gronod/headphones
1
0
Fork 1
mirror of https://github.com/rembo10/headphones.git synced 2026-03-24 13:49:27 +00:00
Code Issues Packages Projects Releases Wiki Activity

More substring result prevention. This time on the full term to prevent

Browse Source 
snatches on "Vol III" when we want "Vol II."
This commit is contained in:
sbuser
2011-08-10 20:06:09 -05:00
parent 5eac6d466e
commit 74b8a88f29
1 changed files with 23 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

39
headphones/searcher.py
View File

@@ -345,7 +345,7 @@ def searchNZB(albumid=None, new=False):
#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
if len(resultlist):
resultlist[:] = [result for result in resultlist if verifyresult(result[0], artistterm)]
resultlist[:] = [result for result in resultlist if verifyresult(result[0], artistterm, term)]
if len(resultlist):
@@ -433,24 +433,31 @@ def searchNZB(albumid=None, new=False):
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])
def verifyresult(title, term):
def verifyresult(title, artistterm, term):
title = re.sub('[\.\-\/\_]', ' ', title)
if term == 'Various Artists':
return True
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
if artistterm <> 'Various Artists':
if not re.search('^' + re.escape(artistterm), title, re.IGNORECASE):
logger.info("Removed from results: " + title + " (artist not at string start).")
return False
elif re.search(re.escape(artistterm) + '\w', title, re.IGNORECASE | re.UNICODE):
logger.info("Removed from results: " + title + " (post substring result).")
return False
elif re.search('\w' + re.escape(artistterm), title, re.IGNORECASE | re.UNICODE):
logger.info("Removed from results: " + title + " (pre substring result).")
return False
#another attempt to weed out substrings. We don't want "Vol III" when we were looking for "Vol II"
tokens = re.split('\W', term, re.IGNORECASE | re.UNICODE)
for token in tokens:
if token == 'Various' or token == 'Artists' or token == 'VA':
continue
if not re.search('(?:\W|^)+' + token + '(?:\W|$)+', title, re.IGNORECASE | re.UNICODE):
logger.info("Removed from results: " + title + " (missing token: " + token + ")")
return False
return True
def getresultNZB(result):
if result[3] == 'newzbin':