def get_torrent_type():
# Move to completed target dir
if os.path.isdir(g_dl_dir + g_target)
os.chdir(g_dl_dir + g_target)
for root, dirs, files in os.walk(os.getcwd()):
for fn in files:
name, ext = os.path.splitext(fn)
fullpath = os.path.join(root, fn)
if ext in g_video_mimes:
guess = guessit.guess_file_info(fullpath)
vid_type = guess['type']
return vid_type
elif ext in g_music_mimes:
return 'music'
else:
name, ext = os.path.splitext(g_target)
if ext in g_video_mimes:
guess = guessit.guess_file_info(g_dl_dir + g_target)
vid_type = guess['type']
return vid_type
elif ext in g_music_mimes:
return 'music'
return 'unknown'
def parse(self, data, type_=None, name=None, **kwargs):
type_ = self._type_map.get(type_)
guessit_options = self._guessit_options(data, type_, name, **kwargs)
if name and name != data:
if not guessit_options.get('strict_name'):
guessit_options['expected_series'] = [name]
guess_result = None
if kwargs.get('metainfo'):
guess_result = guessit.guess_file_info(data, options=guessit_options, type=None)
else:
guess_result = guessit.guess_file_info(data, options=guessit_options, type=type_)
return self.build_parsed(guess_result, data, type_=type_, name=(name if name != data else None), **kwargs)
def Video(torrent, args):
global settings
res = guess_file_info(torrent.name)
args['chain'].append("Video")
if not ('mimetype' in res.keys() and res['mimetype'].startswith('video')):
files = torrent.files()
for fl in files:
res2 = guess_file_info(files[fl]['name'])
if not ('mimetype' in res2.keys() and res2['mimetype'].startswith('video')):
return
return SeriesFunc(torrent, args) or MovieFunc(torrent, args)
def deep_scan_nfo(filename, ratio=deep_scan_ratio):
if verbose:
print('Deep scanning satellite: %s (ratio=%.2f)' % (filename, ratio))
best_guess = None
best_ratio = 0.00
try:
nfo = open(filename)
# Convert file content into iterable words
for word in ''.join([item for item in nfo.readlines()]).split():
try:
guess = guessit.guess_file_info(word + '.nfo',
info=['filename'])
# Series = TV, Title = Movie
if any(item in guess for item in ('title')):
# Compare word against NZB name
diff = difflib.SequenceMatcher(None, word, nzb_name)
# Evaluate ratio against threshold and previous matches
if verbose:
print('Tested: %s (ratio=%.2f)' % (word, diff.ratio()))
if diff.ratio() >= ratio and diff.ratio() > best_ratio:
if verbose:
print('Possible match found: %s (ratio=%.2f)' %
(word, diff.ratio()))
best_guess = guess
best_ratio = diff.ratio()
except UnicodeDecodeError:
# Ignore non-unicode words (common in nfo "artwork")
pass
nfo.close()
except IOError as e:
print('[ERROR] %s' % str(e))
return best_guess
def list(self, filepath, languages):
if not self.config_dict['cache_dir']:
raise Exception('Cache directory is required for this plugin')
possible_languages = self.possible_languages(languages)
if not possible_languages:
return []
guess = guessit.guess_file_info(filepath, 'autodetect')
if guess['type'] != 'episode':
self.logger.debug(u'Not an episode')
return []
# add multiple things to the release group set
release_group = set()
if 'releaseGroup' in guess:
release_group.add(guess['releaseGroup'].lower())
else:
if 'title' in guess:
release_group.add(guess['title'].lower())
if 'screenSize' in guess:
release_group.add(guess['screenSize'].lower())
if 'series' not in guess or len(release_group) == 0:
self.logger.debug(u'Not enough information to proceed')
return []
self.release_group = release_group # used to sort results
return self.query(guess['series'], guess['season'],
guess['episodeNumber'], release_group, filepath,
possible_languages)
def obtain_guessit_query(movie_filename, language):
guess = guessit.guess_file_info(os.path.basename(movie_filename),
info=['filename'])
def extract_query(guess, parts):
result = ['"%s"' % guess.get(k) for k in parts if guess.get(k)]
return ' '.join(result)
result = {}
if guess.get('type') == 'episode':
result['query'] = extract_query(guess,
['series', 'title', 'releaseGroup'])
if 'season' in guess:
result['season'] = guess['season']
if 'episodeNumber' in guess:
result['episode'] = guess['episodeNumber']
elif guess.get('type') == 'movie':
result['query'] = extract_query(guess, ['title', 'year'])
else: # pragma: no cover
assert False, 'internal error: guessit guess: {0}'.format(guess)
result['sublanguageid'] = language
return result
def guess_file(filename, info='filename', options=None, **kwargs):
options = options or {}
filename = u(filename)
print('For:', filename)
guess = guess_file_info(filename, info, options, **kwargs)
if options.get('yaml'):
try:
import yaml
for k, v in guess.items():
if isinstance(v, list) and len(v) == 1:
guess[k] = v[0]
ystr = yaml.safe_dump({filename: dict(guess)},
default_flow_style=False)
i = 0
for yline in ystr.splitlines():
if i == 0:
print("? " + yline[:-1])
elif i == 1:
print(":" + yline[1:])
else:
print(yline)
i = i + 1
return
except ImportError: # pragma: no cover
print('PyYAML not found. Using default output.')
