dinislage/CouchPotatoServer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update guessit with unicode fix

Browse Source
This commit is contained in:
Ruud
2013-06-22 00:34:58 +02:00
parent b2d9a7675d
commit bc8d8dcd04
35 changed files with 517 additions and 218 deletions
Download Patch File Download Diff File Expand all files Collapse all files
libs/guessit/patterns.py
+96 -58
View File
@@ -20,9 +20,10 @@
#
from __future__ import unicode_literals
import re
subtitle_exts = [ 'srt', 'idx', 'sub', 'ssa', 'txt' ]
subtitle_exts = [ 'srt', 'idx', 'sub', 'ssa' ]
video_exts = ['3g2', '3gp', '3gp2', 'asf', 'avi', 'divx', 'flv', 'm4v', 'mk2',
'mka', 'mkv', 'mov', 'mp4', 'mp4a', 'mpeg', 'mpg', 'ogg', 'ogm',
@@ -42,13 +43,13 @@ episode_rexps = [ # ... Season 2 ...
(r'saison (?P<season>[0-9]+)', 1.0, (0, 0)),
# ... s02e13 ...
(r'[Ss](?P<season>[0-9]{1,2}).{,3}(?P<episodeNumber>(?:[Ee][0-9]{1,2})+)[^0-9]', 1.0, (0, -1)),
(r'[Ss](?P<season>[0-9]{1,2}).?(?P<episodeNumber>(?:[Ee-][0-9]{1,2})+)[^0-9]', 1.0, (0, -1)),
# ... s03-x02 ...
(r'[Ss](?P<season>[0-9]{1,2}).{,3}(?P<bonusNumber>(?:[Xx][0-9]{1,2})+)[^0-9]', 1.0, (0, -1)),
(r'[Ss](?P<season>[0-9]{1,2}).?(?P<bonusNumber>(?:[Xx][0-9]{1,2})+)[^0-9]', 1.0, (0, -1)),
# ... 2x13 ...
(r'[^0-9](?P<season>[0-9]{1,2})(?P<episodeNumber>(?:[xX][0-9]{1,2})+)[^0-9]', 0.8, (1, -1)),
(r'[^0-9](?P<season>[0-9]{1,2}).?(?P<episodeNumber>(?:[xX][0-9]{1,2})+)[^0-9]', 0.8, (1, -1)),
# ... s02 ...
#(sep + r's(?P<season>[0-9]{1,2})' + sep, 0.6, (1, -1)),
@@ -61,7 +62,7 @@ episode_rexps = [ # ... Season 2 ...
('ep' + sep + r'(?P<episodeNumber>[0-9]{1,2})[^0-9]', 0.7, (0, -1)),
# ... e13 ... for a mini-series without a season number
(r'e(?P<episodeNumber>[0-9]{1,2})[^0-9]', 0.6, (0, -1))
(sep + r'e(?P<episodeNumber>[0-9]{1,2})' + sep, 0.6, (1, -1))
]
@@ -99,92 +100,129 @@ video_rexps = [ # cd number
(r'f(?P<filmNumber>[0-9]{1,2})', 1.0, (0, 0))
]
websites = [ 'tvu.org.ru', 'emule-island.com', 'UsaBit.com', 'www.divx-overnet.com', 'sharethefiles.com' ]
websites = [ 'tvu.org.ru', 'emule-island.com', 'UsaBit.com', 'www.divx-overnet.com',
'sharethefiles.com' ]
unlikely_series = ['series']
unlikely_series = [ 'series' ]
properties = { 'format': [ 'DVDRip', 'HD-DVD', 'HDDVD', 'HDDVDRip', 'BluRay', 'Blu-ray', 'BDRip', 'BRRip',
'HDRip', 'DVD', 'DVDivX', 'HDTV', 'DVB', 'DVBRip', 'PDTV', 'WEBRip',
'DVDSCR', 'Screener', 'VHS', 'VIDEO_TS', 'WEB-DL', 'WEBDL' ],
'screenSize': [ '720p', '720', '1080p', '1080' ],
