-    pattern = r"(.*?) (&|\b(and|ft|feat(uring)?\b))"
+    pattern = r"(.*?) {0}".format(plugins.feat_tokens())
     match = re.search(pattern, artist, re.IGNORECASE)
     if match:
         artists.append(match.group(1))
@@ -162,8 +157,8 @@ def search_pairs(item):
         titles.append(match.group(1))
 
     # Remove any featuring artists from the title
-    pattern = r"(.*?) \b(ft|feat(uring)?)\b"
-    for title in titles:
+    pattern = r"(.*?) {0}".format(plugins.feat_tokens(for_artist=False))
+    for title in titles[:]:
         match = re.search(pattern, title, re.IGNORECASE)
         if match:
             titles.append(match.group(1))
@@ -189,6 +184,19 @@ def _encode(s):
         s = s.encode('utf8', 'ignore')
     return urllib.quote(s)
 
+# Musixmatch
+
+MUSIXMATCH_URL_PATTERN = 'https://www.musixmatch.com/lyrics/%s/%s'
+
+
+def fetch_musixmatch(artist, title):
+    url = MUSIXMATCH_URL_PATTERN % (_lw_encode(artist.title()),
+                                    _lw_encode(title.title()))
+    html = fetch_url(url)
+    if not html:
+        return
+    lyrics = extract_text_between(html, '"lyrics_body":', '"lyrics_language":')
+    return lyrics.strip(',"').replace('\\n', '\n')
 
 # LyricsWiki.
 
@@ -212,7 +220,7 @@ def fetch_lyricswiki(artist, title):
     if not html:
         return
 
-    lyrics = extract_text(html, "<div class='lyricbox'>")
+    lyrics = extract_text_in(html, u"<div class='lyricbox'>")
     if lyrics and 'Unfortunately, we are not licensed' not in lyrics:
         return lyrics
 
@@ -238,8 +246,8 @@ def fetch_lyricscom(artist, title):
     html = fetch_url(url)
     if not html:
         return
-
-    lyrics = extract_text(html, '<div id="lyric_space">')
+    lyrics = extract_text_between(html, '<div id="lyrics" class="SCREENONLY" '
+                                  'itemprop="description">', '</div>')
     if not lyrics:
         return
     for not_found_str in LYRICSCOM_NOT_FOUND:
@@ -280,7 +288,6 @@ def is_page_candidate(urlLink, urlTitle, title, artist):
     artist = slugify(artist.lower())
     sitename = re.search(u"//([^/]+)/.*", slugify(urlLink.lower())).group(1)
     urlTitle = slugify(urlTitle.lower())
-
     # Check if URL title contains song title (exact match)
     if urlTitle.find(title) != -1:
         return True
@@ -289,41 +296,11 @@ def is_page_candidate(urlLink, urlTitle, title, artist):
     tokens = [by + '_' + artist for by in BY_TRANS] + \
              [artist, sitename, sitename.replace('www.', '')] + LYRICS_TRANS
     songTitle = re.sub(u'(%s)' % u'|'.join(tokens), u'', urlTitle)
-
-    typoRatio = .8
+    songTitle = songTitle.strip('_|')
+    typoRatio = .9
     return difflib.SequenceMatcher(None, songTitle, title).ratio() >= typoRatio
 
 
-def insert_line_feeds(text):
-    """Insert newlines before upper-case characters.
-    """
-    tokensStr = re.split("([a-z][A-Z])", text)
-    for idx in range(1, len(tokensStr), 2):
-        ltoken = list(tokensStr[idx])
-        tokensStr[idx] = ltoken[0] + '\n' + ltoken[1]
-    return ''.join(tokensStr)
-
-
-def sanitize_lyrics(text):
-    """Clean text, returning raw lyrics as output or None if it happens
-    that input text is actually not lyrics content.  Clean (x)html tags
-    in text, correct layout and syntax...
-    """
-    text = strip_cruft(text, False)
-
-    # Restore \n in input text
-    if '\n' not in text:
-        text = insert_line_feeds(text)
-
-    while text.count('\n\n') > text.count('\n') // 4:
-        # Remove first occurrence of \n for each sequence of \n
-        text = re.sub(r'\n(\n+)', '\g<1>', text)
-
-    text = re.sub(r'\n\n+', '\n\n', text)   # keep at most two \n in a row
-
-    return text
-
-
 def remove_credits(text):
     """Remove first/last line of text if it contains the word 'lyrics'
     eg 'Lyrics by songsdatabase.com'
@@ -342,13 +319,12 @@ def is_lyrics(text, artist=None):
     """Determine whether the text seems to be valid lyrics.
     """
     if not text:
-        return
-
+        return False
     badTriggersOcc = []
     nbLines = text.count('\n')
     if nbLines <= 1:
         log.debug(u"Ignoring too short lyrics '{0}'".format(text))
-        return 0
+        return False
     elif nbLines < 5:
         badTriggersOcc.append('too_short')
     else:
@@ -356,7 +332,7 @@ def is_lyrics(text, artist=None):
         # down
         text = remove_credits(text)
 
-    badTriggers = ['lyrics', 'copyright', 'property']
+    badTriggers = ['lyrics', 'copyright', 'property', 'links']
     if artist:
         badTriggersOcc += [artist]
 
@@ -366,62 +342,58 @@ def is_lyrics(text, artist=None):
 
     if badTriggersOcc:
         log.debug(u'Bad triggers detected: {0}'.format(badTriggersOcc))
-
     return len(badTriggersOcc) < 2
 
 
-def scrape_lyrics_from_url(url):
+def _scrape_strip_cruft(html, plain_text_out=False):
+    """Clean up HTML
+    """
+    html = unescape(html)
+
+    html = html.replace('\r', '\n')  # Normalize EOL.
+    html = re.sub(r' +', ' ', html)  # Whitespaces collapse.
+    html = BREAK_RE.sub('\n', html)  # <br> eats up surrounding '\n'.
+    html = re.sub(r'<(script).*?</\1>(?s)', '', html)  # Strip script tags.
+
+    if plain_text_out:  # Strip remaining HTML tags
+        html = COMMENT_RE.sub('', html)
+        html = TAG_RE.sub('', html)
+
+    html = '\n'.join([x.strip() for x in html.strip().split('\n')])
+    html = re.sub(r'\n{3,}', r'\n\n', html)
+    return html
+
+
+def _scrape_merge_paragraphs(html):
+    html = re.sub(r'</p>\s*<p(\s*[^>]*)>', '\n', html)
+    return re.sub(r'<div .*>\s*</div>', '\n', html)
+
+
+def scrape_lyrics_from_html(html):
     """Scrape lyrics from a URL. If no lyrics can be found, return None
     instead.
     """
-    from bs4 import BeautifulSoup, Comment
-    html = fetch_url(url)
+    from bs4 import SoupStrainer, BeautifulSoup
+
     if not html:
         return None
 
-    soup = BeautifulSoup(html)
-
-    for tag in soup.findAll('br'):
-        tag.replaceWith('\n')
-
-    # Remove non relevant html parts
-    [s.extract() for s in soup(['head', 'script'])]
-    comments = soup.findAll(text=lambda text: isinstance(text, Comment))
-    [s.extract() for s in comments]
+    def is_text_notcode(text):
+        length = len(text)
+        return (length > 20 and
+                text.count(' ') > length / 25 and
+                (text.find('{') == -1 or text.find(';') == -1))
+    html = _scrape_strip_cruft(html)
+    html = _scrape_merge_paragraphs(html)
 
+    # extract all long text blocks that are not code
     try:
-        for tag in soup.findAll(True):
-            tag.name = 'p'          # keep tag contents
-
-    except Exception, e:
-        log.debug(u'Error {0} when replacing containing marker by p marker'
-                  .format(e, exc_info=True))
-
-    # Make better soup from current soup! The previous unclosed <p> sections
-    # are now closed.  Use str() rather than prettify() as it's more
-    # conservative concerning EOL
-    soup = BeautifulSoup(str(soup))
-
-    # In case lyrics are nested in no markup but <body>
-    # Insert the whole body in a <p>
-    bodyTag = soup.find('body')
-    if bodyTag:
-        pTag = soup.new_tag("p")
-        bodyTag.parent.insert(0, pTag)
-        pTag.insert(0, bodyTag)
-
-    tagTokens = []
-
-    for tag in soup.findAll('p'):
-        soup2 = BeautifulSoup(str(tag))
-        # Extract all text of <p> section.
-        tagTokens += soup2.findAll(text=True)
-
-    if tagTokens:
-        # Lyrics are expected to be the longest paragraph
-        tagTokens = sorted(tagTokens, key=len, reverse=True)
-        soup = BeautifulSoup(tagTokens[0])
-        return unescape(tagTokens[0].strip("\n\r: "))
+        soup = BeautifulSoup(html, "html.parser",
+                             parse_only=SoupStrainer(text=is_text_notcode))
+    except HTMLParseError:
+        return None
+    soup = sorted(soup.stripped_strings, key=len)[-1]
+    return soup
 
 
 def fetch_google(artist, title):
@@ -443,15 +415,14 @@ def fetch_google(artist, title):
     if 'items' in data.keys():
         for item in data['items']:
             urlLink = item['link']
-            urlTitle = item['title']
+            urlTitle = item.get('title', u'')
             if not is_page_candidate(urlLink, urlTitle, title, artist):
                 continue
-            lyrics = scrape_lyrics_from_url(urlLink)
+            html = fetch_url(urlLink)
+            lyrics = scrape_lyrics_from_html(html)
             if not lyrics:
                 continue
 
-            lyrics = sanitize_lyrics(lyrics)
-
             if is_lyrics(lyrics, artist):
                 log.debug(u'got lyrics from {0}'.format(item['displayLink']))
                 return lyrics
@@ -459,8 +430,16 @@ def fetch_google(artist, title):
 
 # Plugin logic.
 
+SOURCES = ['google', 'lyricwiki', 'lyrics.com', 'musixmatch']
+SOURCE_BACKENDS = {
+    'google': fetch_google,
+    'lyricwiki': fetch_lyricswiki,
+    'lyrics.com': fetch_lyricscom,
+    'musixmatch': fetch_musixmatch,
+}
 
-class LyricsPlugin(BeetsPlugin):
+
+class LyricsPlugin(plugins.BeetsPlugin):
     def __init__(self):
         super(LyricsPlugin, self).__init__()
         self.import_stages = [self.imported]
@@ -469,12 +448,19 @@ class LyricsPlugin(BeetsPlugin):
             'google_API_key': None,
             'google_engine_ID': u'009217259823014548361:lndtuqkycfu',
             'fallback': None,
+            'force': False,
+            'sources': SOURCES,
         })
 
-        self.backends = [fetch_lyricswiki, fetch_lyricscom]
-
-        if self.config['google_API_key'].get():
-            self.backends.insert(0, fetch_google)
+        available_sources = list(SOURCES)
+        if not self.config['google_API_key'].get() and \
+                'google' in SOURCES:
+            available_sources.remove('google')
+        self.config['sources'] = plugins.sanitize_choices(
+            self.config['sources'].as_str_seq(), available_sources)
+        self.backends = []
+        for key in self.config['sources'].as_str_seq():
+            self.backends.append(SOURCE_BACKENDS[key])
 
     def commands(self):
         cmd = ui.Subcommand('lyrics', help='fetch song lyrics')
@@ -490,8 +476,10 @@ class LyricsPlugin(BeetsPlugin):
             # import_write config value.
             write = config['import']['write'].get(bool)
             for item in lib.items(ui.decargs(args)):
-                self.fetch_item_lyrics(lib, logging.INFO, item, write,
-                                       opts.force_refetch)
+                self.fetch_item_lyrics(
+                    lib, logging.INFO, item, write,
+                    opts.force_refetch or self.config['force'],
+                )
                 if opts.printlyr and item.lyrics:
                     ui.print_(item.lyrics)
 
@@ -504,7 +492,7 @@ class LyricsPlugin(BeetsPlugin):
         if self.config['auto']:
             for item in task.imported_items():
                 self.fetch_item_lyrics(session.lib, logging.DEBUG, item,
-                                       False, False)
+                                       False, self.config['force'])
 
     def fetch_item_lyrics(self, lib, loglevel, item, write, force):
         """Fetch and store lyrics for a single item. If ``write``, then the
@@ -551,8 +539,6 @@ class LyricsPlugin(BeetsPlugin):
         for backend in self.backends:
             lyrics = backend(artist, title)
             if lyrics:
-                if isinstance(lyrics, str):
-                    lyrics = lyrics.decode('utf8', 'ignore')
                 log.debug(u'got lyrics from backend: {0}'
                           .format(backend.__name__))
-                return lyrics.strip()
+                return _scrape_strip_cruft(lyrics, True)