print('GuessIt found:', guess.nice_string(options.get('advanced')))
def guess_file(filename, info='filename', options=None, **kwargs):
options = options or {}
filename = u(filename)
if not options.get('yaml') and not options.get('show_property'):
print('For:', filename)
guess = guess_file_info(filename, info, options, **kwargs)
if not options.get('unidentified'):
try:
del guess['unidentified']
except KeyError:
pass
if options.get('show_property'):
print(guess.get(options.get('show_property'), ''))
return
if options.get('yaml'):
import yaml
for k, v in guess.items():
if isinstance(v, list) and len(v) == 1:
guess[k] = v[0]
ystr = yaml.safe_dump({filename: dict(guess)}, default_flow_style=False, allow_unicode=True)
i = 0
for yline in ystr.splitlines():
if i == 0:
print("? " + yline[:-1])
elif i == 1:
print(":" + yline[1:])
else:
print(yline)
i += 1
return
print('GuessIt found:', guess.nice_string(options.get('advanced')))
def from_path(cls, path):
"""Create a :class:`Video` subclass guessing all informations from the given path
:param string path: path
:return: video object
:rtype: :class:`Episode` or :class:`Movie` or :class:`UnknownVideo`
"""
guess = guessit.guess_file_info(path, "autodetect")
result = None
if guess["type"] == "episode" and "series" in guess and "season" in guess and "episodeNumber" in guess:
title = None
if "title" in guess:
title = guess["title"]
result = Episode(path, guess["series"], guess["season"], guess["episodeNumber"], title, guess)
if guess["type"] == "movie" and "title" in guess:
year = None
if "year" in guess:
year = guess["year"]
result = Movie(path, guess["title"], year, guess)
if not result:
result = UnknownVideo(path, guess)
if not isinstance(result, cls):
raise ValueError("Video is not of requested type")
return result
def from_path(cls, path):
"""Create a :class:`Video` subclass guessing all informations from the given path
:param string path: path
:return: video object
:rtype: :class:`Episode` or :class:`Movie` or :class:`UnknownVideo`
"""
guess = guessit.guess_file_info(path, 'autodetect')
result = None
if guess['type'] == 'episode' and 'series' in guess and 'season' in guess and 'episodeNumber' in guess:
title = None
if 'title' in guess:
title = guess['title']
result = Episode(path, guess['series'], guess['season'], guess['episodeNumber'], title, guess)
if guess['type'] == 'movie' and 'title' in guess:
year = None
if 'year' in guess:
year = guess['year']
result = Movie(path, guess['title'], year, guess)
if not result:
result = UnknownVideo(path, guess)
if not isinstance(result, cls):
raise ValueError('Video is not of requested type')
return result
def parse_series(self, data, **kwargs):
guessit_options = self._guessit_options(kwargs)
if kwargs.get('name') and not guessit_options.get('strict_name'):
guessit_options['expected_series'] = [kwargs['name']]
# If no series name is provided, we don't tell guessit what kind of match we are looking for
# This prevents guessit from determining that too general of matches
# are series
#parse_type = 'episode' if kwargs.get('name') else None
parse_type = 'episode'
guess_result = guessit.guess_file_info(
data,
options=guessit_options,
type=parse_type
)
#if guess_result.get('type') != 'episode':
## TODO: All other failures return an invalid parser. This is just a
## hack to match. Maybe exception instead?
#class InvalidParser(object):
#valid = False
#return InvalidParser()
parsed = GuessitParsedSerie(
data,
kwargs.pop(
'name',
None),
guess_result,
**kwargs)
# Passed in quality overrides parsed one
if kwargs.get('quality'):
parsed._old_quality = kwargs['quality']
return parsed
def from_path(cls, path):
"""Create a :class:`Video` subclass guessing all informations from the given path
:param string path: path
:return: video object
:rtype: :class:`Episode` or :class:`Movie` or :class:`UnknownVideo`
"""
guess = guessit.guess_file_info(path, 'autodetect')
result = None
if guess[
'type'] == 'episode' and 'series' in guess and 'season' in guess and 'episodeNumber' in guess:
title = None
if 'title' in guess:
title = guess['title']
result = Episode(path, guess['series'], guess['season'],
guess['episodeNumber'], title, guess)
if guess['type'] == 'movie' and 'title' in guess:
year = None
if 'year' in guess:
year = guess['year']
result = Movie(path, guess['title'], year, guess)
if not result:
result = UnknownVideo(path, guess)
if not isinstance(result, cls):
raise ValueError('Video is not of requested type')
return result
def query(self, filepath, languages, title, season=None, episode=None, year=None, keywords=None):
params = {'sXML': 1, 'sK': title, 'sJ': ','.join([str(self.get_code(l)) for l in languages])}
if season is not None:
params['sTS'] = season
if episode is not None:
params['sTE'] = episode
if year is not None:
params['sY'] = year
if keywords is not None:
params['sR'] = keywords
r = self.session.get(self.server_url + '/ppodnapisi/search', params=params)
if r.status_code != 200:
logger.error(u'Request %s returned status code %d' % (r.url, r.status_code))
return []
subtitles = []
soup = BeautifulSoup(r.content, self.required_features)
for sub in soup('subtitle'):
if 'n' in sub.flags:
logger.debug(u'Skipping hearing impaired')
continue
language = l
confidence = float(sub.rating.text) / 5.0
sub_keywords = set()
for release in sub.release.text.split():
sub_keywords |= get_keywords(guessit.guess_file_info(release + '.srt', 'autodetect'))
sub_path = get_subtitle_path(filepath, language, self.config.multi)
subtitle = ResultSubtitle(sub_path, language, self.__class__.__name__.lower(),
sub.url.text, confidence=confidence, keywords=sub_keywords)
subtitles.append(subtitle)
return subtitles
def GetVidFileData(dir, file, configData):
fileDir = dir
fileName = file
# 3rd party function to guess according to the file name if movie or tv show
fileInfo = guess_file_info(fileName)
if None == fileInfo:
return None
# According to the guessed type we create the object
vidType = fileInfo.get('type', '')
vidFileData = None
if 'episode' == vidType:
# Get Episode information
series = fileInfo.get('series', '')
season = fileInfo.get('season', '')
episodeNumber = fileInfo.get('episodeNumber', '')
# Get episode suffix
format = fileInfo.get('format', '')
basename = os.path.splitext(fileName)[0]
suffix = basename[basename.lower().find(format.lower()) - 1:]
# Create Episode Data Object
vidFileData = EpisodeData(configData, series, season, episodeNumber, fileDir, fileName, suffix)
elif 'movie' == vidType:
# Get Movie information
title = fileInfo.get('title', '')
year = fileInfo.get('year', '')
# Creeate Movie Data Object
vidFileData = MovieData(configData, title, year, fileDir, fileName)
return vidFileData
def handle_file(self, parent_folder, item):
item = self.format_file_name(item)
file_details = guessit.guess_file_info(item)
