Release 0.7.3

Section relative to the library was invalid. Missing references and
arguments.

HISTORY.rst

@@ -1,6 +1,29 @@
 Changelog
 =========
 
+0.7.3
+-----
+**release date:** 2013-11-22
+
+* Fix windows compatibility
+* Improve subtitle validation
+* Improve embedded subtitle languages detection
+* Improve unittests
+
+
+0.7.2
+-----
+**release date:** 2013-11-10
+
+* Fix TVSubtitles for ambiguous series
+* Add a CACHE_VERSION to force cache reloading on version change
+* Set CLI default cache expiration time to 30 days
+* Add podnapisi provider
+* Support script for languages e.g. Latn, Cyrl
+* Improve logging levels
+* Fix subtitle validation in some rare cases
+
+
 0.7.1
 -----
 **release date:** 2013-11-06

README.rst

@@ -19,6 +19,7 @@ best matching subtitles. Providers are extensible through a dedicated entry poin
 * Addic7ed
 * BierDopje
 * OpenSubtitles
+* Podnapisi
 * TheSubDB
 * TvSubtitles
 
@@ -45,7 +46,7 @@ skipping videos that already have subtitles whether they are embedded or not::
     subliminal.cache_region.configure('dogpile.cache.dbm', arguments={'filename': '/path/to/cachefile.dbm'})
 
     # scan for videos in the folder and their subtitles
-    videos = scan_videos(['/path/to/video/folder'], subtitles=True, embedded_subtitles=True) 
+    videos = subliminal.scan_videos(['/path/to/video/folder'], subtitles=True, embedded_subtitles=True, age=timedelta(weeks=1))
 
     # download
     subliminal.download_best_subtitles(videos, {Language('eng'), Language('fra')}, age=timedelta(week=1))

dev-requirements.txt

@@ -2,3 +2,4 @@ sympy>=0.7.3
 sphinx>=1.1.3
 sphinxcontrib-programoutput>=0.8
 Sphinx-PyPI-upload>=0.2.1
+setuptools>=1.4

docs/api/cache.rst

@@ -2,6 +2,10 @@ Cache
 =====
 .. module:: subliminal.cache
 
-.. autodata:: region
+.. autodata:: CACHE_VERSION
+.. autofunction:: subliminal_key_generator
+.. data:: region
 
-Refer to `dogpile.cache's documentation <http://dogpilecache.readthedocs.org>`_ to see how to configure a region
+    The dogpile.cache region
+
+Refer to `dogpile.cache's documentation <http://dogpilecache.readthedocs.org>`_ to see how to configure the region

docs/index.rst

@@ -19,6 +19,7 @@ best matching subtitles. Providers are extensible through a dedicated entry poin
 * Addic7ed
 * BierDopje
 * OpenSubtitles
+* Podnapisi
 * TheSubDB
 * TvSubtitles
 
@@ -47,7 +48,7 @@ skipping videos that already have subtitles whether they are embedded or not::
     subliminal.cache_region.configure('dogpile.cache.dbm', arguments={'filename': '/path/to/cachefile.dbm'})
 
     # scan for videos in the folder and their subtitles
-    videos = scan_videos(['/path/to/video/folder'], subtitles=True, embedded_subtitles=True) 
+    videos = subliminal.scan_videos(['/path/to/video/folder'], subtitles=True, embedded_subtitles=True, age=timedelta(weeks=1))
 
     # download
     subliminal.download_best_subtitles(videos, {Language('eng'), Language('fra')}, age=timedelta(week=1))

requirements.txt

@@ -1,9 +1,9 @@
 beautifulsoup4>=4.3.2
-guessit>=0.6.1
+guessit>=0.6.2
 requests>=2.0.1
 enzyme>=0.4.0
 html5lib>=0.99
-dogpile.cache>=0.5.1
-babelfish>=0.2.1
+dogpile.cache>=0.5.2
+babelfish>=0.4.0
 charade>=1.0.3
 pysrt>=0.5.0

