From 7a013e818dee608fa0081fbc92be52410128fcd0 Mon Sep 17 00:00:00 2001 From: sbuser Date: Sat, 6 Aug 2011 09:50:59 -0500 Subject: [PATCH] Attempt to weed out substring results. When searching for "Foo" we do not want "Foobar" --- headphones/importer.py | 4 +++- headphones/searcher.py | 28 +++++++++++++++++----------- 2 files changed, 20 insertions(+), 12 deletions(-) diff --git a/headphones/importer.py b/headphones/importer.py index c3c6720d..a64e959e 100644 --- a/headphones/importer.py +++ b/headphones/importer.py @@ -329,4 +329,6 @@ def addReleaseById(rid): #start a search for the album import searcher - searcher.searchNZB(rgid, False) \ No newline at end of file + searcher.searchNZB(rgid, False) + else: + logger.info('Release ' + str(rid) + " already exists in the database!") \ No newline at end of file diff --git a/headphones/searcher.py b/headphones/searcher.py index ee1aa2eb..973c8143 100644 --- a/headphones/searcher.py +++ b/headphones/searcher.py @@ -335,17 +335,10 @@ def searchNZB(albumid=None, new=False): logger.info('No results found from NEWZBIN for %s' % term) #attempt to verify that this isn't a substring result - #when looking for "Avant - Avant" we don't want "Avantasia" + #when looking for "Foo - Foo" we don't want "Foobar" #this should be less of an issue when it isn't a self-titled album so we'll only check vs artist - #for reasons I can't explain this still fails on "Lavanttaler" - maybe you have a better regex? - if resultlist: - for result in resultlist: - if re.search('\w' + re.escape(artistterm), result[0], re.IGNORECASE): - logger.info("Removed from results: " + result[0] + " (substring result).") - resultlist.remove(result) - elif re.search(re.escape(artistterm) + '\w', result[0], re.IGNORECASE): - logger.info("Removed from results: " + result[0] + " (substring result).") - resultlist.remove(result) + if len(resultlist): + resultlist[:] = [result for result in resultlist if verifyresult(result[0], artistterm)] if len(resultlist): @@ -472,4 +465,17 @@ def searchNZB(albumid=None, new=False): break myDB.action('UPDATE albums SET status = "Snatched" WHERE AlbumID=?', [albums[2]]) - myDB.action('INSERT INTO snatched VALUES( ?, ?, ?, ?, DATETIME("NOW", "localtime"), ?, ?)', [albums[2], bestqual[0], bestqual[1], bestqual[2], "Snatched", nzb_folder_name]) \ No newline at end of file + myDB.action('INSERT INTO snatched VALUES( ?, ?, ?, ?, DATETIME("NOW", "localtime"), ?, ?)', [albums[2], bestqual[0], bestqual[1], bestqual[2], "Snatched", nzb_folder_name]) + +def verifyresult(title, term): + if not re.search('^' + re.escape(term), title, re.IGNORECASE): + logger.info("Removed from results: " + title + " (artist not at string start).") + return False + elif re.search(re.escape(term) + '\w', title, re.IGNORECASE | re.UNICODE): + logger.info("Removed from results: " + title + " (post substring result).") + return False + elif re.search('\w' + re.escape(term), title, re.IGNORECASE | re.UNICODE): + logger.info("Removed from results: " + title + " (pre substring result).") + return False + else: + return True