# Force type for strange filenames (Mini-series without seasons, etc.)
if "episode" in str(item).lower():
file_details['type'] = 'episode'
if 'container' in file_details:
if file_details['type'] == 'movie':
details = guessit.guess_movie_info(parent_folder+item)
details.update({
'file_dir': "%s/" % parent_folder,
'file_name': item
})
file_object_storage.append(Movie(**details))
elif file_details['type'] == 'episode':
details = guessit.guess_episode_info(parent_folder+item)
details.update({
'file_dir': "%s/" % parent_folder,
'file_name': item
})
if len(file_location_storage) > 0:
temp_file_location_storage = file_location_storage.copy()
for i in temp_file_location_storage:
if i is details['file_dir']:
file_location_storage[i].append(
details['file_name']
)
else:
file_location_storage[details['file_dir']] = [details['file_name']]
else:
file_location_storage[details['file_dir']] = [details['file_name']]
file_object_storage.append(Episode(**details))
def guess_file(filename, info='filename', options=None, **kwargs):
options = options or {}
filename = u(filename)
guess = guess_file_info(filename, info, options, **kwargs)
if options.get('show_property'):
print (guess[options.get('show_property')])
return
print('For:', filename)
if options.get('yaml'):
try:
import yaml
for k, v in guess.items():
if isinstance(v, list) and len(v) == 1:
guess[k] = v[0]
ystr = yaml.safe_dump({filename: dict(guess)}, default_flow_style=False)
i = 0
for yline in ystr.splitlines():
if i == 0:
print("? " + yline[:-1])
elif i == 1:
print(":" + yline[1:])
else:
print(yline)
i = i + 1
return
except ImportError: # pragma: no cover
print('PyYAML not found. Using default output.')
print('GuessIt found:', guess.nice_string(options.get('advanced')))
def deep_scan_nfo(filename, ratio=deep_scan_ratio):
if verbose:
print('Deep scanning satellite: %s (ratio=%.2f)' % (filename, ratio))
best_guess = None
best_ratio = 0.00
try:
nfo = open(os.path.join(root, filename))
# Convert file content into iterable words
for word in ''.join([item for item in nfo.readlines()]).split():
try:
guess = guessit.guess_file_info(word + '.nfo', info=['filename'])
# Series = TV, Title = Movie
if any(item in guess for item in ('series', 'title')):
# Compare word against NZB name
diff = difflib.SequenceMatcher(None, word, nzb_name)
# Evaluate ratio against threshold and previous matches
if verbose:
print('Tested: %s (ratio=%.2f)' % (word, diff.ratio()))
if diff.ratio() >= ratio and diff.ratio() > best_ratio:
if verbose:
print('Possible match found: %s (ratio=%.2f)' % (word, diff.ratio()))
best_guess = guess
best_ratio = diff.ratio()
except UnicodeDecodeError:
# Ignore non-unicode words (common in nfo "artwork")
pass
nfo.close()
except IOError as e:
print('[ERROR] %s' % str(e))
return best_guess
def sort_file(filepath, move):
filename = os.path.basename(filepath)
guess = guessit.guess_file_info(filename)
if not guess.get("type", "") in ["episode", "episodesubtitle"]: return
if not guess.get("mimetype", "").startswith("video"): return
season = str(guess["season"]).zfill(2)
name = format_show_name(guess["series"]).strip()
dst_path = os.path.join(tvroot, name, "Season " + season)
try:
os.makedirs(dst_path)
except OSError as ex:
if ex.errno == 17: pass
else: raise
if move:
logging.info("mv: {0} -> {1}".format(filepath, dst_path))
shutil.move(filepath, dst_path)
else:
# short-circuit if file is found
if filename in os.listdir(dst_path):
raise ValueError("File exists (not overwriting): " + filename)
logging.info("cp: {0} -> {1}".format(filepath, dst_path))
shutil.copy2(filepath, dst_path)
def recursive(path: Path):
for element in path.iterdir():
if element.is_dir():
recursive(element)
if element.suffix in ALL_TYPES['video']:
metadata = guess_file_info(filename=element.name)
if metadata['type'] == 'movie':
if metadata.get('year'):
operations.append((element, self.destination_folders['movies'] / Path(
"{} - {}{}".format(metadata['title'], metadata.get('year'), element.suffix))))
else:
operations.append((element, self.destination_folders['movies'] / Path(
"{}.{}".format(metadata['title'], element.suffix))))
elif metadata['type'] == 'episode':
if not metadata.get('season'):
# Assume it's an anime
operations.append((element, self.destination_folders['tv'] / Path(metadata['series']) / Path(
"{} - {:02d}{}".format(metadata['series'], metadata['episodeNumber'], element.suffix)
)))
else:
# Assume it's a TV Show:
operations.append((element, self.destination_folders['tv'] / Path(metadata['series']) / Path('Season {:02d}'.format(metadata['season'])) / Path(
"{} S{:02d}E{:02d}{}".format(metadata['series'], metadata['episodeNumber'], metadata['season'], element.suffix)
)))
else:
# TODO: More file cases for moving and renaming
continue
def matching_confidence(video, subtitle):
'''Compute the confidence that the subtitle matches the video.
Returns a float between 0 and 1. 1 being the perfect match.'''