setup.py

@@ -4,7 +4,7 @@ from setuptools import setup, find_packages
 
 
 setup(name='subliminal',
-    version='0.7.1',
+    version='0.7.3',
     license='MIT',
     description='Subtitles, faster than your thoughts',
     long_description=open('README.rst').read() + '\n\n' + open('HISTORY.rst').read(),
@@ -29,10 +29,12 @@ setup(name='subliminal',
         'subliminal.providers': ['addic7ed = subliminal.providers.addic7ed:Addic7edProvider',
                                  'bierdopje = subliminal.providers.bierdopje:BierDopjeProvider',
                                  'opensubtitles = subliminal.providers.opensubtitles:OpenSubtitlesProvider',
+                                 'podnapisi = subliminal.providers.podnapisi:PodnapisiProvider',
                                  'thesubdb = subliminal.providers.thesubdb:TheSubDBProvider',
                                  'tvsubtitles = subliminal.providers.tvsubtitles:TVsubtitlesProvider'],
-        'babelfish.converters': ['addic7ed = subliminal.converters.addic7ed:Addic7edConverter',
-                                 'tvsubtitles = subliminal.converters.tvsubtitles:TVsubtitlesConverter']
+        'babelfish.language_converters': ['addic7ed = subliminal.converters.addic7ed:Addic7edConverter',
+                                          'podnapisi = subliminal.converters.podnapisi:PodnapisiConverter',
+                                          'tvsubtitles = subliminal.converters.tvsubtitles:TVsubtitlesConverter']
     },
     install_requires=open('requirements.txt').readlines(),
     test_suite='subliminal.tests.suite')

subliminal/__init__.py

@@ -1,13 +1,13 @@
 # -*- coding: utf-8 -*-
 __title__ = 'subliminal'
-__version__ = '0.7.1'
+__version__ = '0.7.3'
 __author__ = 'Antoine Bertin'
 __license__ = 'MIT'
 __copyright__ = 'Copyright 2013 Antoine Bertin'
 
 import logging
 from .api import PROVIDERS_ENTRY_POINT, list_subtitles, download_subtitles, download_best_subtitles
-from .cache import region as cache_region
+from .cache import MutexLock, region as cache_region
 from .exceptions import Error, ProviderError, ProviderConfigurationError, ProviderNotAvailable, InvalidSubtitle
 from .subtitle import Subtitle
 from .video import VIDEO_EXTENSIONS, SUBTITLE_EXTENSIONS, Video, Episode, Movie, scan_videos, scan_video

subliminal/api.py

@@ -47,11 +47,11 @@ def list_subtitles(videos, languages, providers=None, provider_configs=None):
         Provider = provider_entry_point.load()
         provider_languages = Provider.languages & languages - subtitle_languages
         if not provider_languages:
-            logger.debug('Skipping provider %r: no language to search for', provider_entry_point.name)
+            logger.info('Skipping provider %r: no language to search for', provider_entry_point.name)
             continue
         provider_videos = [v for v in videos if Provider.check(v)]
         if not provider_videos:
-            logger.debug('Skipping provider %r: no video to search for', provider_entry_point.name)
+            logger.info('Skipping provider %r: no video to search for', provider_entry_point.name)
             continue
 
         # list subtitles with the provider
@@ -184,10 +184,10 @@ def download_best_subtitles(videos, languages, providers=None, provider_configs=
             continue
         Provider = provider_entry_point.load()
         if not Provider.languages & languages - subtitle_languages:
-            logger.debug('Skipping provider %r: no language to search for', provider_entry_point.name)
+            logger.info('Skipping provider %r: no language to search for', provider_entry_point.name)
             continue
         if not [v for v in videos if Provider.check(v)]:
-            logger.debug('Skipping provider %r: no video to search for', provider_entry_point.name)
+            logger.info('Skipping provider %r: no video to search for', provider_entry_point.name)
             continue
         provider = Provider(**provider_configs.get(provider_entry_point.name, {}))
         try:

subliminal/cache.py

@@ -1,6 +1,56 @@
 # -*- coding: utf-8 -*-
-import dogpile.cache
+import inspect
+from dogpile.cache import make_region  # @UnresolvedImport
+from dogpile.cache.backends.file import AbstractFileLock  # @UnresolvedImport
+from dogpile.cache.compat import string_type  # @UnresolvedImport
+from dogpile.core.readwrite_lock import ReadWriteMutex  # @UnresolvedImport
 