lib/biplist/__init__.py

@@ -0,0 +1,803 @@
+"""biplist -- a library for reading and writing binary property list files.
+
+Binary Property List (plist) files provide a faster and smaller serialization
+format for property lists on OS X. This is a library for generating binary
+plists which can be read by OS X, iOS, or other clients.
+
+The API models the plistlib API, and will call through to plistlib when
+XML serialization or deserialization is required.
+
+To generate plists with UID values, wrap the values with the Uid object. The
+value must be an int.
+
+To generate plists with NSData/CFData values, wrap the values with the
+Data object. The value must be a string.
+
+Date values can only be datetime.datetime objects.
+
+The exceptions InvalidPlistException and NotBinaryPlistException may be 
+thrown to indicate that the data cannot be serialized or deserialized as
+a binary plist.
+
+Plist generation example:
+    
+    from biplist import *
+    from datetime import datetime
+    plist = {'aKey':'aValue',
+             '0':1.322,
+             'now':datetime.now(),
+             'list':[1,2,3],
+             'tuple':('a','b','c')
+             }
+    try:
+        writePlist(plist, "example.plist")
+    except (InvalidPlistException, NotBinaryPlistException), e:
+        print "Something bad happened:", e
+
+Plist parsing example:
+
+    from biplist import *
+    try:
+        plist = readPlist("example.plist")
+        print plist
+    except (InvalidPlistException, NotBinaryPlistException), e:
+        print "Not a plist:", e
+"""
+
+import sys
+from collections import namedtuple
+import datetime
+import io
+import math
+import plistlib
+from struct import pack, unpack
+from struct import error as struct_error
+import sys
+import time
+
+try:
+    unicode
+    unicodeEmpty = r''
+except NameError:
+    unicode = str
+    unicodeEmpty = ''
+try:
+    long
+except NameError:
+    long = int
+try:
+    {}.iteritems
+    iteritems = lambda x: x.iteritems()
+except AttributeError:
+    iteritems = lambda x: x.items()
+
+__all__ = [
+    'Uid', 'Data', 'readPlist', 'writePlist', 'readPlistFromString',
+    'writePlistToString', 'InvalidPlistException', 'NotBinaryPlistException'
+]
+
+# Apple uses Jan 1, 2001 as a base for all plist date/times.
+apple_reference_date = datetime.datetime.utcfromtimestamp(978307200)
+
+class Uid(int):
+    """Wrapper around integers for representing UID values. This
+       is used in keyed archiving."""
+    def __repr__(self):
+        return "Uid(%d)" % self
+
+class Data(bytes):
+    """Wrapper around str types for representing Data values."""
+    pass
+
+class InvalidPlistException(Exception):
+    """Raised when the plist is incorrectly formatted."""
+    pass
+
+class NotBinaryPlistException(Exception):
+    """Raised when a binary plist was expected but not encountered."""
+    pass
+
+def readPlist(pathOrFile):
+    """Raises NotBinaryPlistException, InvalidPlistException"""
+    didOpen = False
+    result = None
+    if isinstance(pathOrFile, (bytes, unicode)):
+        pathOrFile = open(pathOrFile, 'rb')
+        didOpen = True
+    try:
+        reader = PlistReader(pathOrFile)
+        result = reader.parse()
+    except NotBinaryPlistException as e:
+        try:
+            pathOrFile.seek(0)
+            result = None
+            if hasattr(plistlib, 'loads'):
+                contents = None
+                if isinstance(pathOrFile, (bytes, unicode)):
+                    with open(pathOrFile, 'rb') as f:
+                        contents = f.read()
+                else:
+                    contents = pathOrFile.read()
+                result = plistlib.loads(contents)
+            else:
+                result = plistlib.readPlist(pathOrFile)
+            result = wrapDataObject(result, for_binary=True)
+        except Exception as e:
+            raise InvalidPlistException(e)
+    finally:
+        if didOpen:
+            pathOrFile.close()
+    return result
+
+def wrapDataObject(o, for_binary=False):
+    if isinstance(o, Data) and not for_binary:
+        v = sys.version_info
+        if not (v[0] >= 3 and v[1] >= 4):
+            o = plistlib.Data(o)
+    elif isinstance(o, (bytes, plistlib.Data)) and for_binary:
+        if hasattr(o, 'data'):
+            o = Data(o.data)
+    elif isinstance(o, tuple):
+        o = wrapDataObject(list(o), for_binary)
+        o = tuple(o)
+    elif isinstance(o, list):
+        for i in range(len(o)):
+            o[i] = wrapDataObject(o[i], for_binary)
+    elif isinstance(o, dict):
+        for k in o:
+            o[k] = wrapDataObject(o[k], for_binary)
+    return o
+
+def writePlist(rootObject, pathOrFile, binary=True):
+    if not binary:
+        rootObject = wrapDataObject(rootObject, binary)
+        if hasattr(plistlib, "dump"):
+            if isinstance(pathOrFile, (bytes, unicode)):
+                with open(pathOrFile, 'wb') as f:
+                    return plistlib.dump(rootObject, f)
+            else:
+                return plistlib.dump(rootObject, pathOrFile)
+        else:
+            return plistlib.writePlist(rootObject, pathOrFile)
+    else:
+        didOpen = False
+        if isinstance(pathOrFile, (bytes, unicode)):
+            pathOrFile = open(pathOrFile, 'wb')
+            didOpen = True
+        writer = PlistWriter(pathOrFile)
+        result = writer.writeRoot(rootObject)
+        if didOpen:
+            pathOrFile.close()
+        return result
+
+def readPlistFromString(data):
+    return readPlist(io.BytesIO(data))
+
+def writePlistToString(rootObject, binary=True):
+    if not binary:
+        rootObject = wrapDataObject(rootObject, binary)
+        if hasattr(plistlib, "dumps"):
+            return plistlib.dumps(rootObject)
+        elif hasattr(plistlib, "writePlistToBytes"):
+            return plistlib.writePlistToBytes(rootObject)
+        else:
+            return plistlib.writePlistToString(rootObject)
+    else:
+        ioObject = io.BytesIO()
+        writer = PlistWriter(ioObject)
+        writer.writeRoot(rootObject)
+        return ioObject.getvalue()
+
+def is_stream_binary_plist(stream):
+    stream.seek(0)
+    header = stream.read(7)
+    if header == b'bplist0':
+        return True
+    else:
+        return False
+
+PlistTrailer = namedtuple('PlistTrailer', 'offsetSize, objectRefSize, offsetCount, topLevelObjectNumber, offsetTableOffset')
+PlistByteCounts = namedtuple('PlistByteCounts', 'nullBytes, boolBytes, intBytes, realBytes, dateBytes, dataBytes, stringBytes, uidBytes, arrayBytes, setBytes, dictBytes')
+
+class PlistReader(object):
+    file = None
+    contents = ''
+    offsets = None
+    trailer = None
+    currentOffset = 0
+    
+    def __init__(self, fileOrStream):
+        """Raises NotBinaryPlistException."""
+        self.reset()
+        self.file = fileOrStream
+    
+    def parse(self):
+        return self.readRoot()
+    
+    def reset(self):
+        self.trailer = None
+        self.contents = ''
+        self.offsets = []
+        self.currentOffset = 0
+    
+    def readRoot(self):
+        result = None
+        self.reset()
+        # Get the header, make sure it's a valid file.
+        if not is_stream_binary_plist(self.file):
+            raise NotBinaryPlistException()
+        self.file.seek(0)
+        self.contents = self.file.read()
+        if len(self.contents) < 32:
+            raise InvalidPlistException("File is too short.")
+        trailerContents = self.contents[-32:]
+        try:
+            self.trailer = PlistTrailer._make(unpack("!xxxxxxBBQQQ", trailerContents))
+            offset_size = self.trailer.offsetSize * self.trailer.offsetCount
+            offset = self.trailer.offsetTableOffset
+            offset_contents = self.contents[offset:offset+offset_size]
+            offset_i = 0
+            while offset_i < self.trailer.offsetCount:
+                begin = self.trailer.offsetSize*offset_i
+                tmp_contents = offset_contents[begin:begin+self.trailer.offsetSize]
+                tmp_sized = self.getSizedInteger(tmp_contents, self.trailer.offsetSize)
+                self.offsets.append(tmp_sized)
+                offset_i += 1
+            self.setCurrentOffsetToObjectNumber(self.trailer.topLevelObjectNumber)
+            result = self.readObject()
+        except TypeError as e:
+            raise InvalidPlistException(e)
+        return result
+    
+    def setCurrentOffsetToObjectNumber(self, objectNumber):
+        self.currentOffset = self.offsets[objectNumber]
+    
+    def readObject(self):
+        result = None
+        tmp_byte = self.contents[self.currentOffset:self.currentOffset+1]
+        marker_byte = unpack("!B", tmp_byte)[0]
+        format = (marker_byte >> 4) & 0x0f
+        extra = marker_byte & 0x0f
+        self.currentOffset += 1
+        
+        def proc_extra(extra):
+            if extra == 0b1111:
+                #self.currentOffset += 1
+                extra = self.readObject()
+            return extra
+        
+        # bool, null, or fill byte
+        if format == 0b0000:
+            if extra == 0b0000:
+                result = None
+            elif extra == 0b1000:
+                result = False
+            elif extra == 0b1001:
+                result = True
+            elif extra == 0b1111:
+                pass # fill byte
+            else:
+                raise InvalidPlistException("Invalid object found at offset: %d" % (self.currentOffset - 1))
+        # int
+        elif format == 0b0001:
+            extra = proc_extra(extra)
+            result = self.readInteger(pow(2, extra))
+        # real
+        elif format == 0b0010:
+            extra = proc_extra(extra)
+            result = self.readReal(extra)
+        # date
+        elif format == 0b0011 and extra == 0b0011:
+            result = self.readDate()
+        # data
+        elif format == 0b0100:
+            extra = proc_extra(extra)
+            result = self.readData(extra)
+        # ascii string
+        elif format == 0b0101:
+            extra = proc_extra(extra)
+            result = self.readAsciiString(extra)
+        # Unicode string
+        elif format == 0b0110:
+            extra = proc_extra(extra)
+            result = self.readUnicode(extra)
+        # uid
+        elif format == 0b1000:
+            result = self.readUid(extra)
+        # array
+        elif format == 0b1010:
+            extra = proc_extra(extra)
+            result = self.readArray(extra)
+        # set
+        elif format == 0b1100:
+            extra = proc_extra(extra)
+            result = set(self.readArray(extra))
+        # dict
+        elif format == 0b1101:
+            extra = proc_extra(extra)
+            result = self.readDict(extra)
+        else:    
+            raise InvalidPlistException("Invalid object found: {format: %s, extra: %s}" % (bin(format), bin(extra)))
+        return result
+    
+    def readInteger(self, byteSize):
+        result = 0
+        original_offset = self.currentOffset
+        data = self.contents[self.currentOffset:self.currentOffset + byteSize]
+        result = self.getSizedInteger(data, byteSize, as_number=True)
+        self.currentOffset = original_offset + byteSize
+        return result
+    
+    def readReal(self, length):
+        result = 0.0
+        to_read = pow(2, length)
+        data = self.contents[self.currentOffset:self.currentOffset+to_read]
+        if length == 2: # 4 bytes
+            result = unpack('>f', data)[0]
+        elif length == 3: # 8 bytes
+            result = unpack('>d', data)[0]
+        else:
+            raise InvalidPlistException("Unknown real of length %d bytes" % to_read)
+        return result
+    
+    def readRefs(self, count):    
+        refs = []
+        i = 0
+        while i < count:
+            fragment = self.contents[self.currentOffset:self.currentOffset+self.trailer.objectRefSize]
+            ref = self.getSizedInteger(fragment, len(fragment))
+            refs.append(ref)
+            self.currentOffset += self.trailer.objectRefSize
+            i += 1
+        return refs
+    
+    def readArray(self, count):
+        result = []
+        values = self.readRefs(count)
+        i = 0
+        while i < len(values):
+            self.setCurrentOffsetToObjectNumber(values[i])
+            value = self.readObject()
+            result.append(value)
+            i += 1
+        return result
+    
+    def readDict(self, count):
+        result = {}
+        keys = self.readRefs(count)
+        values = self.readRefs(count)
+        i = 0
+        while i < len(keys):
+            self.setCurrentOffsetToObjectNumber(keys[i])
+            key = self.readObject()
+            self.setCurrentOffsetToObjectNumber(values[i])
+            value = self.readObject()
+            result[key] = value
+            i += 1
+        return result
+    
+    def readAsciiString(self, length):
+        result = unpack("!%ds" % length, self.contents[self.currentOffset:self.currentOffset+length])[0]
+        self.currentOffset += length
+        return result
+    
+    def readUnicode(self, length):
+        actual_length = length*2
+        data = self.contents[self.currentOffset:self.currentOffset+actual_length]
+        # unpack not needed?!! data = unpack(">%ds" % (actual_length), data)[0]
+        self.currentOffset += actual_length
+        return data.decode('utf_16_be')
+    
+    def readDate(self):
+        result = unpack(">d", self.contents[self.currentOffset:self.currentOffset+8])[0]
+        # Use timedelta to workaround time_t size limitation on 32-bit python.
+        result = datetime.timedelta(seconds=result) + apple_reference_date
+        self.currentOffset += 8
+        return result
+    
+    def readData(self, length):
+        result = self.contents[self.currentOffset:self.currentOffset+length]
+        self.currentOffset += length
+        return Data(result)
+    
+    def readUid(self, length):
+        return Uid(self.readInteger(length+1))
+    
+    def getSizedInteger(self, data, byteSize, as_number=False):
+        """Numbers of 8 bytes are signed integers when they refer to numbers, but unsigned otherwise."""
+        result = 0
+        # 1, 2, and 4 byte integers are unsigned
+        if byteSize == 1:
+            result = unpack('>B', data)[0]
+        elif byteSize == 2:
+            result = unpack('>H', data)[0]
+        elif byteSize == 4:
+            result = unpack('>L', data)[0]
+        elif byteSize == 8:
+            if as_number:
+                result = unpack('>q', data)[0]
+            else:
+                result = unpack('>Q', data)[0]
+        elif byteSize <= 16:
+            # Handle odd-sized or integers larger than 8 bytes
+            # Don't naively go over 16 bytes, in order to prevent infinite loops.
+            result = 0
+            if hasattr(int, 'from_bytes'):
+                result = int.from_bytes(data, 'big')
+            else:
+                for byte in data:
+                    result = (result << 8) | unpack('>B', byte)[0]
+        else:
+            raise InvalidPlistException("Encountered integer longer than 16 bytes.")
+        return result
+
+class HashableWrapper(object):
+    def __init__(self, value):
+        self.value = value
+    def __repr__(self):
+        return "<HashableWrapper: %s>" % [self.value]
+
+class BoolWrapper(object):
+    def __init__(self, value):
+        self.value = value
+    def __repr__(self):
+        return "<BoolWrapper: %s>" % self.value
+
+class FloatWrapper(object):
+    _instances = {}
+    def __new__(klass, value):
+        # Ensure FloatWrapper(x) for a given float x is always the same object
+        wrapper = klass._instances.get(value)
+        if wrapper is None:
+            wrapper = object.__new__(klass)
+            wrapper.value = value
+            klass._instances[value] = wrapper
+        return wrapper
+    def __repr__(self):
+        return "<FloatWrapper: %s>" % self.value
+
+class PlistWriter(object):
+    header = b'bplist00bybiplist1.0'
+    file = None
+    byteCounts = None
+    trailer = None
+    computedUniques = None
+    writtenReferences = None
+    referencePositions = None
+    wrappedTrue = None
+    wrappedFalse = None
+    
+    def __init__(self, file):
+        self.reset()
+        self.file = file
+        self.wrappedTrue = BoolWrapper(True)
+        self.wrappedFalse = BoolWrapper(False)
+
+    def reset(self):
+        self.byteCounts = PlistByteCounts(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)
+        self.trailer = PlistTrailer(0, 0, 0, 0, 0)
+        
+        # A set of all the uniques which have been computed.
+        self.computedUniques = set()
+        # A list of all the uniques which have been written.
+        self.writtenReferences = {}
+        # A dict of the positions of the written uniques.
+        self.referencePositions = {}
+        
+    def positionOfObjectReference(self, obj):
+        """If the given object has been written already, return its
+           position in the offset table. Otherwise, return None."""
+        return self.writtenReferences.get(obj)
+        
+    def writeRoot(self, root):
+        """
+        Strategy is:
+        - write header
+        - wrap root object so everything is hashable
+        - compute size of objects which will be written
+          - need to do this in order to know how large the object refs
+            will be in the list/dict/set reference lists
+        - write objects
+          - keep objects in writtenReferences
+          - keep positions of object references in referencePositions
+          - write object references with the length computed previously
+        - computer object reference length
+        - write object reference positions
+        - write trailer
+        """
+        output = self.header
+        wrapped_root = self.wrapRoot(root)
+        should_reference_root = True#not isinstance(wrapped_root, HashableWrapper)
+        self.computeOffsets(wrapped_root, asReference=should_reference_root, isRoot=True)
+        self.trailer = self.trailer._replace(**{'objectRefSize':self.intSize(len(self.computedUniques))})
+        (_, output) = self.writeObjectReference(wrapped_root, output)
+        output = self.writeObject(wrapped_root, output, setReferencePosition=True)
+        
+        # output size at this point is an upper bound on how big the
+        # object reference offsets need to be.
+        self.trailer = self.trailer._replace(**{
+            'offsetSize':self.intSize(len(output)),
+            'offsetCount':len(self.computedUniques),
+            'offsetTableOffset':len(output),
+            'topLevelObjectNumber':0
+            })
+        
+        output = self.writeOffsetTable(output)
+        output += pack('!xxxxxxBBQQQ', *self.trailer)
+        self.file.write(output)
+
+    def wrapRoot(self, root):
+        if isinstance(root, bool):
+            if root is True:
+                return self.wrappedTrue
+            else:
+                return self.wrappedFalse
+        elif isinstance(root, float):
+            return FloatWrapper(root)
+        elif isinstance(root, set):
+            n = set()
+            for value in root:
+                n.add(self.wrapRoot(value))
+            return HashableWrapper(n)
+        elif isinstance(root, dict):
+            n = {}
+            for key, value in iteritems(root):
+                n[self.wrapRoot(key)] = self.wrapRoot(value)
+            return HashableWrapper(n)
+        elif isinstance(root, list):
+            n = []
+            for value in root:
+                n.append(self.wrapRoot(value))
+            return HashableWrapper(n)
+        elif isinstance(root, tuple):
+            n = tuple([self.wrapRoot(value) for value in root])
+            return HashableWrapper(n)
+        else:
+            return root
+
+    def incrementByteCount(self, field, incr=1):
+        self.byteCounts = self.byteCounts._replace(**{field:self.byteCounts.__getattribute__(field) + incr})
+
+    def computeOffsets(self, obj, asReference=False, isRoot=False):
+        def check_key(key):
+            if key is None:
+                raise InvalidPlistException('Dictionary keys cannot be null in plists.')
+            elif isinstance(key, Data):
+                raise InvalidPlistException('Data cannot be dictionary keys in plists.')
+            elif not isinstance(key, (bytes, unicode)):
+                raise InvalidPlistException('Keys must be strings.')
+        
+        def proc_size(size):
+            if size > 0b1110:
+                size += self.intSize(size)
+            return size
+        # If this should be a reference, then we keep a record of it in the
+        # uniques table.
+        if asReference:
+            if obj in self.computedUniques:
+                return
+            else:
+                self.computedUniques.add(obj)
+        
+        if obj is None:
+            self.incrementByteCount('nullBytes')
+        elif isinstance(obj, BoolWrapper):
+            self.incrementByteCount('boolBytes')
+        elif isinstance(obj, Uid):
+            size = self.intSize(obj)
+            self.incrementByteCount('uidBytes', incr=1+size)
+        elif isinstance(obj, (int, long)):
+            size = self.intSize(obj)
+            self.incrementByteCount('intBytes', incr=1+size)
+        elif isinstance(obj, FloatWrapper):
+            size = self.realSize(obj)
+            self.incrementByteCount('realBytes', incr=1+size)
+        elif isinstance(obj, datetime.datetime):    
+            self.incrementByteCount('dateBytes', incr=2)
+        elif isinstance(obj, Data):
+            size = proc_size(len(obj))
+            self.incrementByteCount('dataBytes', incr=1+size)
+        elif isinstance(obj, (unicode, bytes)):
+            size = proc_size(len(obj))
+            self.incrementByteCount('stringBytes', incr=1+size)
+        elif isinstance(obj, HashableWrapper):
+            obj = obj.value
+            if isinstance(obj, set):
+                size = proc_size(len(obj))
+                self.incrementByteCount('setBytes', incr=1+size)
+                for value in obj:
+                    self.computeOffsets(value, asReference=True)
+            elif isinstance(obj, (list, tuple)):
+                size = proc_size(len(obj))
+                self.incrementByteCount('arrayBytes', incr=1+size)
+                for value in obj:
+                    asRef = True
+                    self.computeOffsets(value, asReference=True)
+            elif isinstance(obj, dict):
+                size = proc_size(len(obj))
+                self.incrementByteCount('dictBytes', incr=1+size)
+                for key, value in iteritems(obj):
+                    check_key(key)
+                    self.computeOffsets(key, asReference=True)
+                    self.computeOffsets(value, asReference=True)
+        else:
+            raise InvalidPlistException("Unknown object type.")
+
+    def writeObjectReference(self, obj, output):
+        """Tries to write an object reference, adding it to the references
+           table. Does not write the actual object bytes or set the reference
+           position. Returns a tuple of whether the object was a new reference
+           (True if it was, False if it already was in the reference table)
+           and the new output.
+        """
+        position = self.positionOfObjectReference(obj)
+        if position is None:
+            self.writtenReferences[obj] = len(self.writtenReferences)
+            output += self.binaryInt(len(self.writtenReferences) - 1, byteSize=self.trailer.objectRefSize)
+            return (True, output)
+        else:
+            output += self.binaryInt(position, byteSize=self.trailer.objectRefSize)
+            return (False, output)
+
+    def writeObject(self, obj, output, setReferencePosition=False):
+        """Serializes the given object to the output. Returns output.
+           If setReferencePosition is True, will set the position the
+           object was written.
+        """
+        def proc_variable_length(format, length):
+            result = b''
+            if length > 0b1110:
+                result += pack('!B', (format << 4) | 0b1111)
+                result = self.writeObject(length, result)
+            else:
+                result += pack('!B', (format << 4) | length)
+            return result
+        
+        if isinstance(obj, (str, unicode)) and obj == unicodeEmpty:
+            # The Apple Plist decoder can't decode a zero length Unicode string.
+            obj = b''
+       
+        if setReferencePosition:
+            self.referencePositions[obj] = len(output)
+        
+        if obj is None:
+            output += pack('!B', 0b00000000)
+        elif isinstance(obj, BoolWrapper):
+            if obj.value is False:
+                output += pack('!B', 0b00001000)
+            else:
+                output += pack('!B', 0b00001001)
+        elif isinstance(obj, Uid):
+            size = self.intSize(obj)
+            output += pack('!B', (0b1000 << 4) | size - 1)
+            output += self.binaryInt(obj)
+        elif isinstance(obj, (int, long)):
+            byteSize = self.intSize(obj)
+            root = math.log(byteSize, 2)
+            output += pack('!B', (0b0001 << 4) | int(root))
+            output += self.binaryInt(obj, as_number=True)
+        elif isinstance(obj, FloatWrapper):
+            # just use doubles
+            output += pack('!B', (0b0010 << 4) | 3)
+            output += self.binaryReal(obj)
+        elif isinstance(obj, datetime.datetime):
+            timestamp = (obj - apple_reference_date).total_seconds()
+            output += pack('!B', 0b00110011)
+            output += pack('!d', float(timestamp))
+        elif isinstance(obj, Data):
+            output += proc_variable_length(0b0100, len(obj))
+            output += obj
+        elif isinstance(obj, unicode):
+            byteData = obj.encode('utf_16_be')
+            output += proc_variable_length(0b0110, len(byteData)//2)
+            output += byteData
+        elif isinstance(obj, bytes):
+            output += proc_variable_length(0b0101, len(obj))
+            output += obj
+        elif isinstance(obj, HashableWrapper):
+            obj = obj.value
+            if isinstance(obj, (set, list, tuple)):
+                if isinstance(obj, set):
+                    output += proc_variable_length(0b1100, len(obj))
+                else:
+                    output += proc_variable_length(0b1010, len(obj))
+            
+                objectsToWrite = []
+                for objRef in obj:
+                    (isNew, output) = self.writeObjectReference(objRef, output)
+                    if isNew:
+                        objectsToWrite.append(objRef)
+                for objRef in objectsToWrite:
+                    output = self.writeObject(objRef, output, setReferencePosition=True)
+            elif isinstance(obj, dict):
+                output += proc_variable_length(0b1101, len(obj))
+                keys = []
+                values = []
+                objectsToWrite = []
+                for key, value in iteritems(obj):
+                    keys.append(key)
+                    values.append(value)
+                for key in keys:
+                    (isNew, output) = self.writeObjectReference(key, output)
+                    if isNew:
+                        objectsToWrite.append(key)
+                for value in values:
+                    (isNew, output) = self.writeObjectReference(value, output)
+                    if isNew:
+                        objectsToWrite.append(value)
+                for objRef in objectsToWrite:
+                    output = self.writeObject(objRef, output, setReferencePosition=True)
+        return output
+    
+    def writeOffsetTable(self, output):
+        """Writes all of the object reference offsets."""
+        all_positions = []
+        writtenReferences = list(self.writtenReferences.items())
+        writtenReferences.sort(key=lambda x: x[1])
+        for obj,order in writtenReferences:
+            # Porting note: Elsewhere we deliberately replace empty unicdoe strings
+            # with empty binary strings, but the empty unicode string
+            # goes into writtenReferences.  This isn't an issue in Py2
+            # because u'' and b'' have the same hash; but it is in
+            # Py3, where they don't.
+            if bytes != str and obj == unicodeEmpty:
+                obj = b''
+            position = self.referencePositions.get(obj)
+            if position is None:
+                raise InvalidPlistException("Error while writing offsets table. Object not found. %s" % obj)
+            output += self.binaryInt(position, self.trailer.offsetSize)
+            all_positions.append(position)
+        return output
+    
+    def binaryReal(self, obj):
+        # just use doubles
+        result = pack('>d', obj.value)
+        return result
+    
+    def binaryInt(self, obj, byteSize=None, as_number=False):
+        result = b''
+        if byteSize is None:
+            byteSize = self.intSize(obj)
+        if byteSize == 1:
+            result += pack('>B', obj)
+        elif byteSize == 2:
+            result += pack('>H', obj)
+        elif byteSize == 4:
+            result += pack('>L', obj)
+        elif byteSize == 8:
+            if as_number:
+                result += pack('>q', obj)
+            else:
+                result += pack('>Q', obj)
+        elif byteSize <= 16:
+            try:
+                result = pack('>Q', 0) + pack('>Q', obj)
+            except struct_error as e:
+                raise InvalidPlistException("Unable to pack integer %d: %s" % (obj, e))
+        else:
+            raise InvalidPlistException("Core Foundation can't handle integers with size greater than 16 bytes.")
+        return result
+    
+    def intSize(self, obj):
+        """Returns the number of bytes necessary to store the given integer."""
+        # SIGNED
+        if obj < 0: # Signed integer, always 8 bytes
+            return 8
+        # UNSIGNED
+        elif obj <= 0xFF: # 1 byte
+            return 1
+        elif obj <= 0xFFFF: # 2 bytes
+            return 2
+        elif obj <= 0xFFFFFFFF: # 4 bytes
+            return 4
+        # SIGNED
+        # 0x7FFFFFFFFFFFFFFF is the max.
+        elif obj <= 0x7FFFFFFFFFFFFFFF: # 8 bytes signed
+            return 8
+        elif obj <= 0xffffffffffffffff: # 8 bytes unsigned
+            return 16
+        else:
+            raise InvalidPlistException("Core Foundation can't handle integers with size greater than 8 bytes.")
+    
+    def realSize(self, obj):
+        return 8

lib/cherrypy/wsgiserver/wsgiserver2.py

@@ -1324,8 +1324,8 @@ class HTTPConnection(object):
     def __init__(self, server, sock, makefile=CP_fileobject):
         self.server = server
         self.socket = sock
-        self.rfile = makefile(sock._sock, "rb", self.rbufsize)
-        self.wfile = makefile(sock._sock, "wb", self.wbufsize)
+        self.rfile = makefile(sock, "rb", self.rbufsize)
+        self.wfile = makefile(sock, "wb", self.wbufsize)
         self.requests_seen = 0
 
     def communicate(self):

lib/enum/LICENSE

@@ -0,0 +1,32 @@
+Copyright (c) 2013, Ethan Furman.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+
+    Redistributions of source code must retain the above
+    copyright notice, this list of conditions and the
+    following disclaimer.
+
+    Redistributions in binary form must reproduce the above
+    copyright notice, this list of conditions and the following
+    disclaimer in the documentation and/or other materials
+    provided with the distribution.
+
+    Neither the name Ethan Furman nor the names of any
+    contributors may be used to endorse or promote products
+    derived from this software without specific prior written
+    permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.