guess = guessit.guess_file_info(subtitle.release, 'autodetect')
video_keywords = utils.get_keywords(video.guess)
subtitle_keywords = utils.get_keywords(guess) | subtitle.keywords
replacement = {'keywords': len(video_keywords & subtitle_keywords)}
if isinstance(video, videos.Episode):
replacement.update({'series': 0, 'season': 0, 'episode': 0})
matching_format = '{series:b}{season:b}{episode:b}{keywords:03b}'
best = matching_format.format(series=1, season=1, episode=1, keywords=len(video_keywords))
if guess['type'] in ['episode', 'episodesubtitle']:
if 'series' in guess and guess['series'].lower() == video.series.lower():
replacement['series'] = 1
if 'season' in guess and guess['season'] == video.season:
replacement['season'] = 1
if 'episodeNumber' in guess and guess['episodeNumber'] == video.episode:
replacement['episode'] = 1
elif isinstance(video, videos.Movie):
replacement.update({'title': 0, 'year': 0})
matching_format = '{title:b}{year:b}{keywords:03b}'
best = matching_format.format(title=1, year=1, keywords=len(video_keywords))
if guess['type'] in ['movie', 'moviesubtitle']:
if 'title' in guess and guess['title'].lower() == video.title.lower():
replacement['title'] = 1
if 'year' in guess and guess['year'] == video.year:
replacement['year'] = 1
else:
return 0
confidence = float(int(matching_format.format(**replacement), 2)) / float(int(best, 2))
return confidence
def query(self, filepath, languages, title, season=None, episode=None, year=None, keywords=None):
params = {'sXML': 1, 'sK': title, 'sJ': ','.join([str(self.get_code(l)) for l in languages])}
if season is not None:
params['sTS'] = season
if episode is not None:
params['sTE'] = episode
if year is not None:
params['sY'] = year
if keywords is not None:
params['sR'] = keywords
r = self.session.get(self.server_url + '/ppodnapisi/search', params=params)
if r.status_code != 200:
logger.error(u'Request %s returned status code %d' % (r.url, r.status_code)) if sys.platform != 'win32' else logger.debug('Log line suppressed on windows')
return []
subtitles = []
soup = BeautifulSoup(r.content, self.required_features)
for sub in soup('subtitle'):
if 'n' in sub.flags:
logger.debug(u'Skipping hearing impaired') if sys.platform != 'win32' else logger.debug('Log line suppressed on windows')
continue
language = self.get_language(sub.languageId.text)
confidence = float(sub.rating.text) / 5.0
sub_keywords = set()
for release in sub.release.text.split():
sub_keywords |= get_keywords(guessit.guess_file_info(release + '.srt', 'autodetect'))
sub_path = get_subtitle_path(filepath, language, self.config.multi)
subtitle = ResultSubtitle(sub_path, language, self.__class__.__name__.lower(),
sub.url.text, confidence=confidence, keywords=sub_keywords)
subtitles.append(subtitle)
return subtitles
def fromname(cls, name):
"""Shortcut for :meth:`fromguess` with a `guess` guessed from the `name`.
:param str name: name of the video.
"""
return cls.fromguess(name, guess_file_info(name))
def guess_file(filename, info='filename', options=None, **kwargs):
options = options or {}
filename = u(filename)
if not options.get('yaml') and not options.get('show_property'):
print('For:', filename)
guess = guess_file_info(filename, info, options, **kwargs)
if not options.get('unidentified'):
try:
del guess['unidentified']
except KeyError:
pass
if options.get('show_property'):
print(guess.get(options.get('show_property'), ''))
return
if options.get('yaml'):
import yaml
for k, v in guess.items():
if isinstance(v, list) and len(v) == 1:
guess[k] = v[0]
ystr = yaml.safe_dump({filename: dict(guess)}, default_flow_style=False, allow_unicode=True)
i = 0
for yline in ystr.splitlines():
if i == 0:
print("? " + yline[:-1])
elif i == 1:
print(":" + yline[1:])
else:
print(yline)
i += 1
return
print('GuessIt found:', guess.nice_string(options.get('advanced')))
def sort_file(filepath, move):
filename = os.path.basename(filepath)
guess = guessit.guess_file_info(filename)
if not guess.get("type", "") in ["episode", "episodesubtitle"]: return
if not guess.get("mimetype", "").startswith("video"): return
season = str(guess["season"]).zfill(2)
name = format_show_name(guess["series"]).strip()
dst_path = os.path.join(tvroot, name, "Season " + season)
try:
os.makedirs(dst_path)
except OSError as ex:
if ex.errno == 17: pass
else: raise
if move:
logging.info("mv: {0} -> {1}".format(filepath, dst_path))
shutil.move(filepath, dst_path)
else:
# short-circuit if file is found
if filename in os.listdir(dst_path):
raise ValueError("File exists (not overwriting): " + filename)
logging.info("cp: {0} -> {1}".format(filepath, dst_path))
shutil.copy2(filepath, dst_path)
def matching_confidence(video, subtitle):
"""Compute the confidence that the subtitle matches the video.
Returns a float between 0 and 1. 1 being the perfect match."""
guess = guessit.guess_file_info(subtitle.release, "autodetect")
video_keywords = utils.get_keywords(video.guess)
subtitle_keywords = utils.get_keywords(guess) | subtitle.keywords
replacement = {"keywords": len(video_keywords & subtitle_keywords)}
if isinstance(video, videos.Episode):
replacement.update({"series": 0, "season": 0, "episode": 0})
matching_format = "{series:b}{season:b}{episode:b}{keywords:03b}"
best = matching_format.format(series=1, season=1, episode=1, keywords=len(video_keywords))
if guess["type"] in ["episode", "episodesubtitle"]:
if "series" in guess and guess["series"].lower() == video.series.lower():
replacement["series"] = 1
if "season" in guess and guess["season"] == video.season:
replacement["season"] = 1
if "episodeNumber" in guess and guess["episodeNumber"] == video.episode:
replacement["episode"] = 1
elif isinstance(video, videos.Movie):
replacement.update({"title": 0, "year": 0})
matching_format = "{title:b}{year:b}{keywords:03b}"
best = matching_format.format(title=1, year=1, keywords=len(video_keywords))
if guess["type"] in ["movie", "moviesubtitle"]:
if "title" in guess and guess["title"].lower() == video.title.lower():
replacement["title"] = 1
if "year" in guess and guess["year"] == video.year:
replacement["year"] = 1
else:
return 0
confidence = float(int(matching_format.format(**replacement), 2)) / float(int(best, 2))
return confidence
def get_metadata(self, __episodes):
for key in __episodes:
guessit_result = guessit.guess_file_info(key)
__episodes[key].name = guessit_result.get('series', '')
__episodes[key].season_nr = guessit_result.get('season', '')
__episodes[key].episode_nr = guessit_result.get('episodeNumber', '')
__episodes[key].episode_title = guessit_result.get('title', '')
return __episodes