 
-#: The subliminal's dogpile.cache region
-region = dogpile.cache.make_region()
+#: Subliminal's cache version
+CACHE_VERSION = 1
+
+
+def subliminal_key_generator(namespace, fn, to_str=string_type):
+    """Add a :data:`CACHE_VERSION` to dogpile.cache's default function_key_generator"""
+    if namespace is None:
+        namespace = '%d:%s:%s' % (CACHE_VERSION, fn.__module__, fn.__name__)
+    else:
+        namespace = '%d:%s:%s|%s' % (CACHE_VERSION, fn.__module__, fn.__name__, namespace)
+
+    args = inspect.getargspec(fn)
+    has_self = args[0] and args[0][0] in ('self', 'cls')
+
+    def generate_key(*args, **kw):
+        if kw:
+            raise ValueError('Keyword arguments not supported')
+        if has_self:
+            args = args[1:]
+        return namespace + '|' + ' '.join(map(to_str, args))
+    return generate_key
+
+
+class MutexLock(AbstractFileLock):
+    """:class:`MutexLock` is a thread-based rw lock based on :class:`dogpile.core.ReadWriteMutex`"""
+    def __init__(self, filename):
+        self.mutex = ReadWriteMutex()
+
+    def acquire_read_lock(self, wait):
+        ret = self.mutex.acquire_read_lock(wait)
+        return wait or ret
+
+    def acquire_write_lock(self, wait):
+        ret = self.mutex.acquire_write_lock(wait)
+        return wait or ret
+
+    def release_read_lock(self):
+        return self.mutex.release_read_lock()
+
+    def release_write_lock(self):
+        return self.mutex.release_write_lock()
+
+
+#: The dogpile.cache region (long-lived)
+region = make_region(function_key_generator=subliminal_key_generator)
+
+#: The dogpile.cache region for :meth:`~subliminal.providers.Provider.query` (short-lived)
+query_region = make_region(function_key_generator=subliminal_key_generator)

subliminal/cli.py

@@ -9,7 +9,7 @@ import sys
 import babelfish
 import guessit
 import pkg_resources
-from subliminal import (__version__, PROVIDERS_ENTRY_POINT, cache_region, Video, Episode, Movie, scan_videos,
+from subliminal import (__version__, PROVIDERS_ENTRY_POINT, cache_region, MutexLock, Video, Episode, Movie, scan_videos,
     download_best_subtitles)
 try:
     import colorlog
@@ -29,7 +29,7 @@ def subliminal():
     required_arguments_group = parser.add_argument_group('required arguments')
     required_arguments_group.add_argument('paths', nargs='+', metavar='PATH', help='path to video file or folder')
     required_arguments_group.add_argument('-l', '--languages', nargs='+', required=True, metavar='LANGUAGE',
-                                          help='wanted languages as alpha2 code (ISO-639-1)')
+                                          help='wanted languages as IETF codes e.g. fr, pt-BR, sr-Cyrl ')
 
     # configuration
     configuration_group = parser.add_argument_group('configuration')
@@ -81,9 +81,9 @@ def subliminal():
 
     # parse languages
     try:
-        args.languages = {babelfish.Language.fromalpha2(l) for l in args.languages}
+        args.languages = {babelfish.Language.fromietf(l) for l in args.languages}
     except babelfish.Error:
-        parser.error('argument -l/--languages: codes are not ISO-639-1: %r' % args.languages)
+        parser.error('argument -l/--languages: codes are not IETF: %r' % args.languages)
 
     # parse age
     if args.age is not None:
@@ -138,7 +138,8 @@ def subliminal():
         logging.getLogger('subliminal.api').setLevel(logging.INFO)
 