lib/enum/README

@@ -0,0 +1,2 @@
+enum34 is the new Python stdlib enum module available in Python 3.4
+backported for previous versions of Python from 2.4 to 3.3.

lib/enum/__init__.py

@@ -0,0 +1,790 @@
+"""Python Enumerations"""
+
+import sys as _sys
+
+__all__ = ['Enum', 'IntEnum', 'unique']
+
+version = 1, 0, 4
+
+pyver = float('%s.%s' % _sys.version_info[:2])
+
+try:
+    any
+except NameError:
+    def any(iterable):
+        for element in iterable:
+            if element:
+                return True
+        return False
+
+try:
+    from collections import OrderedDict
+except ImportError:
+    OrderedDict = None
+
+try:
+    basestring
+except NameError:
+    # In Python 2 basestring is the ancestor of both str and unicode
+    # in Python 3 it's just str, but was missing in 3.1
+    basestring = str
+
+try:
+    unicode
+except NameError:
+    # In Python 3 unicode no longer exists (it's just str)
+    unicode = str
+
+class _RouteClassAttributeToGetattr(object):
+    """Route attribute access on a class to __getattr__.
+
+    This is a descriptor, used to define attributes that act differently when
+    accessed through an instance and through a class.  Instance access remains
+    normal, but access to an attribute through a class will be routed to the
+    class's __getattr__ method; this is done by raising AttributeError.
+
+    """
+    def __init__(self, fget=None):
+        self.fget = fget
+
+    def __get__(self, instance, ownerclass=None):
+        if instance is None:
+            raise AttributeError()
+        return self.fget(instance)
+
+    def __set__(self, instance, value):
+        raise AttributeError("can't set attribute")
+
+    def __delete__(self, instance):
+        raise AttributeError("can't delete attribute")
+
+
+def _is_descriptor(obj):
+    """Returns True if obj is a descriptor, False otherwise."""
+    return (
+            hasattr(obj, '__get__') or
+            hasattr(obj, '__set__') or
+            hasattr(obj, '__delete__'))
+
+
+def _is_dunder(name):
+    """Returns True if a __dunder__ name, False otherwise."""
+    return (name[:2] == name[-2:] == '__' and
+            name[2:3] != '_' and
+            name[-3:-2] != '_' and
+            len(name) > 4)
+
+
+def _is_sunder(name):
+    """Returns True if a _sunder_ name, False otherwise."""
+    return (name[0] == name[-1] == '_' and 
+            name[1:2] != '_' and
+            name[-2:-1] != '_' and
+            len(name) > 2)
+
+
+def _make_class_unpicklable(cls):
+    """Make the given class un-picklable."""
+    def _break_on_call_reduce(self, protocol=None):
+        raise TypeError('%r cannot be pickled' % self)
+    cls.__reduce_ex__ = _break_on_call_reduce
+    cls.__module__ = '<unknown>'
+
+
+class _EnumDict(dict):
+    """Track enum member order and ensure member names are not reused.
+
+    EnumMeta will use the names found in self._member_names as the
+    enumeration member names.
+
+    """
+    def __init__(self):
+        super(_EnumDict, self).__init__()
+        self._member_names = []
+
+    def __setitem__(self, key, value):
+        """Changes anything not dundered or not a descriptor.
+
+        If a descriptor is added with the same name as an enum member, the name
+        is removed from _member_names (this may leave a hole in the numerical
+        sequence of values).
+
+        If an enum member name is used twice, an error is raised; duplicate
+        values are not checked for.
+
+        Single underscore (sunder) names are reserved.
+
+        Note:   in 3.x __order__ is simply discarded as a not necessary piece
+                leftover from 2.x
+
+        """
+        if pyver >= 3.0 and key == '__order__':
+                return
+        if _is_sunder(key):
+            raise ValueError('_names_ are reserved for future Enum use')
+        elif _is_dunder(key):
+            pass
+        elif key in self._member_names:
+            # descriptor overwriting an enum?
+            raise TypeError('Attempted to reuse key: %r' % key)
+        elif not _is_descriptor(value):
+            if key in self:
+                # enum overwriting a descriptor?
+                raise TypeError('Key already defined as: %r' % self[key])
+            self._member_names.append(key)
+        super(_EnumDict, self).__setitem__(key, value)
+
+
+# Dummy value for Enum as EnumMeta explicity checks for it, but of course until
+# EnumMeta finishes running the first time the Enum class doesn't exist.  This
+# is also why there are checks in EnumMeta like `if Enum is not None`
+Enum = None
+
+
+class EnumMeta(type):
+    """Metaclass for Enum"""
+    @classmethod
+    def __prepare__(metacls, cls, bases):
+        return _EnumDict()
+
+    def __new__(metacls, cls, bases, classdict):
+        # an Enum class is final once enumeration items have been defined; it
+        # cannot be mixed with other types (int, float, etc.) if it has an
+        # inherited __new__ unless a new __new__ is defined (or the resulting
+        # class will fail).
+        if type(classdict) is dict:
+            original_dict = classdict
+            classdict = _EnumDict()
+            for k, v in original_dict.items():
+                classdict[k] = v
+
+        member_type, first_enum = metacls._get_mixins_(bases)
+        __new__, save_new, use_args = metacls._find_new_(classdict, member_type,
+                                                        first_enum)
+        # save enum items into separate mapping so they don't get baked into
+        # the new class
+        members = dict((k, classdict[k]) for k in classdict._member_names)
+        for name in classdict._member_names:
+            del classdict[name]
+
+        # py2 support for definition order
+        __order__ = classdict.get('__order__')
+        if __order__ is None:
+            if pyver < 3.0:
+                try:
+                    __order__ = [name for (name, value) in sorted(members.items(), key=lambda item: item[1])]
+                except TypeError:
+                    __order__ = [name for name in sorted(members.keys())]
+            else:
+                __order__ = classdict._member_names
+        else:
+            del classdict['__order__']
+            if pyver < 3.0:
+                __order__ = __order__.replace(',', ' ').split()
+                aliases = [name for name in members if name not in __order__]
+                __order__ += aliases
+
+        # check for illegal enum names (any others?)
+        invalid_names = set(members) & set(['mro'])
+        if invalid_names:
+            raise ValueError('Invalid enum member name(s): %s' % (
+                ', '.join(invalid_names), ))
+
+        # create our new Enum type
+        enum_class = super(EnumMeta, metacls).__new__(metacls, cls, bases, classdict)
+        enum_class._member_names_ = []               # names in random order
+        if OrderedDict is not None:
+            enum_class._member_map_ = OrderedDict()
+        else:
+            enum_class._member_map_ = {}             # name->value map
+        enum_class._member_type_ = member_type
+
+        # Reverse value->name map for hashable values.
+        enum_class._value2member_map_ = {}
+
+        # instantiate them, checking for duplicates as we go
+        # we instantiate first instead of checking for duplicates first in case
+        # a custom __new__ is doing something funky with the values -- such as
+        # auto-numbering ;)
+        if __new__ is None:
+            __new__ = enum_class.__new__
+        for member_name in __order__:
+            value = members[member_name]
+            if not isinstance(value, tuple):
+                args = (value, )
+            else:
+                args = value
+            if member_type is tuple:   # special case for tuple enums
+                args = (args, )     # wrap it one more time
+            if not use_args or not args:
+                enum_member = __new__(enum_class)
+                if not hasattr(enum_member, '_value_'):
+                    enum_member._value_ = value
+            else:
+                enum_member = __new__(enum_class, *args)
+                if not hasattr(enum_member, '_value_'):
+                    enum_member._value_ = member_type(*args)
+            value = enum_member._value_
+            enum_member._name_ = member_name
+            enum_member.__objclass__ = enum_class
+            enum_member.__init__(*args)
+            # If another member with the same value was already defined, the
+            # new member becomes an alias to the existing one.
+            for name, canonical_member in enum_class._member_map_.items():
+                if canonical_member.value == enum_member._value_:
+                    enum_member = canonical_member
+                    break
+            else:
+                # Aliases don't appear in member names (only in __members__).
+                enum_class._member_names_.append(member_name)
+            enum_class._member_map_[member_name] = enum_member
+            try:
+                # This may fail if value is not hashable. We can't add the value
+                # to the map, and by-value lookups for this value will be
+                # linear.
+                enum_class._value2member_map_[value] = enum_member
+            except TypeError:
+                pass
+
+
+        # If a custom type is mixed into the Enum, and it does not know how
+        # to pickle itself, pickle.dumps will succeed but pickle.loads will
+        # fail.  Rather than have the error show up later and possibly far
+        # from the source, sabotage the pickle protocol for this class so
+        # that pickle.dumps also fails.
+        #
+        # However, if the new class implements its own __reduce_ex__, do not
+        # sabotage -- it's on them to make sure it works correctly.  We use
+        # __reduce_ex__ instead of any of the others as it is preferred by
+        # pickle over __reduce__, and it handles all pickle protocols.
+        unpicklable = False
+        if '__reduce_ex__' not in classdict:
+            if member_type is not object:
+                methods = ('__getnewargs_ex__', '__getnewargs__',
+                        '__reduce_ex__', '__reduce__')
+                if not any(m in member_type.__dict__ for m in methods):
+                    _make_class_unpicklable(enum_class)
+                    unpicklable = True
+
+
+        # double check that repr and friends are not the mixin's or various
+        # things break (such as pickle)
+        for name in ('__repr__', '__str__', '__format__', '__reduce_ex__'):
+            class_method = getattr(enum_class, name)
+            obj_method = getattr(member_type, name, None)
+            enum_method = getattr(first_enum, name, None)
+            if name not in classdict and class_method is not enum_method:
+                if name == '__reduce_ex__' and unpicklable:
+                    continue
+                setattr(enum_class, name, enum_method)
+
+        # method resolution and int's are not playing nice
+        # Python's less than 2.6 use __cmp__
+
+        if pyver < 2.6:
+
+            if issubclass(enum_class, int):
+                setattr(enum_class, '__cmp__', getattr(int, '__cmp__'))
+
+        elif pyver < 3.0:
+
+            if issubclass(enum_class, int):
+                for method in (
+                        '__le__',
+                        '__lt__',
+                        '__gt__',
+                        '__ge__',
+                        '__eq__',
+                        '__ne__',
+                        '__hash__',
+                        ):
+                    setattr(enum_class, method, getattr(int, method))
+
+        # replace any other __new__ with our own (as long as Enum is not None,
+        # anyway) -- again, this is to support pickle
+        if Enum is not None:
+            # if the user defined their own __new__, save it before it gets
+            # clobbered in case they subclass later
+            if save_new:
+                setattr(enum_class, '__member_new__', enum_class.__dict__['__new__'])
+            setattr(enum_class, '__new__', Enum.__dict__['__new__'])
+        return enum_class
+
+    def __call__(cls, value, names=None, module=None, type=None):
+        """Either returns an existing member, or creates a new enum class.
+
+        This method is used both when an enum class is given a value to match
+        to an enumeration member (i.e. Color(3)) and for the functional API
+        (i.e. Color = Enum('Color', names='red green blue')).
+
+        When used for the functional API: `module`, if set, will be stored in
+        the new class' __module__ attribute; `type`, if set, will be mixed in
+        as the first base class.
+
+        Note: if `module` is not set this routine will attempt to discover the
+        calling module by walking the frame stack; if this is unsuccessful
+        the resulting class will not be pickleable.
+
+        """
+        if names is None:  # simple value lookup
+            return cls.__new__(cls, value)
+        # otherwise, functional API: we're creating a new Enum type
+        return cls._create_(value, names, module=module, type=type)
+
+    def __contains__(cls, member):
+        return isinstance(member, cls) and member.name in cls._member_map_
+
+    def __delattr__(cls, attr):
+        # nicer error message when someone tries to delete an attribute
+        # (see issue19025).
+        if attr in cls._member_map_:
+            raise AttributeError(
+                    "%s: cannot delete Enum member." % cls.__name__)
+        super(EnumMeta, cls).__delattr__(attr)
+
+    def __dir__(self):
+        return (['__class__', '__doc__', '__members__', '__module__'] +
+                self._member_names_)
+
+    @property
+    def __members__(cls):
+        """Returns a mapping of member name->value.
+
+        This mapping lists all enum members, including aliases. Note that this
+        is a copy of the internal mapping.
+
+        """
+        return cls._member_map_.copy()
+
+    def __getattr__(cls, name):
+        """Return the enum member matching `name`
+
+        We use __getattr__ instead of descriptors or inserting into the enum
+        class' __dict__ in order to support `name` and `value` being both
+        properties for enum members (which live in the class' __dict__) and
+        enum members themselves.
+
+        """
+        if _is_dunder(name):
+            raise AttributeError(name)
+        try:
+            return cls._member_map_[name]
+        except KeyError:
+            raise AttributeError(name)
+
+    def __getitem__(cls, name):
+        return cls._member_map_[name]
+
+    def __iter__(cls):
+        return (cls._member_map_[name] for name in cls._member_names_)
+
+    def __reversed__(cls):
+        return (cls._member_map_[name] for name in reversed(cls._member_names_))
+
+    def __len__(cls):
+        return len(cls._member_names_)
+
+    def __repr__(cls):
+        return "<enum %r>" % cls.__name__
+
+    def __setattr__(cls, name, value):
+        """Block attempts to reassign Enum members.
+
+        A simple assignment to the class namespace only changes one of the
+        several possible ways to get an Enum member from the Enum class,
+        resulting in an inconsistent Enumeration.
+
+        """
+        member_map = cls.__dict__.get('_member_map_', {})
+        if name in member_map:
+            raise AttributeError('Cannot reassign members.')
+        super(EnumMeta, cls).__setattr__(name, value)
+
+    def _create_(cls, class_name, names=None, module=None, type=None):
+        """Convenience method to create a new Enum class.
+
+        `names` can be:
+
+        * A string containing member names, separated either with spaces or
+          commas.  Values are auto-numbered from 1.
+        * An iterable of member names.  Values are auto-numbered from 1.
+        * An iterable of (member name, value) pairs.
+        * A mapping of member name -> value.
+
+        """
+        if pyver < 3.0:
+            # if class_name is unicode, attempt a conversion to ASCII
+            if isinstance(class_name, unicode):
+                try:
+                    class_name = class_name.encode('ascii')
+                except UnicodeEncodeError:
+                    raise TypeError('%r is not representable in ASCII' % class_name)
+        metacls = cls.__class__
+        if type is None:
+            bases = (cls, )
+        else:
+            bases = (type, cls)
+        classdict = metacls.__prepare__(class_name, bases)
+        __order__ = []
+
+        # special processing needed for names?
+        if isinstance(names, basestring):
+            names = names.replace(',', ' ').split()
+        if isinstance(names, (tuple, list)) and isinstance(names[0], basestring):
+            names = [(e, i+1) for (i, e) in enumerate(names)]
+
+        # Here, names is either an iterable of (name, value) or a mapping.
+        for item in names:
+            if isinstance(item, basestring):
+                member_name, member_value = item, names[item]
+            else:
+                member_name, member_value = item
+            classdict[member_name] = member_value
+            __order__.append(member_name)
+        # only set __order__ in classdict if name/value was not from a mapping
+        if not isinstance(item, basestring):
+            classdict['__order__'] = ' '.join(__order__)
+        enum_class = metacls.__new__(metacls, class_name, bases, classdict)
+
+        # TODO: replace the frame hack if a blessed way to know the calling
+        # module is ever developed
+        if module is None:
+            try:
+                module = _sys._getframe(2).f_globals['__name__']
+            except (AttributeError, ValueError):
+                pass
+        if module is None:
+            _make_class_unpicklable(enum_class)
+        else:
+            enum_class.__module__ = module
+
+        return enum_class
+
+    @staticmethod
+    def _get_mixins_(bases):
+        """Returns the type for creating enum members, and the first inherited
+        enum class.
+
+        bases: the tuple of bases that was given to __new__
+
+        """
+        if not bases or Enum is None:
+            return object, Enum
+        
+
+        # double check that we are not subclassing a class with existing
+        # enumeration members; while we're at it, see if any other data
+        # type has been mixed in so we can use the correct __new__
+        member_type = first_enum = None
+        for base in bases:
+            if  (base is not Enum and
+                    issubclass(base, Enum) and
+                    base._member_names_):
+                raise TypeError("Cannot extend enumerations")
+        # base is now the last base in bases
+        if not issubclass(base, Enum):
+            raise TypeError("new enumerations must be created as "
+                    "`ClassName([mixin_type,] enum_type)`")
+
+        # get correct mix-in type (either mix-in type of Enum subclass, or
+        # first base if last base is Enum)
+        if not issubclass(bases[0], Enum):
+            member_type = bases[0]     # first data type
+            first_enum = bases[-1]  # enum type
+        else:
+            for base in bases[0].__mro__:
+                # most common: (IntEnum, int, Enum, object)
+                # possible:    (<Enum 'AutoIntEnum'>, <Enum 'IntEnum'>,
+                #               <class 'int'>, <Enum 'Enum'>,
+                #               <class 'object'>)
+                if issubclass(base, Enum):
+                    if first_enum is None:
+                        first_enum = base
+                else:
+                    if member_type is None:
+                        member_type = base
+
+        return member_type, first_enum
+
+    if pyver < 3.0:
+        @staticmethod
+        def _find_new_(classdict, member_type, first_enum):
+            """Returns the __new__ to be used for creating the enum members.
+
+            classdict: the class dictionary given to __new__
+            member_type: the data type whose __new__ will be used by default
+            first_enum: enumeration to check for an overriding __new__
+
+            """
+            # now find the correct __new__, checking to see of one was defined
+            # by the user; also check earlier enum classes in case a __new__ was
+            # saved as __member_new__
+            __new__ = classdict.get('__new__', None)
+            if __new__:
+                return None, True, True      # __new__, save_new, use_args
+
+            N__new__ = getattr(None, '__new__')
+            O__new__ = getattr(object, '__new__')
+            if Enum is None:
+                E__new__ = N__new__
+            else:
+                E__new__ = Enum.__dict__['__new__']
+            # check all possibles for __member_new__ before falling back to
+            # __new__
+            for method in ('__member_new__', '__new__'):
+                for possible in (member_type, first_enum):
+                    try:
+                        target = possible.__dict__[method]
+                    except (AttributeError, KeyError):
+                        target = getattr(possible, method, None)
+                    if target not in [
+                            None,
+                            N__new__,
+                            O__new__,
+                            E__new__,
+                            ]:
+                        if method == '__member_new__':
+                            classdict['__new__'] = target
+                            return None, False, True
+                        if isinstance(target, staticmethod):
+                            target = target.__get__(member_type)
+                        __new__ = target
+                        break
+                if __new__ is not None:
+                    break
+            else:
+                __new__ = object.__new__
+
+            # if a non-object.__new__ is used then whatever value/tuple was
+            # assigned to the enum member name will be passed to __new__ and to the
+            # new enum member's __init__
+            if __new__ is object.__new__:
+                use_args = False
+            else:
+                use_args = True
+
+            return __new__, False, use_args
+    else:
+        @staticmethod
+        def _find_new_(classdict, member_type, first_enum):
+            """Returns the __new__ to be used for creating the enum members.
+
+            classdict: the class dictionary given to __new__
+            member_type: the data type whose __new__ will be used by default
+            first_enum: enumeration to check for an overriding __new__
+
+            """
+            # now find the correct __new__, checking to see of one was defined
+            # by the user; also check earlier enum classes in case a __new__ was
+            # saved as __member_new__
+            __new__ = classdict.get('__new__', None)
+
+            # should __new__ be saved as __member_new__ later?
+            save_new = __new__ is not None
+
+            if __new__ is None:
+                # check all possibles for __member_new__ before falling back to
+                # __new__
+                for method in ('__member_new__', '__new__'):
+                    for possible in (member_type, first_enum):
+                        target = getattr(possible, method, None)
+                        if target not in (
+                                None,
+                                None.__new__,
+                                object.__new__,
+                                Enum.__new__,
+                                ):
+                            __new__ = target
+                            break
+                    if __new__ is not None:
+                        break
+                else:
+                    __new__ = object.__new__
+
+            # if a non-object.__new__ is used then whatever value/tuple was
+            # assigned to the enum member name will be passed to __new__ and to the
+            # new enum member's __init__
+            if __new__ is object.__new__:
+                use_args = False
+            else:
+                use_args = True
+
+            return __new__, save_new, use_args
+
+
+########################################################
+# In order to support Python 2 and 3 with a single
+# codebase we have to create the Enum methods separately
+# and then use the `type(name, bases, dict)` method to
+# create the class.
+########################################################
+temp_enum_dict = {}
+temp_enum_dict['__doc__'] = "Generic enumeration.\n\n    Derive from this class to define new enumerations.\n\n"
+
+def __new__(cls, value):
+    # all enum instances are actually created during class construction
+    # without calling this method; this method is called by the metaclass'
+    # __call__ (i.e. Color(3) ), and by pickle
+    if type(value) is cls:
+        # For lookups like Color(Color.red)
+        value = value.value
+        #return value
+    # by-value search for a matching enum member
+    # see if it's in the reverse mapping (for hashable values)
+    try:
+        if value in cls._value2member_map_:
+            return cls._value2member_map_[value]
+    except TypeError:
+        # not there, now do long search -- O(n) behavior
+        for member in cls._member_map_.values():
+            if member.value == value:
+                return member
+    raise ValueError("%s is not a valid %s" % (value, cls.__name__))
+temp_enum_dict['__new__'] = __new__
+del __new__
+
+def __repr__(self):
+    return "<%s.%s: %r>" % (
+            self.__class__.__name__, self._name_, self._value_)
+temp_enum_dict['__repr__'] = __repr__
+del __repr__
+
+def __str__(self):
+    return "%s.%s" % (self.__class__.__name__, self._name_)
+temp_enum_dict['__str__'] = __str__
+del __str__
+
+def __dir__(self):
+    added_behavior = [
+            m
+            for cls in self.__class__.mro()
+            for m in cls.__dict__
+            if m[0] != '_'
+            ]
+    return (['__class__', '__doc__', '__module__', ] + added_behavior)
+temp_enum_dict['__dir__'] = __dir__
+del __dir__
+
+def __format__(self, format_spec):
+    # mixed-in Enums should use the mixed-in type's __format__, otherwise
+    # we can get strange results with the Enum name showing up instead of
+    # the value
+
+    # pure Enum branch
+    if self._member_type_ is object:
+        cls = str
+        val = str(self)
+    # mix-in branch
+    else:
+        cls = self._member_type_
+        val = self.value
+    return cls.__format__(val, format_spec)
+temp_enum_dict['__format__'] = __format__
+del __format__
+
+
+####################################
+# Python's less than 2.6 use __cmp__
+
+if pyver < 2.6:
+
+    def __cmp__(self, other):
+        if type(other) is self.__class__:
+            if self is other:
+                return 0
+            return -1
+        return NotImplemented
+        raise TypeError("unorderable types: %s() and %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__cmp__'] = __cmp__
+    del __cmp__
+
+else:
+
+    def __le__(self, other):
+        raise TypeError("unorderable types: %s() <= %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__le__'] = __le__
+    del __le__
+
+    def __lt__(self, other):
+        raise TypeError("unorderable types: %s() < %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__lt__'] = __lt__
+    del __lt__
+
+    def __ge__(self, other):
+        raise TypeError("unorderable types: %s() >= %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__ge__'] = __ge__
+    del __ge__
+
+    def __gt__(self, other):
+        raise TypeError("unorderable types: %s() > %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__gt__'] = __gt__
+    del __gt__
+    
+
+def __eq__(self, other):
+    if type(other) is self.__class__:
+        return self is other
+    return NotImplemented
+temp_enum_dict['__eq__'] = __eq__
+del __eq__
+
+def __ne__(self, other):
+    if type(other) is self.__class__:
+        return self is not other
+    return NotImplemented
+temp_enum_dict['__ne__'] = __ne__
+del __ne__
+
+def __hash__(self):
+    return hash(self._name_)
+temp_enum_dict['__hash__'] = __hash__
+del __hash__
+
+def __reduce_ex__(self, proto):
+    return self.__class__, (self._value_, )
+temp_enum_dict['__reduce_ex__'] = __reduce_ex__
+del __reduce_ex__
+
+# _RouteClassAttributeToGetattr is used to provide access to the `name`
+# and `value` properties of enum members while keeping some measure of
+# protection from modification, while still allowing for an enumeration
+# to have members named `name` and `value`.  This works because enumeration
+# members are not set directly on the enum class -- __getattr__ is
+# used to look them up.
+
+@_RouteClassAttributeToGetattr
+def name(self):
+    return self._name_
+temp_enum_dict['name'] = name
+del name
+
+@_RouteClassAttributeToGetattr
+def value(self):
+    return self._value_
+temp_enum_dict['value'] = value
+del value
+
+Enum = EnumMeta('Enum', (object, ), temp_enum_dict)
+del temp_enum_dict
+
+# Enum has now been created
+###########################
+
+class IntEnum(int, Enum):
+    """Enum where members are also (and must be) ints"""
+
+
+def unique(enumeration):
+    """Class decorator that ensures only unique members exist in an enumeration."""
+    duplicates = []
+    for name, member in enumeration.__members__.items():
+        if name != member.name:
+            duplicates.append((name, member.name))
+    if duplicates:
+        duplicate_names = ', '.join(
+                ["%s -> %s" % (alias, name) for (alias, name) in duplicates]
+                )
+        raise ValueError('duplicate names found in %r: %s' %
+                (enumeration, duplicate_names)
+                )
+    return enumeration