def guess_filename(filename):
try:
import guessit
return guessit.guess_file_info(filename)
except Exception as e:
name, extension = os.path.splitext(filename)
log_error(e)
return {'title' : name}
def guess_filename(filename):
try:
import guessit
return guessit.guess_file_info(filename)
except Exception as e:
name, extension = os.path.splitext(filename)
log_error(e)
return {'title': name}
def parse(bot, trigger):
release_name = trigger.group(2).strip()
attrs = guess_file_info(release_name.lower(), options={'name_only': True})
items = [EntityGroup([Entity("RlsParse")])]
for k, v in attrs.items():
items.append(EntityGroup([Entity(k, v)]))
bot.say("{}".format(render(items=items)))
def query(self, filepath, languages, title, season=None, episode=None, year=None, keywords=None):
myDB = db.DBConnection()
myDBcache = db.DBConnection("cache.db")
sql_show_id = myDB.select("SELECT tvdb_id, show_name FROM tv_shows WHERE show_name LIKE ?", ["%" + title + "%"])
if sql_show_id[0][0]:
sql_scene = myDB.select(
"SELECT scene_season, scene_episode FROM tv_episodes WHERE showid = ? and season = ? and episode = ?",
[sql_show_id[0][0], season, episode],
)
real_name = sql_show_id[0][1]
if sql_scene[0][0]:
season = sql_scene[0][0]
episode = sql_scene[0][1]
sql_custom_names = myDBcache.select(
"SELECT show_name FROM scene_exceptions WHERE tvdb_id = ? and show_name<> ? ORDER BY exception_id asc",
[sql_show_id[0][0], real_name],
)
if sql_custom_names:
title = sql_custom_names[0][0]
glog.log(
u"Searching Subtitles on Podnapisiweb with title : %s season : %s episode : %s" % (title, season, episode)
)
params = {"sXML": 1, "sK": title, "sJ": ",".join([str(self.get_code(l)) for l in languages])}
if season is not None:
params["sTS"] = season
if episode is not None:
params["sTE"] = episode
if year is not None:
params["sY"] = year
if keywords is not None:
params["sR"] = keywords
r = self.session.get(self.server_url + "/ppodnapisi/search", params=params)
if r.status_code != 200:
logger.error(u"Request %s returned status code %d" % (r.url, r.status_code))
return []
subtitles = []
soup = BeautifulSoup(r.content, self.required_features)
for sub in soup("subtitle"):
if "n" in sub.flags:
logger.debug(u"Skipping hearing impaired")
continue
language = l
confidence = float(sub.rating.text) / 5.0
sub_keywords = set()
for release in sub.release.text.split():
sub_keywords |= get_keywords(guessit.guess_file_info(release + ".srt", "autodetect"))
sub_path = get_subtitle_path(filepath, language, self.config.multi)
subtitle = ResultSubtitle(
sub_path,
language,
self.__class__.__name__.lower(),
sub.url.text,
confidence=confidence,
keywords=sub_keywords,
)
subtitles.append(subtitle)
return subtitles
def guess(filename):
guess = guess_file_info(filename)
# Country is not unicode, it makes the jsonify crash...
# I don't need it anyway so let's remove it
if 'country' in guess:
del guess['country']
return jsonify(guess)
def hash(self):
"""
Compute the MPC hash of a movie.
"""
if self._hash is None:
self._hash = guessit.guess_file_info(
self.path,
'hash_mpc')['hash_mpc']
return self._hash
def factory(cls, release):
#TODO: Work with lowercase
"""Create a Subtitle object guessing all informations from the given subtitle release filename"""
guess = guessit.guess_file_info(release, 'autodetect')
keywords = set()
for k in ['releaseGroup', 'screenSize', 'videoCodec', 'format', 'container']:
if k in guess:
keywords = keywords | splitKeyword(guess[k])
return Subtitle(release=release, keywords=keywords)
def fromFileName(cls, fileName_, tvdb_ = None) :
guess = guessit.guess_file_info(fileName_)
try:
if guess['type'] == 'episode' :
return EpisodeInfo.fromFileName(guess, fileName_, tvdb_)
elif guess['type'] != 'movie' :
raise ValueError('The file name must refer to an episode or a movie')
except:
raise
def parse_movie(self, data, **kwargs):
log.debug('Parsing movie: `%s` [options: %s]', data, kwargs)
start = time.clock()
guessit_options = self._guessit_options(kwargs)
guess_result = guessit.guess_file_info(data, options=guessit_options, type='movie')
parsed = GuessitParsedMovie(data, kwargs.pop('name', None), guess_result, **kwargs)
end = time.clock()
log.debug('Parsing result: %s (in %s ms)', parsed, (end - start) * 1000)
return parsed
def get_metadata(self, __episodes):
for key in __episodes:
guessit_result = guessit.guess_file_info(key)
__episodes[key].name = guessit_result.get('series', '')
__episodes[key].season_nr = guessit_result.get('season', '')
__episodes[key].episode_nr = guessit_result.get(
'episodeNumber', '')
__episodes[key].episode_title = guessit_result.get('title', '')
return __episodes
def processFile(path):
_, ext = os.path.splitext(path)
if ext[1:] not in settings['types']['video']['extensions']:
return
end = path[len(root_folder):]
info = guessit.guess_file_info(end)
if info['type'] == 'movie':
processMovie(path, info)
elif info['type'] == 'episode':
processSeries(path, info)
def get_movie_info(name):
"""Find movie information"""
movie_info = guess_file_info(name)
if movie_info['type'] == "movie":
if 'year' in movie_info:
return omdb(movie_info['title'], movie_info['year'])
else:
return omdb(movie_info['title'], None)
else:
not_a_movie.append(name)
def get_movie_info(name):
"""Find movie information"""
movie_info = guess_file_info(name)
if movie_info['type'] == "movie":
if 'year' in movie_info:
return omdb(movie_info['title'], movie_info['year'])
else:
return omdb(movie_info['title'], None)
else:
not_a_movie.append(name)
def guess_details(self, path):
show_mapper = {"Scandal (US)":"Scandal (2012)"}
result = {}
guess = guess_file_info(path, info='filename')
if guess['type'] == "episode":
try:
show = guess["series"]
if show in show_mapper:
show = show_mapper[show]
result["type"] = "tv"
result["show"] = show
result["season"] = guess["season"]
if guess.has_key("year"):
result["year"] = guess["year"]
if guess.has_key("episodeList"):
result["double_episode"] = 1
result["episode"] = guess["episodeList"][0]
else:
result["double_episode"] = 0
result["episode"] = guess["episodeNumber"]
print result
t = tvdb_api.Tvdb()
if t[result["show"]][result["season"]][result["episode"]]["episodename"] is None:
return []
else:
return result
except:
return []
elif guess['type'] == "movie":
try:
result["type"] = "movie"
result["name"] = guess["title"]
result["year"] = guess["year"]
search = tmdb.Search()