     # configure cache
-    cache_region.configure('dogpile.cache.dbm', arguments={'filename': args.cache_file})
+    cache_region.configure('dogpile.cache.dbm', expiration_time=datetime.timedelta(days=30),  # @UndefinedVariable
+                           arguments={'filename': args.cache_file, 'lock_factory': MutexLock})
 
     # scan videos
     videos = scan_videos([p for p in args.paths if os.path.exists(p)], subtitles=not args.force,

subliminal/converters/addic7ed.py

@@ -1,29 +1,31 @@
 # -*- coding: utf-8 -*-
 from __future__ import unicode_literals
-from babelfish.converters.name import NameConverter
+from babelfish import LanguageReverseConverter, get_language_converter
 
 
-class Addic7edConverter(NameConverter):
+class Addic7edConverter(LanguageReverseConverter):
     def __init__(self):
-        super(Addic7edConverter, self).__init__()
-        self.from_addic7ed = {'Català': ('cat', None), 'Chinese (Simplified)': ('zho', None),
-                              'Chinese (Traditional)': ('zho', None), 'Euskera': ('eus', None),
-                              'Galego': ('glg', None), 'Greek': ('ell', None),
-                              'Malay': ('msa', None), 'Portuguese (Brazilian)': ('por', 'BR'),
-                              'Serbian (Cyrillic)': ('srp', None), 'Serbian (Latin)': ('srp', None),
-                              'Spanish (Latin America)': ('spa', None), 'Spanish (Spain)': ('spa', None)}
-        self.to_addic7ed = {('cat', None): 'Català', ('zho', None): 'Chinese (Simplified)',
-                            ('eus', None): 'Euskera', ('glg', None): 'Galego',
-                            ('ell', None): 'Greek', ('msa', None): 'Malay',
-                            ('por', 'BR'): 'Portuguese (Brazilian)', ('srp', None): 'Serbian (Cyrillic)'}
-        self.codes |= set(self.from_addic7ed.keys())
+        self.name_converter = get_language_converter('name')
+        self.from_addic7ed = {'Català': ('cat',), 'Chinese (Simplified)': ('zho',), 'Chinese (Traditional)': ('zho',),
+                              'Euskera': ('eus',), 'Galego': ('glg',), 'Greek': ('ell',), 'Malay': ('msa',),
+                              'Portuguese (Brazilian)': ('por', 'BR'), 'Serbian (Cyrillic)': ('srp', None, 'Cyrl'),
+                              'Serbian (Latin)': ('srp',), 'Spanish (Latin America)': ('spa',),
+                              'Spanish (Spain)': ('spa',)}
+        self.to_addic7ed = {('cat',): 'Català', ('zho',): 'Chinese (Simplified)', ('eus',): 'Euskera',
+                            ('glg',): 'Galego', ('ell',): 'Greek', ('msa',): 'Malay',
+                            ('por', 'BR'): 'Portuguese (Brazilian)', ('srp', None, 'Cyrl'): 'Serbian (Cyrillic)'}
+        self.codes = self.name_converter.codes | set(self.from_addic7ed.keys())
 
-    def convert(self, alpha3, country=None):
+    def convert(self, alpha3, country=None, script=None):
+        if (alpha3, country, script) in self.to_addic7ed:
+            return self.to_addic7ed[(alpha3, country, script)]
         if (alpha3, country) in self.to_addic7ed:
             return self.to_addic7ed[(alpha3, country)]
-        return super(Addic7edConverter, self).convert(alpha3, country)
+        if (alpha3,) in self.to_addic7ed:
+            return self.to_addic7ed[(alpha3,)]
+        return self.name_converter.convert(alpha3, country, script)
 
     def reverse(self, addic7ed):
         if addic7ed in self.from_addic7ed:
             return self.from_addic7ed[addic7ed]
-        return super(Addic7edConverter, self).reverse(addic7ed)
+        return self.name_converter.reverse(addic7ed)

