def guess_release_group(string):
group_names = [ r'\.(Xvid)-(?P<releaseGroup>.*?)[ \.]',
r'\.(DivX)-(?P<releaseGroup>.*?)[\. ]',
r'\.(DVDivX)-(?P<releaseGroup>.*?)[\. ]',
]
# first try to see whether we have both a known codec and a known release group
group_names = [ r'\.(?P<videoCodec>' + codec + r')-(?P<releaseGroup>.*?)[ \.]'
for codec in (CODECS + FORMATS) ]
for rexp in group_names:
match = re.search(rexp, string, re.IGNORECASE)
if match:
metadata = match.groupdict()
if canonical_form(metadata['releaseGroup']) in properties['releaseGroup']:
return adjust_metadata(metadata), (match.start(1), match.end(2))
# pick anything as releaseGroup as long as we have a codec in front
# this doesn't include a potential dash ('-') ending the release group
# eg: [...].X264-HiS@SiLUHD-English.[...]
group_names = [ r'\.(?P<videoCodec>' + codec + r')-(?P<releaseGroup>.*?)(-(.*?))?[ \.]'
for codec in (CODECS + FORMATS) ]
for rexp in group_names:
match = re.search(rexp, string, re.IGNORECASE)
if match:
return adjust_metadata(match.groupdict()), (match.start(1), match.end(2))
return None, None
def format_guess(guess):
"""Format all the found values to their natural type.
For instance, a year would be stored as an int value, etc...
Note that this modifies the dictionary given as input.
"""
for prop, value in guess.items():
if prop in ("season", "episodeNumber", "year", "cdNumber", "cdNumberTotal"):
guess[prop] = int(guess[prop])
elif isinstance(value, basestring):
if prop in ("edition",):
value = clean_string(value)
guess[prop] = canonical_form(value)
return guess
def guess_filetype(filename, filetype = 'autodetect'):
other = {}
# look at the extension first
fileext = os.path.splitext(filename)[1][1:].lower()
if fileext in subtitle_exts:
if 'movie' in filetype:
filetype = 'moviesubtitle'
elif 'episode' in filetype:
filetype = 'episodesubtitle'
else:
filetype = 'subtitle'
other = { 'container': fileext }
elif fileext in video_exts:
if filetype == 'autodetect':
filetype = 'video'
other = { 'container': fileext }
else:
if filetype == 'autodetect':
filetype = 'unknown'
other = { 'extension': fileext }
# now look whether there are some specific hints for episode vs movie
if filetype in ('video', 'subtitle'):
for rexp, confidence, span_adjust in episode_rexps:
match = re.search(rexp, filename, re.IGNORECASE)
if match:
if filetype == 'video':
filetype = 'episode'
elif filetype == 'subtitle':
filetype = 'episodesubtitle'
break
for prop, value, start, end in find_properties(filename):
if canonical_form(value) == 'DVB':
if filetype == 'video':
filetype = 'episode'
elif filetype == 'subtitle':
filetype = 'episodesubtitle'
break
# if no episode info found, assume it's a movie
if filetype == 'video':
filetype = 'movie'
elif filetype == 'subtitle':
filetype = 'moviesubtitle'
return filetype, other
def format_guess(guess):
"""Format all the found values to their natural type.
For instance, a year would be stored as an int value, etc...
Note that this modifies the dictionary given as input.
"""
for prop, value in list(guess.items()):
if prop in ('season', 'episodeNumber', 'year', 'cdNumber',
'cdNumberTotal', 'bonusNumber', 'filmNumber'):
guess[prop] = int(guess[prop])
elif isinstance(value, base_text_type):
if prop in ('edition', ):
value = clean_string(value)
guess[prop] = canonical_form(value).replace('\\', '')
return guess
def format_guess(guess):
"""Format all the found values to their natural type.
For instance, a year would be stored as an int value, etc...
Note that this modifies the dictionary given as input.