search.movie(query=result["name"])
for s in search.results:
if 'release_date' in s:
if int(s['release_date'][0:4]) == int(result["year"]):
return result
return []
except:
print traceback.format_exc()
return [traceback.format_exc()]
return []
def subliminal():
parser = subliminal_parser()
args = parser.parse_args()
# parse paths
try:
args.paths = [p.decode('utf-8') for p in args.paths]
except UnicodeDecodeError:
parser.error('argument paths: encodings is not utf-8: %r' % args.paths)
# parse languages
try:
args.languages = {babelfish.Language.fromalpha2(l) for l in args.languages}
except babelfish.Error:
parser.error('argument -l/--languages: codes are not ISO-639-1: %r' % args.languages)
# parse age
if args.age is not None:
match = re.match(r'^(?:(?P<weeks>\d+?)w)?(?:(?P<days>\d+?)d)?(?:(?P<hours>\d+?)h)?$', args.age)
if not match:
parser.error('argument -a/--age: invalid age: %r' % args.age)
args.age = datetime.timedelta(**match.groupdict())
# setup verbosity
if args.verbose:
logging.basicConfig(level=logging.DEBUG)
elif not args.quiet:
logging.basicConfig(level=logging.WARN)
# configure cache
cache_region.configure('dogpile.cache.dbm', arguments={'filename': os.path.expanduser(args.cache_file)})
# scan videos
videos = scan_videos([p for p in args.paths if os.path.exists(p)], subtitles=not args.force, age=args.age)
# guess videos
videos.extend([Video.fromguess(os.path.split(p)[1], guessit.guess_file_info(p, 'autodetect')) for p in args.paths
if not os.path.exists(p)])
# download best subtitles
subtitles = download_best_subtitles(videos, args.languages, providers=args.providers, provider_configs=None,
single=args.single, min_score=args.min_score,
hearing_impaired=args.hearing_impaired)
# output result
if not subtitles:
if not args.quiet:
sys.stderr.write('No subtitles downloaded\n')
exit(1)
if not args.quiet:
subtitles_count = sum([len(s) for s in subtitles.values()])
if subtitles_count == 1:
print('%d subtitle downloaded' % subtitles_count)
else:
print('%d subtitles downloaded' % subtitles_count)
def query(self, name, season, episode, release_group, filepath, languages):
sublinks = []
# get the show id
show_name = name.lower()
if show_name in self.exceptions: # get it from exceptions
show_id = self.exceptions[show_name]
elif show_name in self.showids: # get it from cache
show_id = self.showids[show_name]
else: # retrieve it
show_name_encoded = show_name
if isinstance(show_name_encoded, unicode):
show_name_encoded = show_name_encoded.encode('utf-8')
show_id_url = '%sGetShowByName/%s' % (self.server_url, urllib2.quote(show_name_encoded))
self.logger.debug(u'Retrieving show id from web at %s' % show_id_url)
page = urllib2.urlopen(show_id_url)
dom = minidom.parse(page)
if not dom or len(dom.getElementsByTagName('showid')) == 0: # no proper result
page.close()
return []
show_id = dom.getElementsByTagName('showid')[0].firstChild.data
self.showids[show_name] = show_id
with self.lock:
f = open(self.showid_cache, 'w')
self.logger.debug(u'Writing showid %s to cache file' % show_id)
pickle.dump(self.showids, f)
f.close()
page.close()
# get the subs for the show id we have
for language in languages:
subs_url = '%sGetAllSubsFor/%s/%s/%s/%s' % (self.server_url, show_id, season, episode, language)
self.logger.debug(u'Getting subtitles at %s' % subs_url)
page = urllib2.urlopen(subs_url)
dom = minidom.parse(page)
page.close()
for sub in dom.getElementsByTagName('result'):
sub_release = sub.getElementsByTagName('filename')[0].firstChild.data
if sub_release.endswith('.srt'):
sub_release = sub_release[:-4]
sub_release = sub_release + '.avi' # put a random extension for guessit not to fail guessing that file
# guess information from subtitle
sub_guess = guessit.guess_file_info(sub_release, 'episode')
sub_release_group = set()
if 'releaseGroup' in sub_guess:
sub_release_group.add(sub_guess['releaseGroup'].lower())
else:
if 'title' in sub_guess:
sub_release_group.add(sub_guess['title'].lower())
if 'screenSize' in sub_guess:
sub_release_group.add(sub_guess['screenSize'].lower())
sub_link = sub.getElementsByTagName('downloadlink')[0].firstChild.data
result = Subtitle(filepath, self.getSubtitlePath(filepath, language), self.__class__.__name__, language, sub_link, sub_release, sub_release_group)
sublinks.append(result)
sublinks.sort(self._cmpReleaseGroup)
return sublinks
def matching_confidence(video, subtitle):
"""Compute the probability (confidence) that the subtitle matches the video
:param video: video to match
:type video: :class:`~subliminal.videos.Video`
:param subtitle: subtitle to match
:type subtitle: :class:`~subliminal.subtitles.Subtitle`
:return: the matching probability
:rtype: float
"""
guess = guessit.guess_file_info(subtitle.release, 'autodetect')
video_keywords = get_keywords(video.guess)
subtitle_keywords = get_keywords(guess) | subtitle.keywords
logger.debug(u'Video keywords %r - Subtitle keywords %r' %
(video_keywords, subtitle_keywords))
replacement = {'keywords': len(video_keywords & subtitle_keywords)}
if isinstance(video, Episode):
replacement.update({'series': 0, 'season': 0, 'episode': 0})
matching_format = '{series:b}{season:b}{episode:b}{keywords:03b}'
best = matching_format.format(series=1,
season=1,
episode=1,
keywords=len(video_keywords))
if guess['type'] in ['episode', 'episodesubtitle']:
if 'series' in guess and guess['series'].lower(
) == video.series.lower():
replacement['series'] = 1
if 'season' in guess and guess['season'] == video.season:
replacement['season'] = 1
if 'episodeNumber' in guess and guess[
'episodeNumber'] == video.episode:
replacement['episode'] = 1
elif isinstance(video, Movie):
replacement.update({'title': 0, 'year': 0})
matching_format = '{title:b}{year:b}{keywords:03b}'
best = matching_format.format(title=1,
year=1,
keywords=len(video_keywords))
if guess['type'] in ['movie', 'moviesubtitle']:
if 'title' in guess and guess['title'].lower(
) == video.title.lower():
replacement['title'] = 1
if 'year' in guess and guess['year'] == video.year:
replacement['year'] = 1
else:
logger.debug(u'Not able to compute confidence for %r' % video)
return 0.0
logger.debug(u'Found %r' % replacement)
confidence = float(int(matching_format.format(**replacement), 2)) / float(
int(best, 2))
logger.info(u'Computed confidence %.4f for %r and %r' %
(confidence, video, subtitle))
return confidence
def get_matches(self, video, hearing_impaired=False):
matches = super(LegendasTvSubtitle, self).get_matches(video, hearing_impaired=hearing_impaired)