subliminal/converters/podnapisi.py

@@ -0,0 +1,32 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+from babelfish import LanguageReverseConverter, LanguageConvertError, LanguageReverseError
+
+
+class PodnapisiConverter(LanguageReverseConverter):
+    def __init__(self):
+        self.from_podnapisi = {2: ('eng',), 28: ('spa',), 26: ('pol',), 36: ('srp',), 1: ('slv',), 38: ('hrv',),
+                               9: ('ita',), 8: ('fra',), 48: ('por', 'BR'), 23: ('nld',), 12: ('ara',), 13: ('ron',),
+                               33: ('bul',), 32: ('por',), 16: ('ell',), 15: ('hun',), 31: ('fin',), 30: ('tur',),
+                               7: ('ces',), 25: ('swe',), 27: ('rus',), 24: ('dan',), 22: ('heb',), 51: ('vie',),
+                               52: ('fas',), 5: ('deu',), 14: ('spa', 'AR'), 54: ('ind',), 47: ('srp', None, 'Cyrl'),
+                               3: ('nor',), 20: ('est',), 10: ('bos',), 17: ('zho',), 37: ('slk',), 35: ('mkd',),
+                               11: ('jpn',), 4: ('kor',), 29: ('sqi',), 6: ('isl',), 19: ('lit',), 46: ('ukr',),
+                               44: ('tha',), 53: ('cat',), 56: ('sin',), 21: ('lav',), 40: ('cmn',), 55: ('msa',),
+                               42: ('hin',), 50: ('bel',)}
+        self.to_podnapisi = {v: k for k, v in self.from_podnapisi.items()}
+        self.codes = set(self.from_podnapisi.keys())
+
+    def convert(self, alpha3, country=None, script=None):
+        if (alpha3,) in self.to_podnapisi:
+            return self.to_podnapisi[(alpha3,)]
+        if (alpha3, country) in self.to_podnapisi:
+            return self.to_podnapisi[(alpha3, country)]
+        if (alpha3, country, script) in self.to_podnapisi:
+            return self.to_podnapisi[(alpha3, country, script)]
+        raise LanguageConvertError(alpha3, country, script)
+
+    def reverse(self, podnapisi):
+        if podnapisi not in self.from_podnapisi:
+            raise LanguageReverseError(podnapisi)
+        return self.from_podnapisi[podnapisi]

subliminal/converters/tvsubtitles.py

@@ -1,22 +1,24 @@
 # -*- coding: utf-8 -*-
 from __future__ import unicode_literals
-from babelfish.converters.alpha2 import Alpha2Converter
+from babelfish import LanguageReverseConverter, get_language_converter
 
 
-class TVsubtitlesConverter(Alpha2Converter):
+class TVsubtitlesConverter(LanguageReverseConverter):
     def __init__(self):
-        super(TVsubtitlesConverter, self).__init__()
-        self.from_tvsubtitles = {'br': ('por', 'BR'), 'ua': ('ukr', None), 'gr': ('ell', None), 'cn': ('zho', None),
-                                 'jp': ('jpn', None), 'cz': ('ces', None)}
+        self.alpha2_converter = get_language_converter('alpha2')
+        self.from_tvsubtitles = {'br': ('por', 'BR'), 'ua': ('ukr',), 'gr': ('ell',), 'cn': ('zho',), 'jp': ('jpn',),
+                                 'cz': ('ces',)}
         self.to_tvsubtitles = {v: k for k, v in self.from_tvsubtitles}
-        self.codes |= set(self.from_tvsubtitles.keys())
+        self.codes = self.alpha2_converter.codes | set(self.from_tvsubtitles.keys())
 
-    def convert(self, alpha3, country=None):
+    def convert(self, alpha3, country=None, script=None):
         if (alpha3, country) in self.to_tvsubtitles:
             return self.to_tvsubtitles[(alpha3, country)]
-        return super(TVsubtitlesConverter, self).convert(alpha3, country)
+        if (alpha3,) in self.to_tvsubtitles:
+            return self.to_tvsubtitles[(alpha3,)]
+        return self.alpha2_converter.convert(alpha3, country, script)
 
     def reverse(self, tvsubtitles):
         if tvsubtitles in self.from_tvsubtitles:
             return self.from_tvsubtitles[tvsubtitles]
-        return super(TVsubtitlesConverter, self).reverse(tvsubtitles)
+        return self.alpha2_converter.reverse(tvsubtitles)

