def guess_release_group(string):
# first try to see whether we have both a known codec and a known release group
for rexp in GROUP_NAMES:
match = rexp.search(string)
while match:
metadata = match.groupdict()
# make sure this is an actual release group we caught
release_group = (compute_canonical_form('releaseGroup',
metadata['releaseGroup'])
or compute_canonical_form(
'weakReleaseGroup', metadata['releaseGroup']))
if release_group:
return adjust_metadata(metadata), (match.start(1),
match.end(2))
# we didn't find anything conclusive, keep searching
match = rexp.search(string, match.span()[0] + 1)
# 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.[...]
for rexp in GROUP_NAMES2:
match = rexp.search(string)
if match:
return adjust_metadata(match.groupdict()), (match.start(1),
match.end(2))
return None, None
def guess_release_group(string):
# first try to see whether we have both a known codec and a known release group
for rexp in GROUP_NAMES:
match = rexp.search(string)
while match:
metadata = match.groupdict()
# make sure this is an actual release group we caught
release_group = compute_canonical_form("releaseGroup", metadata["releaseGroup"]) or compute_canonical_form(
"weakReleaseGroup", metadata["releaseGroup"]
)
if release_group:
return adjust_metadata(metadata), (match.start(1), match.end(2))
# we didn't find anything conclusive, keep searching
match = rexp.search(string, match.span()[0] + 1)
# 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.[...]
for rexp in GROUP_NAMES2:
match = rexp.search(string)
if match:
return adjust_metadata(match.groupdict()), (match.start(1), match.end(2))
return None, None
def guess_release_group(string):
# 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]
group_names += [r"(?P<format>" + fmt + r")-?(?P<releaseGroup>.*?)[ \.]" for fmt in FORMATS]
for rexp in group_names:
match = re.search(rexp, string, re.IGNORECASE)
if match:
metadata = match.groupdict()
release_group = compute_canonical_form("releaseGroup", metadata["releaseGroup"])
if release_group:
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]
group_names += [r"\.(?P<format>" + fmt + r")-(?P<releaseGroup>.*?)(-(.*?))?[ \.]" for fmt in 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 guess_release_group(string):
# first try to see whether we have both a known codec and a known release group
for rexp in GROUP_NAMES:
match = rexp.search(string)
if match:
metadata = match.groupdict()
release_group = compute_canonical_form('releaseGroup', metadata['releaseGroup'])
if release_group:
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.[...]
for rexp in GROUP_NAMES2:
match = rexp.search(string)
if match:
return adjust_metadata(match.groupdict()), (match.start(1), match.end(2))
return None, None
def guess_release_group(string):
# 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
]
group_names += [
r'(?P<format>' + fmt + r')-?(?P<releaseGroup>.*?)[ \.]'
for fmt in FORMATS
]
for rexp in group_names:
match = re.search(rexp, string, re.IGNORECASE)
if match:
metadata = match.groupdict()
release_group = compute_canonical_form('releaseGroup',
metadata['releaseGroup'])
if release_group:
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
]
group_names += [
r'\.(?P<format>' + fmt + r')-(?P<releaseGroup>.*?)(-(.*?))?[ \.]'
for fmt in 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 adjust_metadata(md):
return dict(
(property_name, compute_canonical_form(property_name, value) or value) for property_name, value in md.items()
)
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 compute_canonical_form('format', 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):
return dict((property_name, compute_canonical_form(property_name, value) or value)
for property_name, value in md.items())
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 compute_canonical_form('format', 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