lib/enum/doc/enum.rst

@@ -0,0 +1,725 @@
+``enum`` --- support for enumerations
+========================================
+
+.. :synopsis: enumerations are sets of symbolic names bound to unique, constant
+  values.
+.. :moduleauthor:: Ethan Furman <ethan@stoneleaf.us>
+.. :sectionauthor:: Barry Warsaw <barry@python.org>,
+.. :sectionauthor:: Eli Bendersky <eliben@gmail.com>,
+.. :sectionauthor:: Ethan Furman <ethan@stoneleaf.us>
+
+----------------
+
+An enumeration is a set of symbolic names (members) bound to unique, constant
+values.  Within an enumeration, the members can be compared by identity, and
+the enumeration itself can be iterated over.
+
+
+Module Contents
+---------------
+
+This module defines two enumeration classes that can be used to define unique
+sets of names and values: ``Enum`` and ``IntEnum``.  It also defines
+one decorator, ``unique``.
+
+``Enum``
+
+Base class for creating enumerated constants.  See section `Functional API`_
+for an alternate construction syntax.
+
+``IntEnum``
+
+Base class for creating enumerated constants that are also subclasses of ``int``.
+
+``unique``
+
+Enum class decorator that ensures only one name is bound to any one value.
+
+
+Creating an Enum
+----------------
+
+Enumerations are created using the ``class`` syntax, which makes them
+easy to read and write.  An alternative creation method is described in
+`Functional API`_.  To define an enumeration, subclass ``Enum`` as
+follows::
+
+    >>> from enum import Enum
+    >>> class Color(Enum):
+    ...     red = 1
+    ...     green = 2
+    ...     blue = 3
+
+Note: Nomenclature
+
+  - The class ``Color`` is an *enumeration* (or *enum*)
+  - The attributes ``Color.red``, ``Color.green``, etc., are
+    *enumeration members* (or *enum members*).
+  - The enum members have *names* and *values* (the name of
+    ``Color.red`` is ``red``, the value of ``Color.blue`` is
+    ``3``, etc.)
+    
+Note:
+
+    Even though we use the ``class`` syntax to create Enums, Enums
+    are not normal Python classes.  See `How are Enums different?`_ for
+    more details.
+
+Enumeration members have human readable string representations::
+
+    >>> print(Color.red)
+    Color.red
+
+...while their ``repr`` has more information::
+
+    >>> print(repr(Color.red))
+    <Color.red: 1>
+
+The *type* of an enumeration member is the enumeration it belongs to::
+
+    >>> type(Color.red)
+    <enum 'Color'>
+    >>> isinstance(Color.green, Color)
+    True
+    >>>
+
+Enum members also have a property that contains just their item name::
+
+    >>> print(Color.red.name)
+    red
+
+Enumerations support iteration.  In Python 3.x definition order is used; in
+Python 2.x the definition order is not available, but class attribute
+``__order__`` is supported;  otherwise, value order is used::
+
+    >>> class Shake(Enum):
+    ...   __order__ = 'vanilla chocolate cookies mint'  # only needed in 2.x
+    ...   vanilla = 7
+    ...   chocolate = 4
+    ...   cookies = 9
+    ...   mint = 3
+    ...
+    >>> for shake in Shake:
+    ...   print(shake)
+    ...
+    Shake.vanilla
+    Shake.chocolate
+    Shake.cookies
+    Shake.mint
+
+The ``__order__`` attribute is always removed, and in 3.x it is also ignored
+(order is definition order); however, in the stdlib version it will be ignored
+but not removed.
+
+Enumeration members are hashable, so they can be used in dictionaries and sets::
+
+    >>> apples = {}
+    >>> apples[Color.red] = 'red delicious'
+    >>> apples[Color.green] = 'granny smith'
+    >>> apples == {Color.red: 'red delicious', Color.green: 'granny smith'}
+    True
+
+
+Programmatic access to enumeration members and their attributes
+---------------------------------------------------------------
+
+Sometimes it's useful to access members in enumerations programmatically (i.e.
+situations where ``Color.red`` won't do because the exact color is not known
+at program-writing time).  ``Enum`` allows such access::
+
+    >>> Color(1)
+    <Color.red: 1>
+    >>> Color(3)
+    <Color.blue: 3>
+
+If you want to access enum members by *name*, use item access::
+
+    >>> Color['red']
+    <Color.red: 1>
+    >>> Color['green']
+    <Color.green: 2>
+
+If have an enum member and need its ``name`` or ``value``::
+
+    >>> member = Color.red
+    >>> member.name
+    'red'
+    >>> member.value
+    1
+
+
+Duplicating enum members and values
+-----------------------------------
+
+Having two enum members (or any other attribute) with the same name is invalid;
+in Python 3.x this would raise an error, but in Python 2.x the second member
+simply overwrites the first::
+
+    >>> # python 2.x
+    >>> class Shape(Enum):
+    ...   square = 2
+    ...   square = 3
+    ...
+    >>> Shape.square
+    <Shape.square: 3>
+
+    >>> # python 3.x
+    >>> class Shape(Enum):
+    ...   square = 2
+    ...   square = 3
+    Traceback (most recent call last):
+    ...
+    TypeError: Attempted to reuse key: 'square'
+
+However, two enum members are allowed to have the same value.  Given two members
+A and B with the same value (and A defined first), B is an alias to A.  By-value
+lookup of the value of A and B will return A.  By-name lookup of B will also
+return A::
+
+    >>> class Shape(Enum):
+    ...   __order__ = 'square diamond circle alias_for_square'  # only needed in 2.x
+    ...   square = 2
+    ...   diamond = 1
+    ...   circle = 3
+    ...   alias_for_square = 2
+    ...
+    >>> Shape.square
+    <Shape.square: 2>
+    >>> Shape.alias_for_square
+    <Shape.square: 2>
+    >>> Shape(2)
+    <Shape.square: 2>
+
+
+Allowing aliases is not always desirable.  ``unique`` can be used to ensure
+that none exist in a particular enumeration::
+
+    >>> from enum import unique
+    >>> @unique
+    ... class Mistake(Enum):
+    ...   __order__ = 'one two three four'  # only needed in 2.x
+    ...   one = 1
+    ...   two = 2
+    ...   three = 3
+    ...   four = 3
+    Traceback (most recent call last):
+    ...
+    ValueError: duplicate names found in <enum 'Mistake'>: four -> three
+
+Iterating over the members of an enum does not provide the aliases::
+
+    >>> list(Shape)
+    [<Shape.square: 2>, <Shape.diamond: 1>, <Shape.circle: 3>]
+
+The special attribute ``__members__`` is a dictionary mapping names to members.
+It includes all names defined in the enumeration, including the aliases::
+
+    >>> for name, member in sorted(Shape.__members__.items()):
+    ...   name, member
+    ...
+    ('alias_for_square', <Shape.square: 2>)
+    ('circle', <Shape.circle: 3>)
+    ('diamond', <Shape.diamond: 1>)
+    ('square', <Shape.square: 2>)
+
+The ``__members__`` attribute can be used for detailed programmatic access to
+the enumeration members.  For example, finding all the aliases::
+
+    >>> [name for name, member in Shape.__members__.items() if member.name != name]
+    ['alias_for_square']
+
+Comparisons
+-----------
+
+Enumeration members are compared by identity::
+
+    >>> Color.red is Color.red
+    True
+    >>> Color.red is Color.blue
+    False
+    >>> Color.red is not Color.blue
+    True
+
+Ordered comparisons between enumeration values are *not* supported.  Enum
+members are not integers (but see `IntEnum`_ below)::
+
+    >>> Color.red < Color.blue
+    Traceback (most recent call last):
+      File "<stdin>", line 1, in <module>
+    TypeError: unorderable types: Color() < Color()
+
+.. warning::
+
+    In Python 2 *everything* is ordered, even though the ordering may not
+    make sense.  If you want your enumerations to have a sensible ordering
+    check out the `OrderedEnum`_ recipe below.
+
+
+Equality comparisons are defined though::
+
+    >>> Color.blue == Color.red
+    False
+    >>> Color.blue != Color.red
+    True
+    >>> Color.blue == Color.blue
+    True
+
+Comparisons against non-enumeration values will always compare not equal
+(again, ``IntEnum`` was explicitly designed to behave differently, see
+below)::
+
+    >>> Color.blue == 2
+    False
+
+
+Allowed members and attributes of enumerations
+----------------------------------------------
+
+The examples above use integers for enumeration values.  Using integers is
+short and handy (and provided by default by the `Functional API`_), but not
+strictly enforced.  In the vast majority of use-cases, one doesn't care what
+the actual value of an enumeration is.  But if the value *is* important,
+enumerations can have arbitrary values.
+
+Enumerations are Python classes, and can have methods and special methods as
+usual.  If we have this enumeration::
+
+    >>> class Mood(Enum):
+    ...   funky = 1
+    ...   happy = 3
+    ... 
+    ...   def describe(self):
+    ...     # self is the member here
+    ...     return self.name, self.value
+    ... 
+    ...   def __str__(self):
+    ...     return 'my custom str! {0}'.format(self.value)
+    ... 
+    ...   @classmethod
+    ...   def favorite_mood(cls):
+    ...     # cls here is the enumeration
+    ...     return cls.happy
+
+Then::
+
+    >>> Mood.favorite_mood()
+    <Mood.happy: 3>
+    >>> Mood.happy.describe()
+    ('happy', 3)
+    >>> str(Mood.funky)
+    'my custom str! 1'
+
+The rules for what is allowed are as follows: _sunder_ names (starting and
+ending with a single underscore) are reserved by enum and cannot be used;
+all other attributes defined within an enumeration will become members of this
+enumeration, with the exception of *__dunder__* names and descriptors (methods
+are also descriptors).
+
+Note:
+
+    If your enumeration defines ``__new__`` and/or ``__init__`` then
+    whatever value(s) were given to the enum member will be passed into
+    those methods.  See `Planet`_ for an example.
+
+
+Restricted subclassing of enumerations
+--------------------------------------
+
+Subclassing an enumeration is allowed only if the enumeration does not define
+any members.  So this is forbidden::
+
+    >>> class MoreColor(Color):
+    ...   pink = 17
+    Traceback (most recent call last):
+    ...
+    TypeError: Cannot extend enumerations
+
+But this is allowed::
+
+    >>> class Foo(Enum):
+    ...   def some_behavior(self):
+    ...     pass
+    ...
+    >>> class Bar(Foo):
+    ...   happy = 1
+    ...   sad = 2
+    ...
+
+Allowing subclassing of enums that define members would lead to a violation of
+some important invariants of types and instances.  On the other hand, it makes
+sense to allow sharing some common behavior between a group of enumerations.
+(See `OrderedEnum`_ for an example.)
+
+
+Pickling
+--------
+
+Enumerations can be pickled and unpickled::
+
+    >>> from enum.test_enum import Fruit
+    >>> from pickle import dumps, loads
+    >>> Fruit.tomato is loads(dumps(Fruit.tomato, 2))
+    True
+
+The usual restrictions for pickling apply: picklable enums must be defined in
+the top level of a module, since unpickling requires them to be importable
+from that module.
+
+Note:
+
+    With pickle protocol version 4 (introduced in Python 3.4) it is possible
+    to easily pickle enums nested in other classes.
+
+
+
+Functional API
+--------------
+
+The ``Enum`` class is callable, providing the following functional API::
+
+    >>> Animal = Enum('Animal', 'ant bee cat dog')
+    >>> Animal
+    <enum 'Animal'>
+    >>> Animal.ant
+    <Animal.ant: 1>
+    >>> Animal.ant.value
+    1
+    >>> list(Animal)
+    [<Animal.ant: 1>, <Animal.bee: 2>, <Animal.cat: 3>, <Animal.dog: 4>]
+
+The semantics of this API resemble ``namedtuple``. The first argument
+of the call to ``Enum`` is the name of the enumeration. 
+
+The second argument is the *source* of enumeration member names.  It can be a
+whitespace-separated string of names, a sequence of names, a sequence of
+2-tuples with key/value pairs, or a mapping (e.g. dictionary) of names to
+values.  The last two options enable assigning arbitrary values to
+enumerations; the others auto-assign increasing integers starting with 1.  A
+new class derived from ``Enum`` is returned.  In other words, the above
+assignment to ``Animal`` is equivalent to::
+
+    >>> class Animals(Enum):
+    ...   ant = 1
+    ...   bee = 2
+    ...   cat = 3
+    ...   dog = 4
+
+Pickling enums created with the functional API can be tricky as frame stack
+implementation details are used to try and figure out which module the
+enumeration is being created in (e.g. it will fail if you use a utility
+function in separate module, and also may not work on IronPython or Jython).
+The solution is to specify the module name explicitly as follows::
+
+    >>> Animals = Enum('Animals', 'ant bee cat dog', module=__name__)
+
+Derived Enumerations
+--------------------
+
+IntEnum
+^^^^^^^
+
+A variation of ``Enum`` is provided which is also a subclass of
+``int``.  Members of an ``IntEnum`` can be compared to integers;
+by extension, integer enumerations of different types can also be compared
+to each other::
+
+    >>> from enum import IntEnum
+    >>> class Shape(IntEnum):
+    ...   circle = 1
+    ...   square = 2
+    ...
+    >>> class Request(IntEnum):
+    ...   post = 1
+    ...   get = 2
+    ...
+    >>> Shape == 1
+    False
+    >>> Shape.circle == 1
+    True
+    >>> Shape.circle == Request.post
+    True
+
+However, they still can't be compared to standard ``Enum`` enumerations::
+
+    >>> class Shape(IntEnum):
+    ...   circle = 1
+    ...   square = 2
+    ...
+    >>> class Color(Enum):
+    ...   red = 1
+    ...   green = 2
+    ...
+    >>> Shape.circle == Color.red
+    False
+
+``IntEnum`` values behave like integers in other ways you'd expect::
+
+    >>> int(Shape.circle)
+    1
+    >>> ['a', 'b', 'c'][Shape.circle]
+    'b'
+    >>> [i for i in range(Shape.square)]
+    [0, 1]
+
+For the vast majority of code, ``Enum`` is strongly recommended,
+since ``IntEnum`` breaks some semantic promises of an enumeration (by
+being comparable to integers, and thus by transitivity to other
+unrelated enumerations).  It should be used only in special cases where
+there's no other choice; for example, when integer constants are
+replaced with enumerations and backwards compatibility is required with code
+that still expects integers.
+
+
+Others
+^^^^^^
+
+While ``IntEnum`` is part of the ``enum`` module, it would be very
+simple to implement independently::
+
+    class IntEnum(int, Enum):
+        pass
+
+This demonstrates how similar derived enumerations can be defined; for example
+a ``StrEnum`` that mixes in ``str`` instead of ``int``.
+
+Some rules:
+
+1. When subclassing ``Enum``, mix-in types must appear before
+   ``Enum`` itself in the sequence of bases, as in the ``IntEnum``
+   example above.
+2. While ``Enum`` can have members of any type, once you mix in an
+   additional type, all the members must have values of that type, e.g.
+   ``int`` above.  This restriction does not apply to mix-ins which only
+   add methods and don't specify another data type such as ``int`` or
+   ``str``.
+3. When another data type is mixed in, the ``value`` attribute is *not the
+   same* as the enum member itself, although it is equivalant and will compare
+   equal.
+4. %-style formatting:  ``%s`` and ``%r`` call ``Enum``'s ``__str__`` and
+   ``__repr__`` respectively; other codes (such as ``%i`` or ``%h`` for
+   IntEnum) treat the enum member as its mixed-in type.
+
+   Note: Prior to Python 3.4 there is a bug in ``str``'s %-formatting: ``int``
+   subclasses are printed as strings and not numbers when the ``%d``, ``%i``,
+   or ``%u`` codes are used.
+5. ``str.__format__`` (or ``format``) will use the mixed-in
+   type's ``__format__``.  If the ``Enum``'s ``str`` or
+   ``repr`` is desired use the ``!s`` or ``!r`` ``str`` format codes.
+
+
+Decorators
+----------
+
+unique
+^^^^^^
+
+A ``class`` decorator specifically for enumerations.  It searches an
+enumeration's ``__members__`` gathering any aliases it finds; if any are
+found ``ValueError`` is raised with the details::
+
+    >>> @unique
+    ... class NoDupes(Enum):
+    ...    first = 'one'
+    ...    second = 'two'
+    ...    third = 'two'
+    Traceback (most recent call last):
+    ...
+    ValueError: duplicate names found in <enum 'NoDupes'>: third -> second
+
+
+Interesting examples
+--------------------
+
+While ``Enum`` and ``IntEnum`` are expected to cover the majority of
+use-cases, they cannot cover them all.  Here are recipes for some different
+types of enumerations that can be used directly, or as examples for creating
+one's own.
+
+
+AutoNumber
+^^^^^^^^^^
+
+Avoids having to specify the value for each enumeration member::
+
+    >>> class AutoNumber(Enum):
+    ...     def __new__(cls):
+    ...         value = len(cls.__members__) + 1
+    ...         obj = object.__new__(cls)
+    ...         obj._value_ = value
+    ...         return obj
+    ...
+    >>> class Color(AutoNumber):
+    ...     __order__ = "red green blue"  # only needed in 2.x
+    ...     red = ()
+    ...     green = ()
+    ...     blue = ()
+    ...
+    >>> Color.green.value == 2
+    True
+
+Note:
+
+    The `__new__` method, if defined, is used during creation of the Enum
+    members; it is then replaced by Enum's `__new__` which is used after
+    class creation for lookup of existing members.  Due to the way Enums are
+    supposed to behave, there is no way to customize Enum's `__new__`.
+
+
+UniqueEnum
+^^^^^^^^^^
+
+Raises an error if a duplicate member name is found instead of creating an
+alias::
+
+    >>> class UniqueEnum(Enum):
+    ...     def __init__(self, *args):
+    ...         cls = self.__class__
+    ...         if any(self.value == e.value for e in cls):
+    ...             a = self.name
+    ...             e = cls(self.value).name
+    ...             raise ValueError(
+    ...                     "aliases not allowed in UniqueEnum:  %r --> %r"
+    ...                     % (a, e))
+    ... 
+    >>> class Color(UniqueEnum):
+    ...     red = 1
+    ...     green = 2
+    ...     blue = 3
+    ...     grene = 2
+    Traceback (most recent call last):
+    ...
+    ValueError: aliases not allowed in UniqueEnum:  'grene' --> 'green'
+    
+
+OrderedEnum
+^^^^^^^^^^^
+
+An ordered enumeration that is not based on ``IntEnum`` and so maintains
+the normal ``Enum`` invariants (such as not being comparable to other
+enumerations)::
+
+    >>> class OrderedEnum(Enum):
+    ...     def __ge__(self, other):
+    ...         if self.__class__ is other.__class__:
+    ...             return self._value_ >= other._value_
+    ...         return NotImplemented
+    ...     def __gt__(self, other):
+    ...         if self.__class__ is other.__class__:
+    ...             return self._value_ > other._value_
+    ...         return NotImplemented
+    ...     def __le__(self, other):
+    ...         if self.__class__ is other.__class__:
+    ...             return self._value_ <= other._value_
+    ...         return NotImplemented
+    ...     def __lt__(self, other):
+    ...         if self.__class__ is other.__class__:
+    ...             return self._value_ < other._value_
+    ...         return NotImplemented
+    ...
+    >>> class Grade(OrderedEnum):
+    ...     __ordered__ = 'A B C D F'
+    ...     A = 5
+    ...     B = 4
+    ...     C = 3
+    ...     D = 2
+    ...     F = 1
+    ...
+    >>> Grade.C < Grade.A
+    True
+
+
+Planet
+^^^^^^
+
+If ``__new__`` or ``__init__`` is defined the value of the enum member
+will be passed to those methods::
+
+    >>> class Planet(Enum):
+    ...     MERCURY = (3.303e+23, 2.4397e6)
+    ...     VENUS   = (4.869e+24, 6.0518e6)
+    ...     EARTH   = (5.976e+24, 6.37814e6)
+    ...     MARS    = (6.421e+23, 3.3972e6)
+    ...     JUPITER = (1.9e+27,   7.1492e7)
+    ...     SATURN  = (5.688e+26, 6.0268e7)
+    ...     URANUS  = (8.686e+25, 2.5559e7)
+    ...     NEPTUNE = (1.024e+26, 2.4746e7)
+    ...     def __init__(self, mass, radius):
+    ...         self.mass = mass       # in kilograms
+    ...         self.radius = radius   # in meters
+    ...     @property
+    ...     def surface_gravity(self):
+    ...         # universal gravitational constant  (m3 kg-1 s-2)
+    ...         G = 6.67300E-11
+    ...         return G * self.mass / (self.radius * self.radius)
+    ... 
+    >>> Planet.EARTH.value
+    (5.976e+24, 6378140.0)
+    >>> Planet.EARTH.surface_gravity
+    9.802652743337129
+
+
+How are Enums different?
+------------------------
+
+Enums have a custom metaclass that affects many aspects of both derived Enum
+classes and their instances (members).
+
+
+Enum Classes
+^^^^^^^^^^^^
+
+The ``EnumMeta`` metaclass is responsible for providing the
+``__contains__``, ``__dir__``, ``__iter__`` and other methods that
+allow one to do things with an ``Enum`` class that fail on a typical
+class, such as ``list(Color)`` or ``some_var in Color``.  ``EnumMeta`` is
+responsible for ensuring that various other methods on the final ``Enum``
+class are correct (such as ``__new__``, ``__getnewargs__``,
+``__str__`` and ``__repr__``)
+
+
+Enum Members (aka instances)
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The most interesting thing about Enum members is that they are singletons.
+``EnumMeta`` creates them all while it is creating the ``Enum``
+class itself, and then puts a custom ``__new__`` in place to ensure
+that no new ones are ever instantiated by returning only the existing
+member instances.
+
+
+Finer Points
+^^^^^^^^^^^^
+
+Enum members are instances of an Enum class, and even though they are
+accessible as ``EnumClass.member``, they are not accessible directly from
+the member::
+
+    >>> Color.red
+    <Color.red: 1>
+    >>> Color.red.blue
+    Traceback (most recent call last):
+    ...
+    AttributeError: 'Color' object has no attribute 'blue'
+
+Likewise, ``__members__`` is only available on the class.
+
+In Python 3.x ``__members__`` is always an ``OrderedDict``, with the order being
+the definition order.  In Python 2.7 ``__members__`` is an ``OrderedDict`` if
+``__order__`` was specified, and a plain ``dict`` otherwise.  In all other Python
+2.x versions ``__members__`` is a plain ``dict`` even if ``__order__`` was specified
+as the ``OrderedDict`` type didn't exist yet.
+
+If you give your ``Enum`` subclass extra methods, like the `Planet`_
+class above, those methods will show up in a `dir` of the member,
+but not of the class::
+
+    >>> dir(Planet)
+    ['EARTH', 'JUPITER', 'MARS', 'MERCURY', 'NEPTUNE', 'SATURN', 'URANUS',
+    'VENUS', '__class__', '__doc__', '__members__', '__module__']
+    >>> dir(Planet.EARTH)
+    ['__class__', '__doc__', '__module__', 'name', 'surface_gravity', 'value']
+
+A ``__new__`` method will only be used for the creation of the
+``Enum`` members -- after that it is replaced.  This means if you wish to
+change how ``Enum`` members are looked up you either have to write a
+helper function or a ``classmethod``.