subliminal/providers/addic7ed.py

@@ -166,6 +166,9 @@ class Addic7edProvider(Provider):
             if cells[5].string != 'Completed':
                 logger.debug('Skipping incomplete subtitle')
                 continue
+            if not cells[3].string:
+                logger.debug('Skipping empty language')
+                continue
             subtitles.append(Addic7edSubtitle(babelfish.Language.fromaddic7ed(cells[3].string), series, season,
                                               int(cells[1].string), cells[2].string, cells[4].string,
                                               bool(cells[6].string), cells[9].a['href'], link))

subliminal/providers/bierdopje.py

@@ -118,10 +118,7 @@ class BierDopjeProvider(Provider):
                                   result.find('downloadlink').text) for result in root.find('response/results')]
 
     def list_subtitles(self, video, languages):
-        subtitles = []
-        for language in languages:
-            subtitles.extend(self.query(language, video.season, video.episode, video.tvdb_id, video.series))
-        return subtitles
+        return [s for l in languages for s in self.query(l, video.season, video.episode, video.tvdb_id, video.series)]
 
     def download_subtitle(self, subtitle):
         try:

subliminal/providers/opensubtitles.py

@@ -23,8 +23,8 @@ class OpenSubtitlesSubtitle(Subtitle):
     provider_name = 'opensubtitles'
     series_re = re.compile('^"(?P<series_name>.*)" (?P<series_title>.*)$')
 
-    def __init__(self, language, hearing_impaired, id, matched_by, movie_kind, hash, movie_name, movie_release_name, movie_year,
-                 movie_imdb_id, series_season, series_episode):
+    def __init__(self, language, hearing_impaired, id, matched_by, movie_kind, hash, movie_name, movie_release_name,  # @ReservedAssignment
+                 movie_year, movie_imdb_id, series_season, series_episode):
         super(OpenSubtitlesSubtitle, self).__init__(language, hearing_impaired)
         self.id = id
         self.matched_by = matched_by
@@ -83,7 +83,7 @@ class OpenSubtitlesSubtitle(Subtitle):
 
 
 class OpenSubtitlesProvider(Provider):
-    languages = {babelfish.Language.fromopensubtitles(l) for l in babelfish.CONVERTERS['opensubtitles'].codes}
+    languages = {babelfish.Language.fromopensubtitles(l) for l in babelfish.get_language_converter('opensubtitles').codes}
 
     def __init__(self):
         self.server = xmlrpclib.ServerProxy('http://api.opensubtitles.org/xml-rpc')
@@ -106,7 +106,7 @@ class OpenSubtitlesProvider(Provider):
         if response['status'] != '200 OK':
             raise ProviderError('Logout failed with status %r' % response['status'])
 
-    def query(self, languages, hash=None, size=None, imdb_id=None, query=None):
+    def query(self, languages, hash=None, size=None, imdb_id=None, query=None):  # @ReservedAssignment
         searches = []
         if hash and size:
             searches.append({'moviehash': hash, 'moviebytesize': str(size)})
@@ -128,7 +128,6 @@ class OpenSubtitlesProvider(Provider):
         if not response['data']:
             logger.debug('No subtitle found')
             return []
-        logger.debug('Found subtitles %r', response['data'])
         return [OpenSubtitlesSubtitle(babelfish.Language.fromopensubtitles(r['SubLanguageID']),
                                       bool(int(r['SubHearingImpaired'])), r['IDSubtitleFile'], r['MatchedBy'],
                                       r['MovieKind'], r['MovieHash'], r['MovieName'], r['MovieReleaseName'],
@@ -141,7 +140,8 @@ class OpenSubtitlesProvider(Provider):
         query = None
         if ('opensubtitles' not in video.hashes or not video.size) and not video.imdb_id:
             query = video.name.split(os.sep)[-1]
-        return self.query(languages, hash=video.hashes.get('opensubtitles'), size=video.size, imdb_id=video.imdb_id, query=query)
+        return self.query(languages, hash=video.hashes.get('opensubtitles'), size=video.size, imdb_id=video.imdb_id,
+                          query=query)
 