"""
for prop, value in guess.items():
if prop in ('season', 'episodeNumber', 'year', 'cdNumber',
'cdNumberTotal', 'bonusNumber', 'filmNumber'):
guess[prop] = int(guess[prop])
elif isinstance(value, base_text_type):
if prop in ('edition',):
value = clean_string(value)
guess[prop] = canonical_form(value).replace('\\', '')
return guess
def guess_filetype(mtree, filetype):
# put the filetype inside a dummy container to be able to have the
# following functions work correctly as closures
# this is a workaround for python 2 which doesn't have the
# 'nonlocal' keyword (python 3 does have it)
filetype_container = [filetype]
other = {}
filename = mtree.string
def upgrade_episode():
if filetype_container[0] == 'video':
filetype_container[0] = 'episode'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'episodesubtitle'
def upgrade_movie():
if filetype_container[0] == 'video':
filetype_container[0] = 'movie'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'moviesubtitle'
def upgrade_subtitle():
if 'movie' in filetype_container[0]:
filetype_container[0] = 'moviesubtitle'
elif 'episode' in filetype_container[0]:
filetype_container[0] = 'episodesubtitle'
else:
filetype_container[0] = 'subtitle'
def upgrade(type='unknown'):
if filetype_container[0] == 'autodetect':
filetype_container[0] = type
# look at the extension first
fileext = os.path.splitext(filename)[1][1:].lower()
if fileext in subtitle_exts:
upgrade_subtitle()
other = {'container': fileext}
elif fileext in video_exts:
upgrade(type='video')
other = {'container': fileext}
else:
upgrade(type='unknown')
other = {'extension': fileext}
# check whether we are in a 'Movies', 'Tv Shows', ... folder
folder_rexps = [(r'Movies?', upgrade_movie),
(r'Tv ?Shows?', upgrade_episode),
(r'Series', upgrade_episode)]
for frexp, upgrade_func in folder_rexps:
frexp = re.compile(frexp, re.IGNORECASE)
for pathgroup in mtree.children:
if frexp.match(pathgroup.value):
upgrade_func()
# check for a few specific cases which will unintentionally make the
# following heuristics confused (eg: OSS 117 will look like an episode,
# season 1, epnum 17, when it is in fact a movie)
fname = clean_string(filename).lower()
for m in MOVIES:
if m in fname:
upgrade_movie()
for s in SERIES:
if s in fname:
upgrade_episode()
# now look whether there are some specific hints for episode vs movie
if filetype_container[0] in ('video', 'subtitle'):
# if we have an episode_rexp (eg: s02e13), it is an episode
for rexp, _, _ in episode_rexps:
match = re.search(rexp, filename, re.IGNORECASE)
if match:
upgrade_episode()
break
# if we have a 3-4 digit number that's not a year, maybe an episode
match = re.search(r'[^0-9]([0-9]{3,4})[^0-9]', filename)
if match:
fullnumber = int(match.group()[1:-1])
#season = fullnumber // 100
epnumber = fullnumber % 100
possible = True
# check for validity
if epnumber > 40:
possible = False
if valid_year(fullnumber):
possible = False
if possible:
upgrade_episode()
# if we have certain properties characteristic of episodes, it is an ep
for prop, value, _, _ in find_properties(filename):
log.debug('prop: %s = %s' % (prop, value))
if prop == 'episodeFormat':
upgrade_episode()
break
elif canonical_form(value) == 'DVB':
upgrade_episode()
break
# origin-specific type
if 'tvu.org.ru' in filename:
upgrade_episode()
# if no episode info found, assume it's a movie
upgrade_movie()
filetype = filetype_container[0]
return filetype, other
def adjust_metadata(md):
codec = canonical_form(md['videoCodec'])
if codec in FORMATS:
md['format'] = codec
del md['videoCodec']
return md
def guess_filetype(mtree, filetype):
# put the filetype inside a dummy container to be able to have the
# following functions work correctly as closures
# this is a workaround for python 2 which doesn't have the
# 'nonlocal' keyword (python 3 does have it)
filetype_container = [filetype]
other = {}
filename = mtree.string
def upgrade_episode():
if filetype_container[0] == 'video':
filetype_container[0] = 'episode'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'episodesubtitle'
def upgrade_movie():
if filetype_container[0] == 'video':
filetype_container[0] = 'movie'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'moviesubtitle'
def upgrade_subtitle():
if 'movie' in filetype_container[0]:
filetype_container[0] = 'moviesubtitle'
elif 'episode' in filetype_container[0]:
filetype_container[0] = 'episodesubtitle'
else:
filetype_container[0] = 'subtitle'
def upgrade(type='unknown'):
if filetype_container[0] == 'autodetect':
filetype_container[0] = type
# look at the extension first
fileext = os.path.splitext(filename)[1][1:].lower()
if fileext in subtitle_exts:
upgrade_subtitle()
other = { 'container': fileext }
elif fileext in video_exts:
upgrade(type='video')
other = { 'container': fileext }
else:
upgrade(type='unknown')
other = { 'extension': fileext }
# check whether we are in a 'Movies', 'Tv Shows', ... folder
folder_rexps = [ (r'Movies?', upgrade_movie),
(r'Tv ?Shows?', upgrade_episode),
(r'Series', upgrade_episode)