lib/enum/enum.py

@@ -0,0 +1,790 @@
+"""Python Enumerations"""
+
+import sys as _sys
+
+__all__ = ['Enum', 'IntEnum', 'unique']
+
+version = 1, 0, 4
+
+pyver = float('%s.%s' % _sys.version_info[:2])
+
+try:
+    any
+except NameError:
+    def any(iterable):
+        for element in iterable:
+            if element:
+                return True
+        return False
+
+try:
+    from collections import OrderedDict
+except ImportError:
+    OrderedDict = None
+
+try:
+    basestring
+except NameError:
+    # In Python 2 basestring is the ancestor of both str and unicode
+    # in Python 3 it's just str, but was missing in 3.1
+    basestring = str
+
+try:
+    unicode
+except NameError:
+    # In Python 3 unicode no longer exists (it's just str)
+    unicode = str
+
+class _RouteClassAttributeToGetattr(object):
+    """Route attribute access on a class to __getattr__.
+
+    This is a descriptor, used to define attributes that act differently when
+    accessed through an instance and through a class.  Instance access remains
+    normal, but access to an attribute through a class will be routed to the
+    class's __getattr__ method; this is done by raising AttributeError.
+
+    """
+    def __init__(self, fget=None):
+        self.fget = fget
+
+    def __get__(self, instance, ownerclass=None):
+        if instance is None:
+            raise AttributeError()
+        return self.fget(instance)
+
+    def __set__(self, instance, value):
+        raise AttributeError("can't set attribute")
+
+    def __delete__(self, instance):
+        raise AttributeError("can't delete attribute")
+
+
+def _is_descriptor(obj):
+    """Returns True if obj is a descriptor, False otherwise."""
+    return (
+            hasattr(obj, '__get__') or
+            hasattr(obj, '__set__') or
+            hasattr(obj, '__delete__'))
+
+
+def _is_dunder(name):
+    """Returns True if a __dunder__ name, False otherwise."""
+    return (name[:2] == name[-2:] == '__' and
+            name[2:3] != '_' and
+            name[-3:-2] != '_' and
+            len(name) > 4)
+
+
+def _is_sunder(name):
+    """Returns True if a _sunder_ name, False otherwise."""
+    return (name[0] == name[-1] == '_' and 
+            name[1:2] != '_' and
+            name[-2:-1] != '_' and
+            len(name) > 2)
+
+
+def _make_class_unpicklable(cls):
+    """Make the given class un-picklable."""
+    def _break_on_call_reduce(self, protocol=None):
+        raise TypeError('%r cannot be pickled' % self)
+    cls.__reduce_ex__ = _break_on_call_reduce
+    cls.__module__ = '<unknown>'
+
+
+class _EnumDict(dict):
+    """Track enum member order and ensure member names are not reused.
+
+    EnumMeta will use the names found in self._member_names as the
+    enumeration member names.
+
+    """
+    def __init__(self):
+        super(_EnumDict, self).__init__()
+        self._member_names = []
+
+    def __setitem__(self, key, value):
+        """Changes anything not dundered or not a descriptor.
+
+        If a descriptor is added with the same name as an enum member, the name
+        is removed from _member_names (this may leave a hole in the numerical
+        sequence of values).
+
+        If an enum member name is used twice, an error is raised; duplicate
+        values are not checked for.
+
+        Single underscore (sunder) names are reserved.
+
+        Note:   in 3.x __order__ is simply discarded as a not necessary piece
+                leftover from 2.x
+
+        """
+        if pyver >= 3.0 and key == '__order__':
+                return
+        if _is_sunder(key):
+            raise ValueError('_names_ are reserved for future Enum use')
+        elif _is_dunder(key):
+            pass
+        elif key in self._member_names:
+            # descriptor overwriting an enum?
+            raise TypeError('Attempted to reuse key: %r' % key)
+        elif not _is_descriptor(value):
+            if key in self:
+                # enum overwriting a descriptor?
+                raise TypeError('Key already defined as: %r' % self[key])
+            self._member_names.append(key)
+        super(_EnumDict, self).__setitem__(key, value)
+
+
+# Dummy value for Enum as EnumMeta explicity checks for it, but of course until
+# EnumMeta finishes running the first time the Enum class doesn't exist.  This
+# is also why there are checks in EnumMeta like `if Enum is not None`
+Enum = None
+
+
+class EnumMeta(type):
+    """Metaclass for Enum"""
+    @classmethod
+    def __prepare__(metacls, cls, bases):
+        return _EnumDict()
+
+    def __new__(metacls, cls, bases, classdict):
+        # an Enum class is final once enumeration items have been defined; it
+        # cannot be mixed with other types (int, float, etc.) if it has an
+        # inherited __new__ unless a new __new__ is defined (or the resulting
+        # class will fail).
+        if type(classdict) is dict:
+            original_dict = classdict
+            classdict = _EnumDict()
+            for k, v in original_dict.items():
+                classdict[k] = v
+
+        member_type, first_enum = metacls._get_mixins_(bases)
+        __new__, save_new, use_args = metacls._find_new_(classdict, member_type,
+                                                        first_enum)
+        # save enum items into separate mapping so they don't get baked into
+        # the new class
+        members = dict((k, classdict[k]) for k in classdict._member_names)
+        for name in classdict._member_names:
+            del classdict[name]
+
+        # py2 support for definition order
+        __order__ = classdict.get('__order__')
+        if __order__ is None:
+            if pyver < 3.0:
+                try:
+                    __order__ = [name for (name, value) in sorted(members.items(), key=lambda item: item[1])]
+                except TypeError:
+                    __order__ = [name for name in sorted(members.keys())]
+            else:
+                __order__ = classdict._member_names
+        else:
+            del classdict['__order__']
+            if pyver < 3.0:
+                __order__ = __order__.replace(',', ' ').split()
+                aliases = [name for name in members if name not in __order__]
+                __order__ += aliases
+
+        # check for illegal enum names (any others?)
+        invalid_names = set(members) & set(['mro'])
+        if invalid_names:
+            raise ValueError('Invalid enum member name(s): %s' % (
+                ', '.join(invalid_names), ))
+
+        # create our new Enum type
+        enum_class = super(EnumMeta, metacls).__new__(metacls, cls, bases, classdict)
+        enum_class._member_names_ = []               # names in random order
+        if OrderedDict is not None:
+            enum_class._member_map_ = OrderedDict()
+        else:
+            enum_class._member_map_ = {}             # name->value map
+        enum_class._member_type_ = member_type
+
+        # Reverse value->name map for hashable values.
+        enum_class._value2member_map_ = {}
+
+        # instantiate them, checking for duplicates as we go
+        # we instantiate first instead of checking for duplicates first in case
+        # a custom __new__ is doing something funky with the values -- such as
+        # auto-numbering ;)
+        if __new__ is None:
+            __new__ = enum_class.__new__
+        for member_name in __order__:
+            value = members[member_name]
+            if not isinstance(value, tuple):
+                args = (value, )
+            else:
+                args = value
+            if member_type is tuple:   # special case for tuple enums
+                args = (args, )     # wrap it one more time
+            if not use_args or not args:
+                enum_member = __new__(enum_class)
+                if not hasattr(enum_member, '_value_'):
+                    enum_member._value_ = value
+            else:
+                enum_member = __new__(enum_class, *args)
+                if not hasattr(enum_member, '_value_'):
+                    enum_member._value_ = member_type(*args)
+            value = enum_member._value_
+            enum_member._name_ = member_name
+            enum_member.__objclass__ = enum_class
+            enum_member.__init__(*args)
+            # If another member with the same value was already defined, the
+            # new member becomes an alias to the existing one.
+            for name, canonical_member in enum_class._member_map_.items():
+                if canonical_member.value == enum_member._value_:
+                    enum_member = canonical_member
+                    break
+            else:
+                # Aliases don't appear in member names (only in __members__).
+                enum_class._member_names_.append(member_name)
+            enum_class._member_map_[member_name] = enum_member
+            try:
+                # This may fail if value is not hashable. We can't add the value
+                # to the map, and by-value lookups for this value will be
+                # linear.
+                enum_class._value2member_map_[value] = enum_member
+            except TypeError:
+                pass
+
+
+        # If a custom type is mixed into the Enum, and it does not know how
+        # to pickle itself, pickle.dumps will succeed but pickle.loads will
+        # fail.  Rather than have the error show up later and possibly far
+        # from the source, sabotage the pickle protocol for this class so
+        # that pickle.dumps also fails.
+        #
+        # However, if the new class implements its own __reduce_ex__, do not
+        # sabotage -- it's on them to make sure it works correctly.  We use
+        # __reduce_ex__ instead of any of the others as it is preferred by
+        # pickle over __reduce__, and it handles all pickle protocols.
+        unpicklable = False
+        if '__reduce_ex__' not in classdict:
+            if member_type is not object:
+                methods = ('__getnewargs_ex__', '__getnewargs__',
+                        '__reduce_ex__', '__reduce__')
+                if not any(m in member_type.__dict__ for m in methods):
+                    _make_class_unpicklable(enum_class)
+                    unpicklable = True
+
+
+        # double check that repr and friends are not the mixin's or various
+        # things break (such as pickle)
+        for name in ('__repr__', '__str__', '__format__', '__reduce_ex__'):
+            class_method = getattr(enum_class, name)
+            obj_method = getattr(member_type, name, None)
+            enum_method = getattr(first_enum, name, None)
+            if name not in classdict and class_method is not enum_method:
+                if name == '__reduce_ex__' and unpicklable:
+                    continue
+                setattr(enum_class, name, enum_method)
+
+        # method resolution and int's are not playing nice
+        # Python's less than 2.6 use __cmp__
+
+        if pyver < 2.6:
+
+            if issubclass(enum_class, int):
+                setattr(enum_class, '__cmp__', getattr(int, '__cmp__'))
+
+        elif pyver < 3.0:
+
+            if issubclass(enum_class, int):
+                for method in (
+                        '__le__',
+                        '__lt__',
+                        '__gt__',
+                        '__ge__',
+                        '__eq__',
+                        '__ne__',
+                        '__hash__',
+                        ):
+                    setattr(enum_class, method, getattr(int, method))
+
+        # replace any other __new__ with our own (as long as Enum is not None,
+        # anyway) -- again, this is to support pickle
+        if Enum is not None:
+            # if the user defined their own __new__, save it before it gets
+            # clobbered in case they subclass later
+            if save_new:
+                setattr(enum_class, '__member_new__', enum_class.__dict__['__new__'])
+            setattr(enum_class, '__new__', Enum.__dict__['__new__'])
+        return enum_class
+
+    def __call__(cls, value, names=None, module=None, type=None):
+        """Either returns an existing member, or creates a new enum class.
+
+        This method is used both when an enum class is given a value to match
+        to an enumeration member (i.e. Color(3)) and for the functional API
+        (i.e. Color = Enum('Color', names='red green blue')).
+
+        When used for the functional API: `module`, if set, will be stored in
+        the new class' __module__ attribute; `type`, if set, will be mixed in
+        as the first base class.
+
+        Note: if `module` is not set this routine will attempt to discover the
+        calling module by walking the frame stack; if this is unsuccessful
+        the resulting class will not be pickleable.
+
+        """
+        if names is None:  # simple value lookup
+            return cls.__new__(cls, value)
+        # otherwise, functional API: we're creating a new Enum type
+        return cls._create_(value, names, module=module, type=type)
+
+    def __contains__(cls, member):
+        return isinstance(member, cls) and member.name in cls._member_map_
+
+    def __delattr__(cls, attr):
+        # nicer error message when someone tries to delete an attribute
+        # (see issue19025).
+        if attr in cls._member_map_:
+            raise AttributeError(
+                    "%s: cannot delete Enum member." % cls.__name__)
+        super(EnumMeta, cls).__delattr__(attr)
+
+    def __dir__(self):
+        return (['__class__', '__doc__', '__members__', '__module__'] +
+                self._member_names_)
+
+    @property
+    def __members__(cls):
+        """Returns a mapping of member name->value.
+
+        This mapping lists all enum members, including aliases. Note that this
+        is a copy of the internal mapping.
+
+        """
+        return cls._member_map_.copy()
+
+    def __getattr__(cls, name):
+        """Return the enum member matching `name`
+
+        We use __getattr__ instead of descriptors or inserting into the enum
+        class' __dict__ in order to support `name` and `value` being both
+        properties for enum members (which live in the class' __dict__) and
+        enum members themselves.
+
+        """
+        if _is_dunder(name):
+            raise AttributeError(name)
+        try:
+            return cls._member_map_[name]
+        except KeyError:
+            raise AttributeError(name)
+
+    def __getitem__(cls, name):
+        return cls._member_map_[name]
+
+    def __iter__(cls):
+        return (cls._member_map_[name] for name in cls._member_names_)
+
+    def __reversed__(cls):
+        return (cls._member_map_[name] for name in reversed(cls._member_names_))
+
+    def __len__(cls):
+        return len(cls._member_names_)
+
+    def __repr__(cls):
+        return "<enum %r>" % cls.__name__
+
+    def __setattr__(cls, name, value):
+        """Block attempts to reassign Enum members.
+
+        A simple assignment to the class namespace only changes one of the
+        several possible ways to get an Enum member from the Enum class,
+        resulting in an inconsistent Enumeration.
+
+        """
+        member_map = cls.__dict__.get('_member_map_', {})
+        if name in member_map:
+            raise AttributeError('Cannot reassign members.')
+        super(EnumMeta, cls).__setattr__(name, value)
+
+    def _create_(cls, class_name, names=None, module=None, type=None):
+        """Convenience method to create a new Enum class.
+
+        `names` can be:
+
+        * A string containing member names, separated either with spaces or
+          commas.  Values are auto-numbered from 1.
+        * An iterable of member names.  Values are auto-numbered from 1.
+        * An iterable of (member name, value) pairs.
+        * A mapping of member name -> value.
+
+        """
+        if pyver < 3.0:
+            # if class_name is unicode, attempt a conversion to ASCII
+            if isinstance(class_name, unicode):
+                try:
+                    class_name = class_name.encode('ascii')
+                except UnicodeEncodeError:
+                    raise TypeError('%r is not representable in ASCII' % class_name)
+        metacls = cls.__class__
+        if type is None:
+            bases = (cls, )
+        else:
+            bases = (type, cls)
+        classdict = metacls.__prepare__(class_name, bases)
+        __order__ = []
+
+        # special processing needed for names?
+        if isinstance(names, basestring):
+            names = names.replace(',', ' ').split()
+        if isinstance(names, (tuple, list)) and isinstance(names[0], basestring):
+            names = [(e, i+1) for (i, e) in enumerate(names)]
+
+        # Here, names is either an iterable of (name, value) or a mapping.
+        for item in names:
+            if isinstance(item, basestring):
+                member_name, member_value = item, names[item]
+            else:
+                member_name, member_value = item
+            classdict[member_name] = member_value
+            __order__.append(member_name)
+        # only set __order__ in classdict if name/value was not from a mapping
+        if not isinstance(item, basestring):
+            classdict['__order__'] = ' '.join(__order__)
+        enum_class = metacls.__new__(metacls, class_name, bases, classdict)
+
+        # TODO: replace the frame hack if a blessed way to know the calling
+        # module is ever developed
+        if module is None:
+            try:
+                module = _sys._getframe(2).f_globals['__name__']
+            except (AttributeError, ValueError):
+                pass
+        if module is None:
+            _make_class_unpicklable(enum_class)
+        else:
+            enum_class.__module__ = module
+
+        return enum_class
+
+    @staticmethod
+    def _get_mixins_(bases):
+        """Returns the type for creating enum members, and the first inherited
+        enum class.
+
+        bases: the tuple of bases that was given to __new__
+
+        """
+        if not bases or Enum is None:
+            return object, Enum
+        
+
+        # double check that we are not subclassing a class with existing
+        # enumeration members; while we're at it, see if any other data
+        # type has been mixed in so we can use the correct __new__
+        member_type = first_enum = None
+        for base in bases:
+            if  (base is not Enum and
+                    issubclass(base, Enum) and
+                    base._member_names_):
+                raise TypeError("Cannot extend enumerations")
+        # base is now the last base in bases
+        if not issubclass(base, Enum):
+            raise TypeError("new enumerations must be created as "
+                    "`ClassName([mixin_type,] enum_type)`")
+
+        # get correct mix-in type (either mix-in type of Enum subclass, or
+        # first base if last base is Enum)
+        if not issubclass(bases[0], Enum):
+            member_type = bases[0]     # first data type
+            first_enum = bases[-1]  # enum type
+        else:
+            for base in bases[0].__mro__:
+                # most common: (IntEnum, int, Enum, object)
+                # possible:    (<Enum 'AutoIntEnum'>, <Enum 'IntEnum'>,
+                #               <class 'int'>, <Enum 'Enum'>,
+                #               <class 'object'>)
+                if issubclass(base, Enum):
+                    if first_enum is None:
+                        first_enum = base
+                else:
+                    if member_type is None:
+                        member_type = base
+
+        return member_type, first_enum
+
+    if pyver < 3.0:
+        @staticmethod
+        def _find_new_(classdict, member_type, first_enum):
+            """Returns the __new__ to be used for creating the enum members.
+
+            classdict: the class dictionary given to __new__
+            member_type: the data type whose __new__ will be used by default
+            first_enum: enumeration to check for an overriding __new__
+
+            """
+            # now find the correct __new__, checking to see of one was defined
+            # by the user; also check earlier enum classes in case a __new__ was
+            # saved as __member_new__
+            __new__ = classdict.get('__new__', None)
+            if __new__:
+                return None, True, True      # __new__, save_new, use_args
+
+            N__new__ = getattr(None, '__new__')
+            O__new__ = getattr(object, '__new__')
+            if Enum is None:
+                E__new__ = N__new__
+            else:
+                E__new__ = Enum.__dict__['__new__']
+            # check all possibles for __member_new__ before falling back to
+            # __new__
+            for method in ('__member_new__', '__new__'):
+                for possible in (member_type, first_enum):
+                    try:
+                        target = possible.__dict__[method]
+                    except (AttributeError, KeyError):
+                        target = getattr(possible, method, None)
+                    if target not in [
+                            None,
+                            N__new__,
+                            O__new__,
+                            E__new__,
+                            ]:
+                        if method == '__member_new__':
+                            classdict['__new__'] = target
+                            return None, False, True
+                        if isinstance(target, staticmethod):
+                            target = target.__get__(member_type)
+                        __new__ = target
+                        break
+                if __new__ is not None:
+                    break
+            else:
+                __new__ = object.__new__
+
+            # if a non-object.__new__ is used then whatever value/tuple was
+            # assigned to the enum member name will be passed to __new__ and to the
+            # new enum member's __init__
+            if __new__ is object.__new__:
+                use_args = False
+            else:
+                use_args = True
+
+            return __new__, False, use_args
+    else:
+        @staticmethod
+        def _find_new_(classdict, member_type, first_enum):
+            """Returns the __new__ to be used for creating the enum members.
+
+            classdict: the class dictionary given to __new__
+            member_type: the data type whose __new__ will be used by default
+            first_enum: enumeration to check for an overriding __new__
+
+            """
+            # now find the correct __new__, checking to see of one was defined
+            # by the user; also check earlier enum classes in case a __new__ was
+            # saved as __member_new__
+            __new__ = classdict.get('__new__', None)
+
+            # should __new__ be saved as __member_new__ later?
+            save_new = __new__ is not None
+
+            if __new__ is None:
+                # check all possibles for __member_new__ before falling back to
+                # __new__
+                for method in ('__member_new__', '__new__'):
+                    for possible in (member_type, first_enum):
+                        target = getattr(possible, method, None)
+                        if target not in (
+                                None,
+                                None.__new__,
+                                object.__new__,
+                                Enum.__new__,
+                                ):
+                            __new__ = target
+                            break
+                    if __new__ is not None:
+                        break
+                else:
+                    __new__ = object.__new__
+
+            # if a non-object.__new__ is used then whatever value/tuple was
+            # assigned to the enum member name will be passed to __new__ and to the
+            # new enum member's __init__
+            if __new__ is object.__new__:
+                use_args = False
+            else:
+                use_args = True
+
+            return __new__, save_new, use_args
+
+
+########################################################
+# In order to support Python 2 and 3 with a single
+# codebase we have to create the Enum methods separately
+# and then use the `type(name, bases, dict)` method to
+# create the class.
+########################################################
+temp_enum_dict = {}
+temp_enum_dict['__doc__'] = "Generic enumeration.\n\n    Derive from this class to define new enumerations.\n\n"
+
+def __new__(cls, value):
+    # all enum instances are actually created during class construction
+    # without calling this method; this method is called by the metaclass'
+    # __call__ (i.e. Color(3) ), and by pickle
+    if type(value) is cls:
+        # For lookups like Color(Color.red)
+        value = value.value
+        #return value
+    # by-value search for a matching enum member
+    # see if it's in the reverse mapping (for hashable values)
+    try:
+        if value in cls._value2member_map_:
+            return cls._value2member_map_[value]
+    except TypeError:
+        # not there, now do long search -- O(n) behavior
+        for member in cls._member_map_.values():
+            if member.value == value:
+                return member
+    raise ValueError("%s is not a valid %s" % (value, cls.__name__))
+temp_enum_dict['__new__'] = __new__
+del __new__
+
+def __repr__(self):
+    return "<%s.%s: %r>" % (
+            self.__class__.__name__, self._name_, self._value_)
+temp_enum_dict['__repr__'] = __repr__
+del __repr__
+
+def __str__(self):
+    return "%s.%s" % (self.__class__.__name__, self._name_)
+temp_enum_dict['__str__'] = __str__
+del __str__
+
+def __dir__(self):
+    added_behavior = [
+            m
+            for cls in self.__class__.mro()
+            for m in cls.__dict__
+            if m[0] != '_'
+            ]
+    return (['__class__', '__doc__', '__module__', ] + added_behavior)
+temp_enum_dict['__dir__'] = __dir__
+del __dir__
+
+def __format__(self, format_spec):
+    # mixed-in Enums should use the mixed-in type's __format__, otherwise
+    # we can get strange results with the Enum name showing up instead of
+    # the value
+
+    # pure Enum branch
+    if self._member_type_ is object:
+        cls = str
+        val = str(self)
+    # mix-in branch
+    else:
+        cls = self._member_type_
+        val = self.value
+    return cls.__format__(val, format_spec)
+temp_enum_dict['__format__'] = __format__
+del __format__
+
+
+####################################
+# Python's less than 2.6 use __cmp__
+
+if pyver < 2.6:
+
+    def __cmp__(self, other):
+        if type(other) is self.__class__:
+            if self is other:
+                return 0
+            return -1
+        return NotImplemented
+        raise TypeError("unorderable types: %s() and %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__cmp__'] = __cmp__
+    del __cmp__
+
+else:
+
+    def __le__(self, other):
+        raise TypeError("unorderable types: %s() <= %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__le__'] = __le__
+    del __le__
+
+    def __lt__(self, other):
+        raise TypeError("unorderable types: %s() < %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__lt__'] = __lt__
+    del __lt__
+
+    def __ge__(self, other):
+        raise TypeError("unorderable types: %s() >= %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__ge__'] = __ge__
+    del __ge__
+
+    def __gt__(self, other):
+        raise TypeError("unorderable types: %s() > %s()" % (self.__class__.__name__, other.__class__.__name__))
+    temp_enum_dict['__gt__'] = __gt__
+    del __gt__
+    
+
+def __eq__(self, other):
+    if type(other) is self.__class__:
+        return self is other
+    return NotImplemented
+temp_enum_dict['__eq__'] = __eq__
+del __eq__
+
+def __ne__(self, other):
+    if type(other) is self.__class__:
+        return self is not other
+    return NotImplemented
+temp_enum_dict['__ne__'] = __ne__
+del __ne__
+
+def __hash__(self):
+    return hash(self._name_)
+temp_enum_dict['__hash__'] = __hash__
+del __hash__
+
+def __reduce_ex__(self, proto):
+    return self.__class__, (self._value_, )
+temp_enum_dict['__reduce_ex__'] = __reduce_ex__
+del __reduce_ex__
+
+# _RouteClassAttributeToGetattr is used to provide access to the `name`
+# and `value` properties of enum members while keeping some measure of
+# protection from modification, while still allowing for an enumeration
+# to have members named `name` and `value`.  This works because enumeration
+# members are not set directly on the enum class -- __getattr__ is
+# used to look them up.
+
+@_RouteClassAttributeToGetattr
+def name(self):
+    return self._name_
+temp_enum_dict['name'] = name
+del name
+
+@_RouteClassAttributeToGetattr
+def value(self):
+    return self._value_
+temp_enum_dict['value'] = value
+del value
+
+Enum = EnumMeta('Enum', (object, ), temp_enum_dict)
+del temp_enum_dict
+
+# Enum has now been created
+###########################
+
+class IntEnum(int, Enum):
+    """Enum where members are also (and must be) ints"""
+
+
+def unique(enumeration):
+    """Class decorator that ensures only unique members exist in an enumeration."""
+    duplicates = []
+    for name, member in enumeration.__members__.items():
+        if name != member.name:
+            duplicates.append((name, member.name))
+    if duplicates:
+        duplicate_names = ', '.join(
+                ["%s -> %s" % (alias, name) for (alias, name) in duplicates]
+                )
+        raise ValueError('duplicate names found in %r: %s' %
+                (enumeration, duplicate_names)
+                )
+    return enumeration