# prop_multi is a dict of { property_name: { canonical_form: [ pattern ] } }
# pattern is a string considered as a regexp, with the addition that dashes are
# replaced with '([ \.-_])?' which matches more types of separators (or none)
# note: simpler patterns need to be at the end of the list to not shadow more
# complete ones, eg: 'AAC' needs to come after 'He-AAC'
# ie: from most specific to less specific
prop_multi = { 'format': { 'DVD': [ 'DVD', 'DVD-Rip', 'VIDEO-TS', 'DVDivX' ],
'HD-DVD': [ 'HD-(?:DVD)?-Rip', 'HD-DVD' ],
'BluRay': [ 'Blu-ray', 'B[DR]Rip' ],
'HDTV': [ 'HD-TV' ],
'DVB': [ 'DVB-Rip', 'DVB', 'PD-TV' ],
'WEBRip': [ 'WEB-Rip' ],
'Screener': [ 'DVD-SCR', 'Screener' ],
'VHS': [ 'VHS' ],
'WEB-DL': [ 'WEB-DL' ] },
'videoCodec': [ 'XviD', 'DivX', 'x264', 'h264', 'Rv10' ],
'screenSize': { '480p': [ '480p?' ],
'720p': [ '720p?' ],
'1080p': [ '1080p?' ] },
'audioCodec': [ 'AC3', 'DTS', 'He-AAC', 'AAC-He', 'AAC' ],
'videoCodec': { 'XviD': [ 'Xvid' ],
'DivX': [ 'DVDivX', 'DivX' ],
'h264': [ '[hx]-264' ],
'Rv10': [ 'Rv10' ] },
'audioChannels': [ '5.1' ],
'audioCodec': { 'AC3': [ 'AC3' ],
'DTS': [ 'DTS' ],
'AAC': [ 'He-AAC', 'AAC-He', 'AAC' ] },
'releaseGroup': [ 'ESiR', 'WAF', 'SEPTiC', '[XCT]', 'iNT', 'PUKKA',
'CHD', 'ViTE', 'TLF', 'DEiTY', 'FLAiTE',
'MDX', 'GM4F', 'DVL', 'SVD', 'iLUMiNADOS', ' FiNaLe',
'UnSeeN', 'aXXo', 'KLAXXON', 'NoTV', 'ZeaL', 'LOL',
'SiNNERS', 'DiRTY', 'REWARD', 'ECI', 'KiNGS', 'CLUE',
'CtrlHD', 'POD', 'WiKi', 'DIMENSION', 'IMMERSE', 'FQM',
'2HD', 'REPTiLE', 'CTU', 'HALCYON', 'EbP', 'SiTV', 'SAiNTS',
'HDBRiSe', 'AlFleNi-TeaM', 'EVOLVE', '0TV' ],
'audioChannels': { '5.1': [ r'5\.1', 'DD5\.1', '5ch' ] },
'episodeFormat': [ 'Minisode', 'Minisodes' ],
'episodeFormat': { 'Minisode': [ 'Minisodes?' ] }
'other': [ '5ch', 'PROPER', 'REPACK', 'LIMITED', 'DualAudio', 'iNTERNAL', 'Audiofixed', 'R5',
'complete', 'classic', # not so sure about these ones, could appear in a title
'ws', # widescreen
],
}
# prop_single dict of { property_name: [ canonical_form ] }
prop_single = { 'releaseGroup': [ 'ESiR', 'WAF', 'SEPTiC', r'\[XCT\]', 'iNT', 'PUKKA',
'CHD', 'ViTE', 'TLF', 'DEiTY', 'FLAiTE',
'MDX', 'GM4F', 'DVL', 'SVD', 'iLUMiNADOS', 'FiNaLe',
'UnSeeN', 'aXXo', 'KLAXXON', 'NoTV', 'ZeaL', 'LOL',
'SiNNERS', 'DiRTY', 'REWARD', 'ECI', 'KiNGS', 'CLUE',
'CtrlHD', 'POD', 'WiKi', 'DIMENSION', 'IMMERSE', 'FQM',
'2HD', 'REPTiLE', 'CTU', 'HALCYON', 'EbP', 'SiTV',
'SAiNTS', 'HDBRiSe', 'AlFleNi-TeaM', 'EVOLVE', '0TV',
'TLA', 'NTB', 'ASAP', 'MOMENTUM', 'FoV', 'D-Z0N3' ],
def find_properties(filename):
'other': [ 'PROPER', 'REPACK', 'LIMITED', 'DualAudio', 'Audiofixed', 'R5',
'complete', 'classic', # not so sure about these ones, could appear in a title
'ws' ] # widescreen
}
_dash = '-'
_psep = '[-\. _]?'