# The best available information about a subtitle is its name. Using guessit to parse it.
guess = self.guess if self.guess else guess_file_info(self.name + '.mkv', type=self.type)
matches |= guess_matches(video, guess)
# imdb_id match used only for movies
if self.type == 'movie' and video.imdb_id and self.imdb_id == video.imdb_id:
matches.add('imdb_id')
return matches
def factory(cls, release):
#TODO: Work with lowercase
"""Create a Subtitle object guessing all informations from the given subtitle release filename"""
guess = guessit.guess_file_info(release, 'autodetect')
keywords = set()
for k in [
'releaseGroup', 'screenSize', 'videoCodec', 'format',
'container'
]:
if k in guess:
keywords = keywords | splitKeyword(guess[k])
return Subtitle(release=release, keywords=keywords)
def lights_camera_action():
# assumes there is at most one directory
target_full_path = g_dl_dir + "/" + g_target
guess = []
# is a directory
if os.path.isdir(target_full_path):
for fn in os.listdir('.'):
name, ext = os.path.splitext(fn)
if ext in g_video_mimes:
guess = guessit.guess_file_info(g_dl_dir + "/" + fn)
break
mv_title = guess['title']
mv_year = guess['year']
if mv_year:
mv_link_name = g_movie_dir + mv_title + " (" + mv_year + ")"
else:
mv_link_name = g_movie_dir + mv_title
os.symlink(target_full_path, mv_link_name)
# was originally just a single video file
else:
guess = guessit.guess_file_info(target_full_path)
mv_title = guess['title']
mv_year = guess['year']
if mv_year:
mv_link_name = g_movie_dir + mv_title + " (" + mv_year + ")"
else:
mv_link_name = g_movie_dir + mv_title
os.mkdir(mv_link_name)
shutil.copy(target_full_path, mv_link_name)
def parse(self, data, type_=None, name=None, **kwargs):
type_ = self._type_map.get(type_)
guessit_options = self._guessit_options(data, type_, name, **kwargs)
if name and name != data:
if not guessit_options.get('strict_name'):
guessit_options['expected_series'] = [name]
guess_result = None
if kwargs.get('metainfo'):
guess_result = guessit.guess_file_info(data,
options=guessit_options,
type=None)
else:
guess_result = guessit.guess_file_info(data,
options=guessit_options,
type=type_)
return self.build_parsed(guess_result,
data,
type_=type_,
name=(name if name != data else None),
**kwargs)
def guessInfo(fileName, tvdbid=None):
if not settings.fullpathguess:
fileName = os.path.basename(fileName)
guess = guessit.guess_file_info(fileName)
try:
if guess['type'] == 'movie':
return tmdbInfo(guess)
elif guess['type'] == 'episode':
return tvdbInfo(guess, tvdbid)
else:
return None
except Exception as e:
print(e)
return None
def get_path_video(filename):
guess = guessit.guess_file_info(filename)
if guess[u'type'] == u'episode':
series = guess.get(u'series', u'').title()
season = guess.get(u'season', u'')
return config.TV_PATH.format(series=series, season=season)
elif guess[u'type'] == u'movie':
title = guess.get(u'title', u'').title()
year = guess.get(u'year', u'')
return config.MOVIE_PATH.format(title=title, year=year)
else:
return None
def guess_file_info_get():
"""
@api {get} /guess Detect properties for a given filename
@apiName GuessFileInfoGet
@apiGroup Guess
@apiParam {String} filename Filename out of which to guess information.
@apiParam {String} * Other fields you pass will be forwarded as options to the guesser.
@apiSuccess {Object}   Object containing all detected fields.
For a list of detected properties see <a href="https://guessit.readthedocs.org/en/latest/#features">here</a>
@apiExample Example usage:
curl "http://guessit.io/guess?filename=House.of.Cards.2013.S02E03.1080p.NF.WEBRip.DD5.1.x264-NTb.mkv"
@apiSuccessExample Success-Response:
HTTP/1.1 200 OK
{
"audioChannels": "5.1",
"audioCodec": "DolbyDigital",
"container": "mkv",
"episodeNumber": 3,
"format": "WEBRip",
"mimetype": "video/x-matroska",
"releaseGroup": "NTb",
"screenSize": "1080p",
"season": 2,
"series": "House of Cards",
"title": "NF",
"type": "episode",
"videoCodec": "h264",
"year": 2013
}
"""
args = parse_options_dict(request.args)
filename = args.pop('filename')
filetype = args.pop('type', None)
options = args
log.info('[GET] Guess request: %s -- options: %s' % (filename, options))
# TODO: store request in DB
# TODO: if exception, store in list of bugs
g = guessit.guess_file_info(filename, type=filetype, options=options)
return jsonify(g)
def get_movie_info(path):
"""Find movie information from a `path` to file."""
# I've added this string to files that don't exist on OMDB
if 'omdb' in path:
return None
# Use the guessit module to find details of a movie from name
file = guess_file_info(os.path.basename(path))
# BUG: Use some heuristics here too?
if 'title' not in file:
return None
if not file['title']:
return None
# Use omdb to find ratings, genre etc. from title and year
data, url = omdb(file['title'], file.get('year'))
# Use the longest word as a title
if not data:
logger.warning('\033[35m' +
"OMDB 404 - %s. Retrying with longest word!" % url +
'\033[0m')
data, url = omdb(max(file['title'].split(), key=len), file.get('year'))
# Use the first word as title
if not data:
logger.warning('\033[35m' +
"OMDB 404 - %s. Retrying with first word!" % url +
'\033[0m')
data, url = omdb(file['title'].split()[0], file.get('year'))
# Still no luck :'(
if not data:
logger.warning('\033[35m' + "OMDB 404 - %s." % url + '\033[0m')
return data
# BUG: What if we end up fetching data of some other movie?
if file['title'] != data['title']:
logger.warning('\033[32m' + "Titles don't match: %s - %s" %
(file['title'], data['title']) + '\033[0m')
# Save the path to this movie in the data
data['movie_path'] = path
return data
def move_satellites(videofile, dest):
""" Moves satellite files such as subtitles that are associated with base
and stored in root to the correct dest.