lib/mako/LICENSE

@@ -0,0 +1,20 @@
+This is the MIT license: http://www.opensource.org/licenses/mit-license.php
+
+Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>.
+Mako is a trademark of Michael Bayer.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy of this
+software and associated documentation files (the "Software"), to deal in the Software
+without restriction, including without limitation the rights to use, copy, modify, merge,
+publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons
+to whom the Software is furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all copies or
+substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
+INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
+PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE
+FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+DEALINGS IN THE SOFTWARE.

lib/mako/README.rst

@@ -0,0 +1,52 @@
+=========================
+Mako Templates for Python
+=========================
+
+Mako is a template library written in Python. It provides a familiar, non-XML 
+syntax which compiles into Python modules for maximum performance. Mako's 
+syntax and API borrows from the best ideas of many others, including Django
+templates, Cheetah, Myghty, and Genshi. Conceptually, Mako is an embedded 
+Python (i.e. Python Server Page) language, which refines the familiar ideas
+of componentized layout and inheritance to produce one of the most 
+straightforward and flexible models available, while also maintaining close 
+ties to Python calling and scoping semantics.
+
+Nutshell
+========
+
+::
+
+    <%inherit file="base.html"/>
+    <%
+        rows = [[v for v in range(0,10)] for row in range(0,10)]
+    %>
+    <table>
+        % for row in rows:
+            ${makerow(row)}
+        % endfor
+    </table>
+
+    <%def name="makerow(row)">
+        <tr>
+        % for name in row:
+            <td>${name}</td>\
+        % endfor
+        </tr>
+    </%def>
+
+Philosophy
+===========
+
+Python is a great scripting language. Don't reinvent the wheel...your templates can handle it !
+
+Documentation
+==============
+
+See documentation for Mako at http://www.makotemplates.org/docs/
+
+License
+========
+
+Mako is licensed under an MIT-style license (see LICENSE).
+Other incorporated projects may be licensed under different licenses.
+All licenses allow for non-commercial and commercial use.

lib/mako/__init__.py

@@ -1,9 +1,8 @@
 # mako/__init__.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
 
 
-__version__ = '0.9.1'
-
+__version__ = '1.0.1'

lib/mako/_ast_util.py

@@ -1,5 +1,5 @@
 # mako/_ast_util.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
@@ -34,42 +34,42 @@ from _ast import *
 from mako.compat import arg_stringname
 
 BOOLOP_SYMBOLS = {
-    And:        'and',
-    Or:         'or'
+    And: 'and',
+    Or: 'or'
 }
 
 BINOP_SYMBOLS = {
-    Add:        '+',
-    Sub:        '-',
-    Mult:       '*',
-    Div:        '/',
-    FloorDiv:   '//',
-    Mod:        '%',
-    LShift:     '<<',
-    RShift:     '>>',
-    BitOr:      '|',
-    BitAnd:     '&',
-    BitXor:     '^'
+    Add: '+',
+    Sub: '-',
+    Mult: '*',
+    Div: '/',
+    FloorDiv: '//',
+    Mod: '%',
+    LShift: '<<',
+    RShift: '>>',
+    BitOr: '|',
+    BitAnd: '&',
+    BitXor: '^'
 }
 
 CMPOP_SYMBOLS = {
-    Eq:         '==',
-    Gt:         '>',
-    GtE:        '>=',
-    In:         'in',
-    Is:         'is',
-    IsNot:      'is not',
-    Lt:         '<',
-    LtE:        '<=',
-    NotEq:      '!=',
-    NotIn:      'not in'
+    Eq: '==',
+    Gt: '>',
+    GtE: '>=',
+    In: 'in',
+    Is: 'is',
+    IsNot: 'is not',
+    Lt: '<',
+    LtE: '<=',
+    NotEq: '!=',
+    NotIn: 'not in'
 }
 
 UNARYOP_SYMBOLS = {
-    Invert:     '~',
-    Not:        'not',
-    UAdd:       '+',
-    USub:       '-'
+    Invert: '~',
+    Not: 'not',
+    UAdd: '+',
+    USub: '-'
 }
 
 ALL_SYMBOLS = {}
@@ -215,8 +215,8 @@ def get_compile_mode(node):
     if not isinstance(node, mod):
         raise TypeError('expected mod node, got %r' % node.__class__.__name__)
     return {
-        Expression:     'eval',
-        Interactive:    'single'
+        Expression: 'eval',
+        Interactive: 'single'
     }.get(node.__class__, 'expr')
 
 

lib/mako/ast.py

@@ -1,5 +1,5 @@
 # mako/ast.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
@@ -8,7 +8,6 @@
 code, as well as generating Python from AST nodes"""
 
 from mako import exceptions, pyparser, compat
-from mako.compat import arg_stringname
 import re
 
 class PythonCode(object):
@@ -107,43 +106,70 @@ class FunctionDecl(object):
         f.visit(expr)
         if not hasattr(self, 'funcname'):
             raise exceptions.CompileException(
-                              "Code '%s' is not a function declaration" % code,
-                              **exception_kwargs)
+                            "Code '%s' is not a function declaration" % code,
+                            **exception_kwargs)
         if not allow_kwargs and self.kwargs:
             raise exceptions.CompileException(
                                 "'**%s' keyword argument not allowed here" %
-                                self.argnames[-1], **exception_kwargs)
+                                self.kwargnames[-1], **exception_kwargs)
 
-    def get_argument_expressions(self, include_defaults=True):
-        """return the argument declarations of this FunctionDecl as a printable
-        list."""
+    def get_argument_expressions(self, as_call=False):
+        """Return the argument declarations of this FunctionDecl as a printable
+        list.
+
+        By default the return value is appropriate for writing in a ``def``;
+        set `as_call` to true to build arguments to be passed to the function
+        instead (assuming locals with the same names as the arguments exist).
+        """
 
         namedecls = []
-        defaults = [d for d in self.defaults]
-        kwargs = self.kwargs
-        varargs = self.varargs
-        argnames = [f for f in self.argnames]
-        argnames.reverse()
-        for arg in argnames:
-            default = None
-            if kwargs:
-                arg = "**" + arg_stringname(arg)
-                kwargs = False
-            elif varargs:
-                arg = "*" + arg_stringname(arg)
-                varargs = False
+
+        # Build in reverse order, since defaults and slurpy args come last
+        argnames = self.argnames[::-1]
+        kwargnames = self.kwargnames[::-1]
+        defaults = self.defaults[::-1]
+        kwdefaults = self.kwdefaults[::-1]
+
+        # Named arguments
+        if self.kwargs:
+            namedecls.append("**" + kwargnames.pop(0))
+
+        for name in kwargnames:
+            # Keyword-only arguments must always be used by name, so even if
+            # this is a call, print out `foo=foo`
+            if as_call:
+                namedecls.append("%s=%s" % (name, name))
+            elif kwdefaults:
+                default = kwdefaults.pop(0)
+                if default is None:
+                    # The AST always gives kwargs a default, since you can do
+                    # `def foo(*, a=1, b, c=3)`
+                    namedecls.append(name)
+                else:
+                    namedecls.append("%s=%s" % (
+                        name, pyparser.ExpressionGenerator(default).value()))
             else:
-                default = len(defaults) and defaults.pop() or None
-            if include_defaults and default:
-                namedecls.insert(0, "%s=%s" %
-                            (arg,
-                            pyparser.ExpressionGenerator(default).value()
-                            )
-                        )
+                namedecls.append(name)
+
+        # Positional arguments
+        if self.varargs:
+            namedecls.append("*" + argnames.pop(0))
+
+        for name in argnames:
+            if as_call or not defaults:
+                namedecls.append(name)
             else:
-                namedecls.insert(0, arg)
+                default = defaults.pop(0)
+                namedecls.append("%s=%s" % (
+                    name, pyparser.ExpressionGenerator(default).value()))
+
+        namedecls.reverse()
         return namedecls
 
+    @property
+    def allargnames(self):
+        return tuple(self.argnames) + tuple(self.kwargnames)
+
 class FunctionArgs(FunctionDecl):
     """the argument portion of a function declaration"""
 

lib/mako/cache.py

@@ -1,5 +1,5 @@
 # mako/cache.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
@@ -88,9 +88,10 @@ class Cache(object):
         if not self.template.cache_enabled:
             return creation_function()
 