     def download_subtitle(self, subtitle):
         try:

subliminal/providers/podnapisi.py

@@ -0,0 +1,163 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+import io
+import logging
+import re
+import xml.etree.ElementTree
+import zipfile
+import babelfish
+import bs4
+import charade
+import guessit
+import requests
+from . import Provider
+from .. import __version__
+from ..exceptions import InvalidSubtitle, ProviderNotAvailable, ProviderError
+from ..subtitle import Subtitle, is_valid_subtitle, compute_guess_matches
+from ..video import Episode, Movie
+
+
+logger = logging.getLogger(__name__)
+
+
+class PodnapisiSubtitle(Subtitle):
+    provider_name = 'podnapisi'
+
+    def __init__(self, language, id, releases, hearing_impaired, link, series=None, season=None, episode=None,  # @ReservedAssignment
+                 title=None, year=None):
+        super(PodnapisiSubtitle, self).__init__(language, hearing_impaired)
+        self.id = id
+        self.releases = releases
+        self.hearing_impaired = hearing_impaired
+        self.link = link
+        self.series = series
+        self.season = season
+        self.episode = episode
+        self.title = title
+        self.year = year
+
+    def compute_matches(self, video):
+        matches = set()
+        # episode
+        if isinstance(video, Episode):
+            # series
+            if video.series and self.series.lower() == video.series.lower():
+                matches.add('series')
+            # season
+            if video.season and self.season == video.season:
+                matches.add('season')
+            # episode
+            if video.episode and self.episode == video.episode:
+                matches.add('episode')
+            # guess
+            for release in self.releases:
+                matches |= compute_guess_matches(video, guessit.guess_episode_info(release + '.mkv'))
+        # movie
+        elif isinstance(video, Movie):
+            # title
+            if video.title and self.title.lower() == video.title.lower():
+                matches.add('title')
+            # year
+            if video.year and self.year == video.year:
+                matches.add('year')
+            # guess
+            for release in self.releases:
+                matches |= compute_guess_matches(video, guessit.guess_movie_info(release + '.mkv'))
+        return matches
+
+
+class PodnapisiProvider(Provider):
+    languages = {babelfish.Language.frompodnapisi(l) for l in babelfish.get_language_converter('podnapisi').codes}
+    video_types = (Episode, Movie)
+    server = 'http://simple.podnapisi.net'
+    link_re = re.compile('^.*(?P<link>/ppodnapisi/download/i/\d+/k/.*$)')
+
+    def initialize(self):
+        self.session = requests.Session()
+        self.session.headers = {'User-Agent': 'Subliminal/%s' % __version__}
+
+    def terminate(self):
+        self.session.close()
+
+    def get(self, url, params=None, is_xml=True):
+        """Make a GET request on `url` with the given parameters
+
+        :param string url: part of the URL to reach with the leading slash
+        :param dict params: params of the request
+        :param bool xml: whether the response content is XML or not
+        :return: the response
+        :rtype: :class:`xml.etree.ElementTree.Element` or :class:`bs4.BeautifulSoup`
+        :raise: :class:`~subliminal.exceptions.ProviderNotAvailable`
+
+        """
+        try:
+            r = self.session.get(self.server + '/ppodnapisi' + url, params=params, timeout=10)
+        except requests.Timeout:
+            raise ProviderNotAvailable('Timeout after 10 seconds')
+        if r.status_code != 200:
+            raise ProviderNotAvailable('Request failed with status code %d' % r.status_code)
+        if is_xml:
+            return xml.etree.ElementTree.fromstring(r.content)
+        else:
+            return bs4.BeautifulSoup(r.content, ['permissive'])
+
+    def query(self, language, series=None, season=None, episode=None, title=None, year=None):
+        params = {'sXML': 1, 'sJ': language.podnapisi}
+        if series and season and episode:
+            params['sK'] = series
+            params['sTS'] = season
+            params['sTE'] = episode
+        elif title:
+            params['sK'] = title
+            if year:
+                params['sY'] = year
+        else:
+            raise ValueError('Missing parameters series and season and episode or title')
+        logger.debug('Searching episode %r', params)
+        subtitles = []
+        while True:
+            root = self.get('/search', params)
+            if not int(root.find('pagination/results').text):
+                logger.debug('No subtitle found')
+                break
+            if series and season and episode:
+                subtitles.extend([PodnapisiSubtitle(language, int(s.find('id').text), s.find('release').text.split(),
+                                                    'h' in (s.find('flags').text or ''), s.find('url').text[38:],
+                                                    series=series, season=season, episode=episode)
+                                  for s in root.findall('subtitle')])
+            elif title:
+                subtitles.extend([PodnapisiSubtitle(language, int(s.find('id').text), s.find('release').text.split(),
+                                                    'h' in (s.find('flags').text or ''), s.find('url').text[38:],
+                                                    title=title, year=year)
+                                  for s in root.findall('subtitle')])
+            if int(root.find('pagination/current').text) >= int(root.find('pagination/count').text):
+                break
+            params['page'] = int(root.find('pagination/current').text) + 1
+        return subtitles
+
+    def list_subtitles(self, video, languages):
+        if isinstance(video, Episode):
+            return [s for l in languages for s in self.query(l, series=video.series, season=video.season,
+                                                             episode=video.episode)]
+        elif isinstance(video, Movie):
+            return [s for l in languages for s in self.query(l, title=video.title, year=video.year)]
+
+    def download_subtitle(self, subtitle):
+        soup = self.get(subtitle.link, is_xml=False)
+        link = soup.find('a', href=self.link_re)
+        if not link:
+            raise ProviderError('Cannot find the download link')
+        try:
+            r = self.session.get(self.server + self.link_re.match(link['href']).group('link'), timeout=10)
+        except requests.Timeout:
+            raise ProviderNotAvailable('Timeout after 10 seconds')
+        if r.status_code != 200:
+            raise ProviderNotAvailable('Request failed with status code %d' % r.status_code)
+        with zipfile.ZipFile(io.BytesIO(r.content)) as zf:
+            if len(zf.namelist()) > 1:
+                raise ProviderError('More than one file to unzip')
+            subtitle_bytes = zf.read(zf.namelist()[0])
+        subtitle_text = subtitle_bytes.decode(charade.detect(subtitle_bytes)['encoding'], 'replace')
+        if not is_valid_subtitle(subtitle_text):
+            raise InvalidSubtitle
+        return subtitle_text