]
for frexp, upgrade_func in folder_rexps:
frexp = re.compile(frexp, re.IGNORECASE)
for pathgroup in mtree.children:
if frexp.match(pathgroup.value):
upgrade_func()
# check for a few specific cases which will unintentionally make the
# following heuristics confused (eg: OSS 117 will look like an episode,
# season 1, epnum 17, when it is in fact a movie)
fname = clean_string(filename).lower()
for m in MOVIES:
if m in fname:
upgrade_movie()
for s in SERIES:
if s in fname:
upgrade_episode()
# now look whether there are some specific hints for episode vs movie
if filetype_container[0] in ('video', 'subtitle'):
# if we have an episode_rexp (eg: s02e13), it is an episode
for rexp, _, _ in episode_rexps:
match = re.search(rexp, filename, re.IGNORECASE)
if match:
upgrade_episode()
break
# if we have a 3-4 digit number that's not a year, maybe an episode
match = re.search(r'[^0-9]([0-9]{3,4})[^0-9]', filename)
if match:
fullnumber = int(match.group()[1:-1])
#season = fullnumber // 100
epnumber = fullnumber % 100
possible = True
# check for validity
if epnumber > 40:
possible = False
if valid_year(fullnumber):
possible = False
if possible:
upgrade_episode()
# if we have certain properties characteristic of episodes, it is an ep
for prop, value, _, _ in find_properties(filename):
log.debug('prop: %s = %s' % (prop, value))
if prop == 'episodeFormat':
upgrade_episode()
break
elif canonical_form(value) == 'DVB':
upgrade_episode()
break
# origin-specific type
if 'tvu.org.ru' in filename:
upgrade_episode()
# if no episode info found, assume it's a movie
upgrade_movie()
filetype = filetype_container[0]
return filetype, other
def guess_filetype(filename, filetype):
other = {}
# look at the extension first
fileext = os.path.splitext(filename)[1][1:].lower()
if fileext in subtitle_exts:
if 'movie' in filetype:
filetype = 'moviesubtitle'
elif 'episode' in filetype:
filetype = 'episodesubtitle'
else:
filetype = 'subtitle'
other = { 'container': fileext }
elif fileext in video_exts:
if filetype == 'autodetect':
filetype = 'video'
other = { 'container': fileext }
else:
if filetype == 'autodetect':
filetype = 'unknown'
other = { 'extension': fileext }
# put the filetype inside a dummy container to be able to have the
# following functions work correctly as closures
# this is a workaround for python 2 which doesn't have the
# 'nonlocal' keyword (python 3 does have it)
filetype_container = [filetype]
def upgrade_episode():
if filetype_container[0] == 'video':
filetype_container[0] = 'episode'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'episodesubtitle'
def upgrade_movie():
if filetype_container[0] == 'video':
filetype_container[0] = 'movie'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'moviesubtitle'
# now look whether there are some specific hints for episode vs movie
if filetype in ('video', 'subtitle'):
for rexp, _, _ in episode_rexps:
match = re.search(rexp, filename, re.IGNORECASE)
if match:
upgrade_episode()
break
for prop, value, _, _ in find_properties(filename):
log.debug('prop: %s = %s' % (prop, value))
if prop == 'episodeFormat':
upgrade_episode()
break
elif canonical_form(value) == 'DVB':
upgrade_episode()
break
# if no episode info found, assume it's a movie
upgrade_movie()
filetype = filetype_container[0]
return filetype, other
def guess_filetype(filename, filetype):
other = {}
# look at the extension first
fileext = os.path.splitext(filename)[1][1:].lower()
if fileext in subtitle_exts:
if 'movie' in filetype:
filetype = 'moviesubtitle'
elif 'episode' in filetype:
filetype = 'episodesubtitle'
else:
filetype = 'subtitle'
other = {'container': fileext}
elif fileext in video_exts:
if filetype == 'autodetect':
filetype = 'video'
other = {'container': fileext}
else:
if filetype == 'autodetect':
filetype = 'unknown'
other = {'extension': fileext}
# put the filetype inside a dummy container to be able to have the
# following functions work correctly as closures
# this is a workaround for python 2 which doesn't have the
# 'nonlocal' keyword (python 3 does have it)
filetype_container = [filetype]
def upgrade_episode():
if filetype_container[0] == 'video':
filetype_container[0] = 'episode'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'episodesubtitle'
def upgrade_movie():
if filetype_container[0] == 'video':
filetype_container[0] = 'movie'
elif filetype_container[0] == 'subtitle':
filetype_container[0] = 'moviesubtitle'
# now look whether there are some specific hints for episode vs movie
if filetype in ('video', 'subtitle'):
for rexp, _, _ in episode_rexps:
match = re.search(rexp, filename, re.IGNORECASE)
if match:
upgrade_episode()
break
for prop, value, _, _ in find_properties(filename):
log.debug('prop: %s = %s' % (prop, value))
if prop == 'episodeFormat':
upgrade_episode()
break
elif canonical_form(value) == 'DVB':
upgrade_episode()
break
if 'tvu.org.ru' in filename:
upgrade_episode()
# if no episode info found, assume it's a movie
upgrade_movie()
filetype = filetype_container[0]
return filetype, other