-        return self.impl.get_or_create(key,
-                        creation_function,
-                        **self._get_cache_kw(kw, context))
+        return self.impl.get_or_create(
+            key,
+            creation_function,
+            **self._get_cache_kw(kw, context))
 
     def set(self, key, value, **kw):
         """Place a value in the cache.
@@ -178,6 +179,7 @@ class Cache(object):
             tmpl_kw.setdefault('context', context)
         return tmpl_kw
 
+
 class CacheImpl(object):
     """Provide a cache implementation for use by :class:`.Cache`."""
 

lib/mako/cmd.py

@@ -0,0 +1,62 @@
+# mako/cmd.py
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
+#
+# This module is part of Mako and is released under
+# the MIT License: http://www.opensource.org/licenses/mit-license.php
+from argparse import ArgumentParser
+from os.path import isfile, dirname
+import sys
+from mako.template import Template
+from mako.lookup import TemplateLookup
+from mako import exceptions
+
+def varsplit(var):
+    if "=" not in var:
+        return (var, "")
+    return var.split("=", 1)
+
+def _exit():
+    sys.stderr.write(exceptions.text_error_template().render())
+    sys.exit(1)
+
+def cmdline(argv=None):
+
+    parser = ArgumentParser("usage: %prog [FILENAME]")
+    parser.add_argument("--var", default=[], action="append",
+                  help="variable (can be used multiple times, use name=value)")
+    parser.add_argument("--template-dir", default=[], action="append",
+                  help="Directory to use for template lookup (multiple "
+                    "directories may be provided). If not given then if the "
+                    "template is read from stdin, the value defaults to be "
+                    "the current directory, otherwise it defaults to be the "
+                    "parent directory of the file provided.")
+    parser.add_argument('input', nargs='?', default='-')
+
+    options = parser.parse_args(argv)
+    if options.input == '-':
+        lookup_dirs = options.template_dir or ["."]
+        lookup = TemplateLookup(lookup_dirs)
+        try:
+            template = Template(sys.stdin.read(), lookup=lookup)
+        except:
+            _exit()
+    else:
+        filename = options.input
+        if not isfile(filename):
+            raise SystemExit("error: can't find %s" % filename)
+        lookup_dirs = options.template_dir or [dirname(filename)]
+        lookup = TemplateLookup(lookup_dirs)
+        try:
+            template = Template(filename=filename, lookup=lookup)
+        except:
+            _exit()
+
+    kw = dict([varsplit(var) for var in options.var])
+    try:
+        print(template.render(**kw))
+    except:
+        _exit()
+
+
+if __name__ == "__main__":
+    cmdline()

lib/mako/codegen.py

@@ -1,5 +1,5 @@
 # mako/codegen.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
@@ -14,7 +14,7 @@ from mako import util, ast, parsetree, filters, exceptions
 from mako import compat
 
 
-MAGIC_NUMBER = 9
+MAGIC_NUMBER = 10
 
 # names which are hardwired into the
 # template and are not accessed via the
@@ -99,7 +99,6 @@ class _GenerateRenderMethod(object):
     """
     def __init__(self, printer, compiler, node):
         self.printer = printer
-        self.last_source_line = -1
         self.compiler = compiler
         self.node = node
         self.identifier_stack = [None]
@@ -146,6 +145,26 @@ class _GenerateRenderMethod(object):
             for node in defs:
                 _GenerateRenderMethod(printer, compiler, node)
 
+        if not self.in_def:
+            self.write_metadata_struct()
+
+    def write_metadata_struct(self):
+        self.printer.source_map[self.printer.lineno] = \
+                    max(self.printer.source_map)
+        struct = {
+            "filename": self.compiler.filename,
+            "uri": self.compiler.uri,
+            "source_encoding": self.compiler.source_encoding,
+            "line_map": self.printer.source_map,
+        }
+        self.printer.writelines(
+            '"""',
+            '__M_BEGIN_METADATA',
+            compat.json.dumps(struct),
+            '__M_END_METADATA\n'
+            '"""'
+        )
+
     @property
     def identifiers(self):
         return self.identifier_stack[-1]
@@ -187,7 +206,7 @@ class _GenerateRenderMethod(object):
 
         # module-level names, python code
         if self.compiler.generate_magic_comment and \
-            self.compiler.source_encoding:
+                self.compiler.source_encoding:
             self.printer.writeline("# -*- coding:%s -*-" %
                                     self.compiler.source_encoding)
 
@@ -232,7 +251,7 @@ class _GenerateRenderMethod(object):
                             [n.name for n in
                             main_identifiers.topleveldefs.values()]
                         )
-        self.printer.write("\n\n")
+        self.printer.write_blanks(2)
 
         if len(module_code):
             self.write_module_code(module_code)
@@ -255,8 +274,9 @@ class _GenerateRenderMethod(object):
             decorator = node.decorator
             if decorator:
                 self.printer.writeline(
-                                 "@runtime._decorate_toplevel(%s)" % decorator)
+                                "@runtime._decorate_toplevel(%s)" % decorator)
 
+        self.printer.start_source(node.lineno)
         self.printer.writelines(
             "def %s(%s):" % (name, ','.join(args)),
                 # push new frame, assign current frame to __M_caller
@@ -267,7 +287,7 @@ class _GenerateRenderMethod(object):
             self.printer.writeline("context._push_buffer()")
 
         self.identifier_stack.append(
-                                   self.compiler.identifiers.branch(self.node))
+                                self.compiler.identifiers.branch(self.node))
         if (not self.in_def or self.node.is_block) and '**pageargs' in args:
             self.identifier_stack[-1].argument_declared.add('pageargs')
 
@@ -288,7 +308,7 @@ class _GenerateRenderMethod(object):
 
         self.write_def_finish(self.node, buffered, filtered, cached)
         self.printer.writeline(None)
-        self.printer.write("\n\n")
+        self.printer.write_blanks(2)
         if cached:
             self.write_cache_decorator(
                                 node, name,
@@ -299,7 +319,7 @@ class _GenerateRenderMethod(object):
         """write module-level template code, i.e. that which
         is enclosed in <%! %> tags in the template."""
         for n in module_code:
-            self.write_source_comment(n)
+            self.printer.start_source(n.lineno)
             self.printer.write_indented_block(n.text)
 
     def write_inherit(self, node):
@@ -309,9 +329,9 @@ class _GenerateRenderMethod(object):
             "def _mako_inherit(template, context):",
                 "_mako_generate_namespaces(context)",
                 "return runtime._inherit_from(context, %s, _template_uri)" %
-                 (node.parsed_attributes['file']),
+                (node.parsed_attributes['file']),
                 None
-            )
+        )
 
     def write_namespaces(self, namespaces):
         """write the module-level namespace-generating callable."""
@@ -323,14 +343,14 @@ class _GenerateRenderMethod(object):
                     "_mako_generate_namespaces(context)",
                 "return context.namespaces[(__name__, name)]",
             None, None
-            )
+        )
         self.printer.writeline("def _mako_generate_namespaces(context):")
 
 
         for node in namespaces.values():
             if 'import' in node.attributes:
                 self.compiler.has_ns_imports = True
-            self.write_source_comment(node)
+            self.printer.start_source(node.lineno)
             if len(node.nodes):
                 self.printer.writeline("def make_namespace():")
                 export = []
@@ -369,9 +389,9 @@ class _GenerateRenderMethod(object):
                                 " templateuri=%s, callables=%s, "
                                 " calling_uri=_template_uri)" %
                                 (
-                                   node.name,
-                                   node.parsed_attributes.get('file', 'None'),
-                                   callable_name,
+                                    node.name,
+                                    node.parsed_attributes.get('file', 'None'),
+                                    callable_name,
                                 )
                             )
             elif 'module' in node.parsed_attributes:
@@ -381,9 +401,10 @@ class _GenerateRenderMethod(object):
                                 " callables=%s, calling_uri=_template_uri,"
                                 " module=%s)" %
                                 (
-                                   node.name,
-                                   callable_name,
-                                   node.parsed_attributes.get('module', 'None')
+                                    node.name,
+                                    callable_name,
+                                    node.parsed_attributes.get(
+                                                'module', 'None')
                                 )
                             )
             else:
@@ -400,8 +421,8 @@ class _GenerateRenderMethod(object):
                 self.printer.writeline("context['self'].%s = ns" % (node.name))
 
             self.printer.writeline(
-                   "context.namespaces[(__name__, %s)] = ns" % repr(node.name))
-            self.printer.write("\n")
+                "context.namespaces[(__name__, %s)] = ns" % repr(node.name))
+            self.printer.write_blanks(1)
         if not len(namespaces):
             self.printer.writeline("pass")
         self.printer.writeline(None)
@@ -466,7 +487,7 @@ class _GenerateRenderMethod(object):
             for ident, ns in self.compiler.namespaces.items():
                 if 'import' in ns.attributes:
                     self.printer.writeline(
-                            "_mako_get_namespace(context, %r)."\
+                            "_mako_get_namespace(context, %r)."
                                     "_populate(_import_ns, %r)" %
                             (
                                 ident,
@@ -532,18 +553,11 @@ class _GenerateRenderMethod(object):
 
         self.printer.writeline("__M_writer = context.writer()")
 
-    def write_source_comment(self, node):
-        """write a source comment containing the line number of the
-        corresponding template line."""
-        if self.last_source_line != node.lineno:
-            self.printer.writeline("# SOURCE LINE %d" % node.lineno)
-            self.last_source_line = node.lineno
-
     def write_def_decl(self, node, identifiers):
         """write a locally-available callable referencing a top-level def"""
         funcname = node.funcname
         namedecls = node.get_argument_expressions()
-        nameargs = node.get_argument_expressions(include_defaults=False)
+        nameargs = node.get_argument_expressions(as_call=True)
 
         if not self.in_def and (
                                 len(self.identifiers.locally_assigned) > 0 or
@@ -553,7 +567,7 @@ class _GenerateRenderMethod(object):
             nameargs.insert(0, 'context')
         self.printer.writeline("def %s(%s):" % (funcname, ",".join(namedecls)))
         self.printer.writeline(
-                       "return render_%s(%s)" % (funcname, ",".join(nameargs)))
+                    "return render_%s(%s)" % (funcname, ",".join(nameargs)))
         self.printer.writeline(None)
 
     def write_inline_def(self, node, identifiers, nested):
@@ -564,9 +578,9 @@ class _GenerateRenderMethod(object):
         decorator = node.decorator
         if decorator:
             self.printer.writeline(
-                          "@runtime._decorate_inline(context, %s)" % decorator)
+                        "@runtime._decorate_inline(context, %s)" % decorator)
         self.printer.writeline(
-                          "def %s(%s):" % (node.funcname, ",".join(namedecls)))
+                        "def %s(%s):" % (node.funcname, ",".join(namedecls)))
         filtered = len(node.filter_args.args) > 0
         buffered = eval(node.attributes.get('buffered', 'False'))
         cached = eval(node.attributes.get('cached', 'False'))
@@ -574,11 +588,11 @@ class _GenerateRenderMethod(object):
             # push new frame, assign current frame to __M_caller
             "__M_caller = context.caller_stack._push_frame()",
             "try:"
-            )
+        )
         if buffered or filtered or cached:
             self.printer.writelines(
                 "context._push_buffer()",
-                )
+            )
 
         identifiers = identifiers.branch(node, nested=nested)
 
@@ -627,8 +641,8 @@ class _GenerateRenderMethod(object):
                 )
             else:
                 self.printer.writelines(
-                   "finally:",
-                       "__M_buf, __M_writer = context._pop_buffer_and_writer()"
+                    "finally:",
+                    "__M_buf, __M_writer = context._pop_buffer_and_writer()"
                 )
 
             if callstack:
@@ -684,7 +698,7 @@ class _GenerateRenderMethod(object):
 
         # form "arg1, arg2, arg3=arg3, arg4=arg4", etc.
         pass_args = [
-                        '=' in a and "%s=%s" % ((a.split('=')[0],)*2) or a
+                        "%s=%s" % ((a.split('=')[0],) * 2) if '=' in a else a
                         for a in args
                     ]
 
@@ -696,11 +710,11 @@ class _GenerateRenderMethod(object):
         if buffered:
             s = "context.get('local')."\
                 "cache._ctx_get_or_create("\
-                "%s, lambda:__M_%s(%s),  context, %s__M_defname=%r)" % \
-                            (cachekey, name, ','.join(pass_args),
-                            ''.join(["%s=%s, " % (k, v)
-                            for k, v in cache_args.items()]),
-                            name
+                "%s, lambda:__M_%s(%s),  context, %s__M_defname=%r)" % (
+                                cachekey, name, ','.join(pass_args),
+                                ''.join(["%s=%s, " % (k, v)
+                                for k, v in cache_args.items()]),
+                                name
                             )
             # apply buffer_filters
             s = self.create_filter_callable(self.compiler.buffer_filters, s,
@@ -709,12 +723,13 @@ class _GenerateRenderMethod(object):
         else:
             self.printer.writelines(
                     "__M_writer(context.get('local')."
-                    "cache._ctx_get_or_create("\
+                    "cache._ctx_get_or_create("
                     "%s, lambda:__M_%s(%s), context, %s__M_defname=%r))" %
-                    (cachekey, name, ','.join(pass_args),
-                    ''.join(["%s=%s, " % (k, v)
+                    (
+                        cachekey, name, ','.join(pass_args),
+                        ''.join(["%s=%s, " % (k, v)
                         for k, v in cache_args.items()]),
-                    name,
+                        name,
                     ),
                     "return ''",
                 None
@@ -745,18 +760,17 @@ class _GenerateRenderMethod(object):
                 continue
             m = re.match(r'(.+?)(\(.*\))', e)
             if m:
-                (ident, fargs) = m.group(1,2)
+                ident, fargs = m.group(1, 2)
                 f = locate_encode(ident)
                 e = f + fargs
             else:
-                x = e
                 e = locate_encode(e)
                 assert e is not None
             target = "%s(%s)" % (e, target)
         return target
 
     def visitExpression(self, node):
-        self.write_source_comment(node)
+        self.printer.start_source(node.lineno)
         if len(node.escapes) or \
                 (
                     self.compiler.pagetag is not None and
@@ -778,7 +792,7 @@ class _GenerateRenderMethod(object):
                 self.printer.writeline("loop = __M_loop._exit()")
                 self.printer.writeline(None)
         else:
-            self.write_source_comment(node)
+            self.printer.start_source(node.lineno)
             if self.compiler.enable_loop and node.keyword == 'for':
                 text = mangle_mako_loop(node, self.printer)
             else:
@@ -800,7 +814,7 @@ class _GenerateRenderMethod(object):
                 self.printer.writeline("pass")
 
     def visitText(self, node):
-        self.write_source_comment(node)
+        self.printer.start_source(node.lineno)
         self.printer.writeline("__M_writer(%s)" % repr(node.content))
 
     def visitTextTag(self, node):
@@ -822,11 +836,11 @@ class _GenerateRenderMethod(object):
                                 "__M_buf.getvalue()",
                                 False),
                 None
-                )
+            )
 
     def visitCode(self, node):
         if not node.ismodule:
-            self.write_source_comment(node)
+            self.printer.start_source(node.lineno)
             self.printer.write_indented_block(node.text)
 
             if not self.in_def and len(self.identifiers.locally_assigned) > 0:
@@ -835,20 +849,20 @@ class _GenerateRenderMethod(object):
                 # which is used for def calls within the same template,
                 # to simulate "enclosing scope"
                 self.printer.writeline(
-                      '__M_locals_builtin_stored = __M_locals_builtin()')
+                    '__M_locals_builtin_stored = __M_locals_builtin()')
                 self.printer.writeline(
-                      '__M_locals.update(__M_dict_builtin([(__M_key,'
-                      ' __M_locals_builtin_stored[__M_key]) for __M_key in'
-                      ' [%s] if __M_key in __M_locals_builtin_stored]))' %
-                      ','.join([repr(x) for x in node.declared_identifiers()]))
+                    '__M_locals.update(__M_dict_builtin([(__M_key,'
+                    ' __M_locals_builtin_stored[__M_key]) for __M_key in'
+                    ' [%s] if __M_key in __M_locals_builtin_stored]))' %
+                    ','.join([repr(x) for x in node.declared_identifiers()]))
 
     def visitIncludeTag(self, node):
-        self.write_source_comment(node)
+        self.printer.start_source(node.lineno)
         args = node.attributes.get('args')
         if args:
             self.printer.writeline(
-                      "runtime._include_file(context, %s, _template_uri, %s)" %
-                      (node.parsed_attributes['file'], args))
+                    "runtime._include_file(context, %s, _template_uri, %s)" %
+                    (node.parsed_attributes['file'], args))
         else:
             self.printer.writeline(
                         "runtime._include_file(context, %s, _template_uri)" %
@@ -864,7 +878,7 @@ class _GenerateRenderMethod(object):
         if node.is_anonymous:
             self.printer.writeline("%s()" % node.funcname)
         else:
-            nameargs = node.get_argument_expressions(include_defaults=False)
+            nameargs = node.get_argument_expressions(as_call=True)
             nameargs += ['**pageargs']
             self.printer.writeline("if 'parent' not in context._data or "
                                   "not hasattr(context._data['parent'], '%s'):"
@@ -941,9 +955,9 @@ class _GenerateRenderMethod(object):
             # push on caller for nested call
             "context.caller_stack.nextcaller = "
                 "runtime.Namespace('caller', context, "
-                                  "callables=ccall(__M_caller))",
+                                "callables=ccall(__M_caller))",
             "try:")
-        self.write_source_comment(node)
+        self.printer.start_source(node.lineno)
         self.printer.writelines(
                 "__M_writer(%s)" % self.create_filter_callable(
                                                     [], node.expression, True),
@@ -966,9 +980,9 @@ class _Identifiers(object):
                 # things that have already been declared
                 # in an enclosing namespace (i.e. names we can just use)
                 self.declared = set(parent.declared).\
-                         union([c.name for c in parent.closuredefs.values()]).\
-                         union(parent.locally_declared).\
-                         union(parent.argument_declared)
+                        union([c.name for c in parent.closuredefs.values()]).\
+                        union(parent.locally_declared).\
+                        union(parent.argument_declared)
 
                 # if these identifiers correspond to a "nested"
                 # scope, it means whatever the parent identifiers
@@ -1012,7 +1026,7 @@ class _Identifiers(object):
             node.accept_visitor(self)
 
         illegal_names = self.compiler.reserved_names.intersection(
-                                                         self.locally_declared)
+                                                        self.locally_declared)
         if illegal_names:
             raise exceptions.NameConflictError(
                 "Reserved words declared in template: %s" %
@@ -1047,7 +1061,7 @@ class _Identifiers(object):
 
         for ident in node.undeclared_identifiers():
             if ident != 'context' and\
-                       ident not in self.declared.union(self.locally_declared):
+                    ident not in self.declared.union(self.locally_declared):
                 self.undeclared.add(ident)
         for ident in node.declared_identifiers():
             self.locally_declared.add(ident)
@@ -1067,7 +1081,7 @@ class _Identifiers(object):
         if not node.ismodule:
             self.check_declared(node)
             self.locally_assigned = self.locally_assigned.union(
-                                                   node.declared_identifiers())
+                                                node.declared_identifiers())
 
     def visitNamespaceTag(self, node):
         # only traverse into the sub-elements of a
@@ -1095,8 +1109,8 @@ class _Identifiers(object):
             self._check_name_exists(self.closuredefs, node)
 
         for ident in node.undeclared_identifiers():
-            if ident != 'context' and\
-                       ident not in self.declared.union(self.locally_declared):
+            if ident != 'context' and \
+                    ident not in self.declared.union(self.locally_declared):
                 self.undeclared.add(ident)
 