"""
if verbose:
print('Move satellites for %s' % videofile)
root = os.path.dirname(videofile)
destbasenm = os.path.splitext(dest)[0]
base = os.path.basename(os.path.splitext(videofile)[0])
for (dirpath, dirnames, filenames) in os.walk(root):
for filename in filenames:
fbase, fext = os.path.splitext(filename)
fextlo = fext.lower()
fpath = os.path.join(dirpath, filename)
if fextlo in satellite_extensions:
# Handle subtitles and nfo files
subpart = ''
# We support GuessIt supported subtitle extensions
if fextlo[1:] in guessit.patterns.extension.subtitle_exts:
guess = guessit.guess_file_info(filename,
info=['filename'])
if guess and 'subtitleLanguage' in guess:
fbase = fbase[:fbase.rfind('.')]
# Use alpha2 subtitle language from GuessIt (en, es, de, etc.)
subpart = '.' + guess['subtitleLanguage'][0].alpha2
if verbose:
if subpart != '':
print('Satellite: %s is a subtitle [%s]' %
(filename, guess['subtitleLanguage'][0]))
else:
# English (or undetermined)
print('Satellite: %s is a subtitle' % filename)
elif (fbase.lower() != base.lower()) and fextlo == '.nfo':
# Aggressive match attempt
if deep_scan:
guess = deep_scan_nfo(fpath)
if guess is not None:
# Guess details are not important, just that there was a match
fbase = base
if fbase.lower() == base.lower():
old = fpath
new = destbasenm + subpart + fext
if verbose:
print('Satellite: %s' % os.path.basename(new))
rename(old, new)
def download(self, subtitle):
logger.info(u'Downloading %s in %s' % (subtitle.link, subtitle.path)
) if sys.platform != 'win32' else logger.debug(
'Log line suppressed on windows')
try:
r = self.session.get(subtitle.link,
headers={
'Referer': self.server_url,
'User-Agent': self.user_agent
})
zipcontent = StringIO.StringIO(r.content)
zipsub = zipfile.ZipFile(zipcontent)
# if not zipsub.is_zipfile(zipcontent):
# raise DownloadFailedError('Downloaded file is not a zip file')
subfile = ''
if len(zipsub.namelist()) == 1:
subfile = zipsub.namelist()[0]
else:
#Season Zip Retrive Season and episode Numbers from path
guess = guessit.guess_file_info(subtitle.path, 'episode')
ep_string = "s%(seasonnumber)02de%(episodenumber)02d" % {
'seasonnumber': guess['season'],
'episodenumber': guess['episodeNumber']
}
for file in zipsub.namelist():
if re.search(ep_string, file, re.I):
subfile = file
break
if os.path.splitext(subfile)[1] in EXTENSIONS:
with open(subtitle.path, 'wb') as f:
f.write(zipsub.open(subfile).read())
else:
zipsub.close()
raise DownloadFailedError('No subtitles found in zip file')
zipsub.close()
except Exception as e:
if os.path.exists(subtitle.path):
os.remove(subtitle.path)
raise DownloadFailedError(str(e))
logger.debug(
u'Download finished') if sys.platform != 'win32' else logger.debug(
'Log line suppressed on windows')
def process_file(dirname, filename):
"""
Process a file with guessit and construct the wanted_item dict.
Items used in wanted_item for type = 'episode':
- 'type'
- 'title'
- 'year'
- 'season'
- 'episode'
- 'source'
- 'quality'
- 'codec'
- 'releasegrp'
Items used in wanted_item for type = 'movie':
- 'type'
- 'title'
- 'year'
- 'source'
- 'quality'
- 'codec'
- 'releasegrp'
"""
log.info("Processing file: %s" % filename)
file_path = os.path.join(dirname, filename)
# Check minimal video file size if needed
if autosubliminal.MINVIDEOFILESIZE:
file_size = os.path.getsize(file_path)
# MINVIDEOFILESIZE is size in MB
if file_size < autosubliminal.MINVIDEOFILESIZE * 1024 * 1024:
log.warning("File size (%s) is lower than %sMB, skipping" %
(utils.humanize_bytes(file_size),
autosubliminal.MINVIDEOFILESIZE))
return
# Guess
try:
log.debug("Guessing file info")
guess = guessit.guess_file_info(file_path)
log.debug("Guess: %r" % guess)
except Exception, e:
log.error("Could not guess file info for: %s" % file_path)
log.error(e)
return
def classify():
if request.content_type == "application/json":
form = ClassifyForm(data=request.get_json(force=True))
else:
form = ClassifyForm(request.form)
if form.validate():
release_name = form.release_name.data
options = {'name_only': True}
if form.media_type.data == "unknown":
data = guessit.guess_file_info(release_name, options=options)
elif form.media_type.data == "tv":
data = guessit.guess_episode_info(release_name, options=options)
else:
data = guessit.guess_movie_info(release_name, options=options)
try:
jsonify()
return json.dumps(data, default=json_serial)
except Exception as err:
return json.dumps({"err": str(err)}, default=json_serial)
def parse_series(self, data, **kwargs):
log.debug('Parsing series: `%s` [options: %s]', data, kwargs)
guessit_options = self._guessit_options(kwargs)
if kwargs.get('name') and not guessit_options.get('strict_name'):
guessit_options['expected_series'] = [kwargs['name']]
start = time.clock()
# If no series name is provided, we don't tell guessit what kind of match we are looking for
# This prevents guessit from determining that too general of matches are series
parse_type = 'episode' if kwargs.get('name') else None
guess_result = guessit.guess_file_info(data, options=guessit_options, type=parse_type)
if guess_result.get('type') != 'episode':
# TODO: All other failures return an invalid parser. This is just a hack to match. Maybe exception instead?
class InvalidParser(object):
valid = False
return InvalidParser()
parsed = GuessitParsedSerie(data, kwargs.pop('name', None), guess_result, **kwargs)
end = time.clock()
log.debug('Parsing result: %s (in %s ms)', parsed, (end - start) * 1000)
return parsed