def _to_rexp(prop):
return re.compile(prop.replace(_dash, _psep), re.IGNORECASE)
# properties_rexps dict of { property_name: { canonical_form: [ rexp ] } }
# containing the rexps compiled from both prop_multi and prop_single
properties_rexps = dict((type, dict((canonical_form,
[ _to_rexp(pattern) for pattern in patterns ])
for canonical_form, patterns in props.items()))
for type, props in prop_multi.items())
properties_rexps.update(dict((type, dict((canonical_form, [ _to_rexp(canonical_form) ])
for canonical_form in props))
for type, props in prop_single.items()))
def find_properties(string):
result = []
clow = filename.lower()
for prop, values in properties.items():
for value in values:
pos = clow.find(value.lower())
if pos != -1:
end = pos + len(value)
# make sure our word is always surrounded by separators
if ((pos > 0 and clow[pos - 1] not in sep) or
(end < len(clow) and clow[end] not in sep)):
for property_name, props in properties_rexps.items():
for canonical_form, rexps in props.items():
for value_rexp in rexps:
match = value_rexp.search(string)
if match:
start, end = match.span()
# make sure our word is always surrounded by separators
# note: sep is a regexp, but in this case using it as
# a sequence achieves the same goal
continue
# a char sequence achieves the same goal
if ((start > 0 and string[start-1] not in sep) or
(end < len(string) and string[end] not in sep)):
continue
result.append((prop, value, pos, end))
result.append((property_name, canonical_form, start, end))
return result
property_synonyms = { 'DVD': [ 'DVDRip', 'VIDEO_TS' ],
'HD-DVD': [ 'HDDVD', 'HDDVDRip' ],
'BluRay': [ 'BDRip', 'BRRip', 'Blu-ray' ],
'WEB-DL': [ 'WEBDL' ],
'DVB': [ 'DVBRip', 'PDTV' ],
'Screener': [ 'DVDSCR' ],
'DivX': [ 'DVDivX' ],
'h264': [ 'x264' ],
'720p': [ '720' ],
'1080p': [ '1080' ],
'AAC': [ 'He-AAC', 'AAC-He' ],
'Special Edition': [ 'Special' ],
property_synonyms = { 'Special Edition': [ 'Special' ],
'Collector Edition': [ 'Collector' ],
'Criterion Edition': [ 'Criterion' ],
'Minisode': [ 'Minisodes' ]
'Criterion Edition': [ 'Criterion' ]
}
def revert_synonyms():
reverse = {}
for _, values in properties.items():
for value in values:
reverse[value.lower()] = value
for canonical, synonyms in property_synonyms.items():
for synonym in synonyms:
reverse[synonym.lower()] = canonical
return reverse
reverse_synonyms = revert_synonyms()
def canonical_form(string):
return reverse_synonyms.get(string.lower(), string)
def compute_canonical_form(property_name, value):
"""Return the canonical form of a property given its type if it is a valid
one, None otherwise."""
for canonical_form, rexps in properties_rexps[property_name].items():
for rexp in rexps:
if rexp.match(value):
return canonical_form
return None