         # visit defs only one level deep
@@ -1108,8 +1122,7 @@ class _Identifiers(object):
                 n.accept_visitor(self)
 
     def visitBlockTag(self, node):
-        if node is not self.node and \
-            not node.is_anonymous:
+        if node is not self.node and not node.is_anonymous:
 
             if isinstance(self.node, parsetree.DefTag):
                 raise exceptions.CompileException(
@@ -1123,7 +1136,7 @@ class _Identifiers(object):
 
         for ident in node.undeclared_identifiers():
             if ident != 'context' and \
-                       ident not in self.declared.union(self.locally_declared):
+                    ident not in self.declared.union(self.locally_declared):
                 self.undeclared.add(ident)
 
         if not node.is_anonymous:
@@ -1139,7 +1152,7 @@ class _Identifiers(object):
     def visitTextTag(self, node):
         for ident in node.undeclared_identifiers():
             if ident != 'context' and \
-                ident not in self.declared.union(self.locally_declared):
+                    ident not in self.declared.union(self.locally_declared):
                 self.undeclared.add(ident)
 
     def visitIncludeTag(self, node):
@@ -1156,8 +1169,9 @@ class _Identifiers(object):
     def visitCallTag(self, node):
         if node is self.node:
             for ident in node.undeclared_identifiers():
-                if ident != 'context' and\
-                       ident not in self.declared.union(self.locally_declared):
+                if ident != 'context' and \
+                        ident not in self.declared.union(
+                                                self.locally_declared):
                     self.undeclared.add(ident)
             for ident in node.declared_identifiers():
                 self.argument_declared.add(ident)
@@ -1165,15 +1179,16 @@ class _Identifiers(object):
                 n.accept_visitor(self)
         else:
             for ident in node.undeclared_identifiers():
-                if ident != 'context' and\
-                       ident not in self.declared.union(self.locally_declared):
+                if ident != 'context' and \
+                        ident not in self.declared.union(
+                                                self.locally_declared):
                     self.undeclared.add(ident)
 
 
 _FOR_LOOP = re.compile(
         r'^for\s+((?:\(?)\s*[A-Za-z_][A-Za-z_0-9]*'
         r'(?:\s*,\s*(?:[A-Za-z_][A-Za-z0-9_]*),??)*\s*(?:\)?))\s+in\s+(.*):'
-    )
+)
 
 def mangle_mako_loop(node, printer):
     """converts a for loop into a context manager wrapped around a for loop
@@ -1189,7 +1204,7 @@ def mangle_mako_loop(node, printer):
                     'loop = __M_loop._enter(%s)' % match.group(2),
                     'try:'
                     #'with __M_loop(%s) as loop:' % match.group(2)
-                )
+            )
             text = 'for %s in loop:' % match.group(1)
         else:
             raise SyntaxError("Couldn't apply loop context: %s" % node.text)

lib/mako/compat.py

@@ -3,8 +3,8 @@ import time
 
 py3k = sys.version_info >= (3, 0)
 py33 = sys.version_info >= (3, 3)
+py2k = sys.version_info < (3,)
 py26 = sys.version_info >= (2, 6)
-py25 = sys.version_info >= (2, 5)
 jython = sys.platform.startswith('java')
 win32 = sys.platform.startswith('win')
 pypy = hasattr(sys, 'pypy_version_info')
@@ -23,6 +23,9 @@ if py3k:
     def u(s):
         return s
 
+    def b(s):
+        return s.encode("latin-1")
+
     def octal(lit):
         return eval("0o" + lit)
 
@@ -44,6 +47,9 @@ else:
     def u(s):
         return unicode(s, "utf-8")
 
+    def b(s):
+        return s
+
     def octal(lit):
         return eval("0" + lit)
 
@@ -62,6 +68,18 @@ else:
             fp.close()
 
 
+if py3k:
+    def reraise(tp, value, tb=None, cause=None):
+        if cause is not None:
+            value.__cause__ = cause
+        if value.__traceback__ is not tb:
+            raise value.with_traceback(tb)
+        raise value
+else:
+    exec("def reraise(tp, value, tb=None, cause=None):\n"
+            "    raise tp, value, tb\n")
+
+
 def exception_as():
     return sys.exc_info()[1]
 
@@ -93,23 +111,12 @@ except:
             return func(*(args + fargs), **newkeywords)
         return newfunc
 
-if not py25:
-    def all(iterable):
-        for i in iterable:
-            if not i:
-                return False
-        return True
 
-    def exception_name(exc):
-        try:
-            return exc.__class__.__name__
-        except AttributeError:
-            return exc.__name__
-else:
-    all = all
+all = all
+import json
 
-    def exception_name(exc):
-        return exc.__class__.__name__
+def exception_name(exc):
+    return exc.__class__.__name__
 
 try:
     from inspect import CO_VARKEYWORDS, CO_VARARGS

lib/mako/exceptions.py

@@ -1,5 +1,5 @@
 # mako/exceptions.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
@@ -8,7 +8,6 @@
 
 import traceback
 import sys
-import re
 from mako import util, compat
 
 class MakoException(Exception):
@@ -28,7 +27,7 @@ class CompileException(MakoException):
     def __init__(self, message, source, lineno, pos, filename):
         MakoException.__init__(self,
                               message + _format_filepos(lineno, pos, filename))
-        self.lineno =lineno
+        self.lineno = lineno
         self.pos = pos
         self.filename = filename
         self.source = source
@@ -37,7 +36,7 @@ class SyntaxException(MakoException):
     def __init__(self, message, source, lineno, pos, filename):
         MakoException.__init__(self,
                               message + _format_filepos(lineno, pos, filename))
-        self.lineno =lineno
+        self.lineno = lineno
         self.pos = pos
         self.filename = filename
         self.source = source
@@ -77,7 +76,6 @@ class RichTraceback(object):
         self.records = self._init(traceback)
 
         if isinstance(self.error, (CompileException, SyntaxException)):
-            import mako.template
             self.source = self.error.source
             self.lineno = self.error.lineno
             self._has_source = True
@@ -167,19 +165,19 @@ class RichTraceback(object):
                                             None, None, None, None))
                     continue
 
-                template_ln = module_ln = 1
-                line_map = {}
-                for line in module_source.split("\n"):
-                    match = re.match(r'\s*# SOURCE LINE (\d+)', line)
-                    if match:
-                        template_ln = int(match.group(1))
-                    module_ln += 1
-                    line_map[module_ln] = template_ln
+                template_ln = 1
+
+                source_map = mako.template.ModuleInfo.\
+                                get_module_source_metadata(
+                                    module_source, full_line_map=True)
+                line_map = source_map['full_line_map']
+
                 template_lines = [line for line in
                                     template_source.split("\n")]
                 mods[filename] = (line_map, template_lines)
 
-            template_ln = line_map[lineno]
+            template_ln = line_map[lineno - 1]
+
             if template_ln <= len(template_lines):
                 template_line = template_lines[template_ln - 1]
             else:
@@ -188,7 +186,7 @@ class RichTraceback(object):
                                 line, template_filename, template_ln,
                                 template_line, template_source))
         if not self.source:
-            for l in range(len(new_trcback)-1, 0, -1):
+            for l in range(len(new_trcback) - 1, 0, -1):
                 if new_trcback[l][5]:
                     self.source = new_trcback[l][7]
                     self.lineno = new_trcback[l][5]
@@ -260,10 +258,11 @@ def html_error_template():
     filenames, line numbers and code for that of the originating source
     template, as applicable.
 
-    The template's default ``encoding_errors`` value is ``'htmlentityreplace'``. The
-    template has two options. With the ``full`` option disabled, only a section of
-    an HTML document is returned. With the ``css`` option disabled, the default
-    stylesheet won't be included.
+    The template's default ``encoding_errors`` value is
+    ``'htmlentityreplace'``. The template has two options. With the
+    ``full`` option disabled, only a section of an HTML document is
+    returned. With the ``css`` option disabled, the default stylesheet
+    won't be included.
 
     """
     import mako.template

lib/mako/ext/autohandler.py

@@ -1,5 +1,5 @@
 # ext/autohandler.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
@@ -48,7 +48,7 @@ def autohandler(template, context, name='autohandler'):
         if len(tokens) == 1:
             break
         tokens[-2:] = [name]
-
+ 
     if not lookup.filesystem_checks:
         return lookup._uri_cache.setdefault(
                             (autohandler, _template_uri, name), None)
@@ -62,4 +62,4 @@ def _file_exists(lookup, path):
             return True
     else:
         return False
-
+ 

lib/mako/ext/babelplugin.py

@@ -1,14 +1,36 @@
 # ext/babelplugin.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
 
 """gettext message extraction via Babel: http://babel.edgewall.org/"""
 from babel.messages.extract import extract_python
-from mako.compat import StringIO
-from mako import compat
-from mako import lexer, parsetree
+from mako.ext.extract import MessageExtractor
+
+
+class BabelMakoExtractor(MessageExtractor):
+    def __init__(self, keywords, comment_tags, options):
+        self.keywords = keywords
+        self.options = options
+        self.config = {
+                'comment-tags': u' '.join(comment_tags),
+                'encoding': options.get('input_encoding',
+                    options.get('encoding', None)),
+            }
+        super(BabelMakoExtractor, self).__init__()
+
+    def __call__(self, fileobj):
+        return self.process_file(fileobj)
+
+    def process_python(self, code, code_lineno, translator_strings):
+        comment_tags = self.config['comment-tags']
+        for lineno, funcname, messages, python_translator_comments \
+                in extract_python(code,
+                        self.keywords, comment_tags, self.options):
+            yield (code_lineno + (lineno - 1), funcname, messages,
+                   translator_strings + python_translator_comments)
+
 
 def extract(fileobj, keywords, comment_tags, options):
     """Extract messages from Mako templates.
@@ -22,105 +44,6 @@ def extract(fileobj, keywords, comment_tags, options):
     :return: an iterator over ``(lineno, funcname, message, comments)`` tuples
     :rtype: ``iterator``
     """
-    encoding = options.get('input_encoding', options.get('encoding', None))
-
-    template_node = lexer.Lexer(fileobj.read(),
-                                input_encoding=encoding).parse()
-    for extracted in extract_nodes(template_node.get_children(),
-                                   keywords, comment_tags, options):
-        yield extracted
-
-def extract_nodes(nodes, keywords, comment_tags, options):
-    """Extract messages from Mako's lexer node objects
-
-    :param nodes: an iterable of Mako parsetree.Node objects to extract from
-    :param keywords: a list of keywords (i.e. function names) that should be
-                     recognized as translation functions
-    :param comment_tags: a list of translator tags to search for and include
-                         in the results
-    :param options: a dictionary of additional options (optional)
-    :return: an iterator over ``(lineno, funcname, message, comments)`` tuples
-    :rtype: ``iterator``
-    """
-    translator_comments = []
-    in_translator_comments = False
-
-    for node in nodes:
-        child_nodes = None
-        if in_translator_comments and isinstance(node, parsetree.Text) and \
-                not node.content.strip():
-            # Ignore whitespace within translator comments
-            continue
-
-        if isinstance(node, parsetree.Comment):
-            value = node.text.strip()
-            if in_translator_comments:
-                translator_comments.extend(_split_comment(node.lineno, value))
-                continue
-            for comment_tag in comment_tags:
-                if value.startswith(comment_tag):
-                    in_translator_comments = True
-                    translator_comments.extend(_split_comment(node.lineno,
-                                                              value))
-            continue
-
-        if isinstance(node, parsetree.DefTag):
-            code = node.function_decl.code
-            child_nodes = node.nodes
-        elif isinstance(node, parsetree.BlockTag):
-            code = node.body_decl.code
-            child_nodes = node.nodes
-        elif isinstance(node, parsetree.CallTag):
-            code = node.code.code
-            child_nodes = node.nodes
-        elif isinstance(node, parsetree.PageTag):
-            code = node.body_decl.code
-        elif isinstance(node, parsetree.CallNamespaceTag):
-            code = node.expression
-            child_nodes = node.nodes
-        elif isinstance(node, parsetree.ControlLine):
-            if node.isend:
-                in_translator_comments = False
-                continue
-            code = node.text
-        elif isinstance(node, parsetree.Code):
-            in_translator_comments = False
-            code = node.code.code
-        elif isinstance(node, parsetree.Expression):
-            code = node.code.code
-        else:
-            continue
-
-        # Comments don't apply unless they immediately preceed the message
-        if translator_comments and \
-                translator_comments[-1][0] < node.lineno - 1:
-            translator_comments = []
-
-        translator_strings = [comment[1] for comment in translator_comments]
-
-        if isinstance(code, compat.text_type):
-            code = code.encode('ascii', 'backslashreplace')
-
-        used_translator_comments = False
-        code = compat.byte_buffer(code)
-        for lineno, funcname, messages, python_translator_comments \
-                in extract_python(code, keywords, comment_tags, options):
-            yield (node.lineno + (lineno - 1), funcname, messages,
-                   translator_strings + python_translator_comments)
-            used_translator_comments = True
-
-        if used_translator_comments:
-            translator_comments = []
-        in_translator_comments = False
-
-        if child_nodes:
-            for extracted in extract_nodes(child_nodes, keywords, comment_tags,
-                                           options):
-                yield extracted
-
-
-def _split_comment(lineno, comment):
-    """Return the multiline comment at lineno split into a list of comment line
-    numbers and the accompanying comment line"""
-    return [(lineno + index, line) for index, line in
-            enumerate(comment.splitlines())]
+    extractor = BabelMakoExtractor(keywords, comment_tags, options)
+    for message in extractor(fileobj):
+        yield message

lib/mako/ext/beaker_cache.py

@@ -4,7 +4,16 @@ from mako import exceptions
 
 from mako.cache import CacheImpl
 
+try:
+    from beaker import cache as beaker_cache
+except:
+    has_beaker = False
+else:
+    has_beaker = True
+
 _beaker_cache = None
+
+
 class BeakerCacheImpl(CacheImpl):
     """A :class:`.CacheImpl` provided for the Beaker caching system.
 
@@ -15,15 +24,11 @@ class BeakerCacheImpl(CacheImpl):
     """
 
     def __init__(self, cache):
+        if not has_beaker:
+            raise exceptions.RuntimeException(
+                "Can't initialize Beaker plugin; Beaker is not installed.")
         global _beaker_cache
         if _beaker_cache is None:
-            try:
-                from beaker import cache as beaker_cache
-            except ImportError:
-                raise exceptions.RuntimeException(
-                            "the Beaker package is required to use cache "
-                            "functionality.")
-
             if 'manager' in cache.template.cache_args:
                 _beaker_cache = cache.template.cache_args['manager']
             else:
@@ -48,7 +53,7 @@ class BeakerCacheImpl(CacheImpl):
             cache = _beaker_cache.get_cache_region(self.cache.id, region, **kw)
         else:
             cache = _beaker_cache.get_cache(self.cache.id, **kw)
-        cache_args = {'starttime':self.cache.starttime}
+        cache_args = {'starttime': self.cache.starttime}
         if expiretime:
             cache_args['expiretime'] = expiretime
         return cache, cache_args

lib/mako/ext/extract.py

@@ -0,0 +1,101 @@
+import re
+from mako import compat
+from mako import lexer
+from mako import parsetree
+
+
+class MessageExtractor(object):
+    def process_file(self, fileobj):
+        template_node = lexer.Lexer(
+            fileobj.read(),
+            input_encoding=self.config['encoding']).parse()
+        for extracted in self.extract_nodes(template_node.get_children()):
+            yield extracted
+
+    def extract_nodes(self, nodes):
+        translator_comments = []
+        in_translator_comments = False
+        comment_tags = list(
+            filter(None, re.split(r'\s+', self.config['comment-tags'])))
+
+        for node in nodes:
+            child_nodes = None
+            if in_translator_comments and \
+                    isinstance(node, parsetree.Text) and \
+                    not node.content.strip():
+                # Ignore whitespace within translator comments
+                continue
+
+            if isinstance(node, parsetree.Comment):
+                value = node.text.strip()
+                if in_translator_comments:
+                    translator_comments.extend(
+                        self._split_comment(node.lineno, value))
+                    continue
+                for comment_tag in comment_tags:
+                    if value.startswith(comment_tag):
+                        in_translator_comments = True
+                        translator_comments.extend(
+                            self._split_comment(node.lineno, value))
+                continue
+
+            if isinstance(node, parsetree.DefTag):
+                code = node.function_decl.code
+                child_nodes = node.nodes
+            elif isinstance(node, parsetree.BlockTag):
+                code = node.body_decl.code
+                child_nodes = node.nodes
+            elif isinstance(node, parsetree.CallTag):
+                code = node.code.code
+                child_nodes = node.nodes
+            elif isinstance(node, parsetree.PageTag):
+                code = node.body_decl.code
+            elif isinstance(node, parsetree.CallNamespaceTag):
+                code = node.expression
+                child_nodes = node.nodes
+            elif isinstance(node, parsetree.ControlLine):
+                if node.isend:
+                    in_translator_comments = False
+                    continue
+                code = node.text
+            elif isinstance(node, parsetree.Code):
+                in_translator_comments = False
+                code = node.code.code
+            elif isinstance(node, parsetree.Expression):
+                code = node.code.code
+            else:
+                continue
+
+            # Comments don't apply unless they immediately preceed the message
+            if translator_comments and \
+                    translator_comments[-1][0] < node.lineno - 1:
+                translator_comments = []
+
+            translator_strings = [
+                comment[1] for comment in translator_comments]
+
+            if isinstance(code, compat.text_type):
+                code = code.encode('ascii', 'backslashreplace')
+
+            used_translator_comments = False
+            code = compat.byte_buffer(code)
+
+            for message in self.process_python(
+                    code, node.lineno, translator_strings):
+                yield message
+                used_translator_comments = True
+
+            if used_translator_comments:
+                translator_comments = []
+            in_translator_comments = False
+
+            if child_nodes:
+                for extracted in self.extract_nodes(child_nodes):
+                    yield extracted
+
+    @staticmethod
+    def _split_comment(lineno, comment):
+        """Return the multiline comment at lineno split into a list of
+        comment line numbers and the accompanying comment line"""
+        return [(lineno + index, line) for index, line in
+                enumerate(comment.splitlines())]

lib/mako/ext/linguaplugin.py

@@ -0,0 +1,38 @@
+import io
+from lingua.extractors import Extractor
+from lingua.extractors import Message
+from lingua.extractors import get_extractor
+from mako.ext.extract import MessageExtractor
+from mako import compat
+
+
+class LinguaMakoExtractor(Extractor, MessageExtractor):
+    '''Mako templates'''
+    extensions = ['.mako']
+    default_config = {
+        'encoding': 'utf-8',
+        'comment-tags': '',
+    }
+
+    def __call__(self, filename, options, fileobj=None):
+        self.options = options
+        self.filename = filename
+        self.python_extractor = get_extractor('x.py')
+        if fileobj is None:
+            fileobj = open(filename, 'rb')
+        return self.process_file(fileobj)
+
+    def process_python(self, code, code_lineno, translator_strings):
+        source = code.getvalue().strip()
+        if source.endswith(compat.b(':')):
+            source += compat.b(' pass')
+            code = io.BytesIO(source)
+        for msg in self.python_extractor(
+                self.filename, self.options, code, code_lineno):
+            if translator_strings:
+                msg = Message(msg.msgctxt, msg.msgid, msg.msgid_plural,
+                              msg.flags,
+                              compat.u(' ').join(
+                                  translator_strings + [msg.comment]),
+                              msg.tcomment, msg.location)
+            yield msg

lib/mako/ext/preprocessors.py

@@ -1,20 +1,20 @@
 # ext/preprocessors.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
 
-"""preprocessing functions, used with the 'preprocessor'
+"""preprocessing functions, used with the 'preprocessor' 
 argument on Template, TemplateLookup"""
 
 import re
 
 def convert_comments(text):
     """preprocess old style comments.
-
+ 
     example:
-
+ 
     from mako.ext.preprocessors import convert_comments
-    t = Template(..., preprocessor=preprocess_comments)"""
+    t = Template(..., preprocessor=convert_comments)"""
     return re.sub(r'(?<=\n)\s*#[^#]', "##", text)
 

lib/mako/ext/pygmentplugin.py

@@ -1,5 +1,5 @@
 # ext/pygmentplugin.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php

lib/mako/ext/turbogears.py

@@ -1,5 +1,5 @@
 # ext/turbogears.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php

lib/mako/filters.py

@@ -1,5 +1,5 @@
 # mako/filters.py
-# Copyright (C) 2006-2013 the Mako authors and contributors <see AUTHORS file>
+# Copyright (C) 2006-2015 the Mako authors and contributors <see AUTHORS file>
 #
 # This module is part of Mako and is released under
 # the MIT License: http://www.opensource.org/licenses/mit-license.php
@@ -48,6 +48,10 @@ def url_escape(string):
     string = string.encode("utf8")
     return quote_plus(string)
 
+def legacy_url_escape(string):
+    # convert into a list of octets
+    return quote_plus(string)
+
 def url_unescape(string):
     text = unquote_plus(string)
     if not is_ascii_str(text):
@@ -64,7 +68,7 @@ class Decode(object):
             if isinstance(x, compat.text_type):
                 return x
             elif not isinstance(x, compat.binary_type):
-                return compat.text_type(str(x), encoding=key)
+                return decode(str(x))
             else:
                 return compat.text_type(x, encoding=key)
         return decode
@@ -193,4 +197,5 @@ if compat.py3k:
 
 NON_UNICODE_ESCAPES = DEFAULT_ESCAPES.copy()
 NON_UNICODE_ESCAPES['h'] = 'filters.legacy_html_escape'
+NON_UNICODE_ESCAPES['u'] = 'filters.legacy_url_escape'