subliminal/providers/thesubdb.py

@@ -16,7 +16,7 @@ logger = logging.getLogger(__name__)
 class TheSubDBSubtitle(Subtitle):
     provider_name = 'thesubdb'
 
-    def __init__(self, language, hash):
+    def __init__(self, language, hash):  # @ReservedAssignment
         super(TheSubDBSubtitle, self).__init__(language)
         self.hash = hash
 
@@ -55,7 +55,7 @@ class TheSubDBProvider(Provider):
             raise ProviderNotAvailable('Timeout after 10 seconds')
         return r
 
-    def query(self, hash):
+    def query(self, hash):  # @ReservedAssignment
         params = {'action': 'search', 'hash': hash}
         logger.debug('Searching subtitles %r', params)
         r = self.get(params)

subliminal/providers/tvsubtitles.py

@@ -22,7 +22,7 @@ logger = logging.getLogger(__name__)
 class TVsubtitlesSubtitle(Subtitle):
     provider_name = 'tvsubtitles'
 
-    def __init__(self, language, series, season, episode, id, rip, release):
+    def __init__(self, language, series, season, episode, id, rip, release):  # @ReservedAssignment
         super(TVsubtitlesSubtitle, self).__init__(language)
         self.series = series
         self.season = season
@@ -61,8 +61,9 @@ class TVsubtitlesProvider(Provider):
                            'nld', 'pol', 'por', 'ron', 'rus', 'spa', 'swe', 'tur', 'ukr', 'zho']}
     video_types = (Episode,)
     server = 'http://www.tvsubtitles.net'
-    episode_id_re = re.compile('^episode-(\d+)\.html$')
-    subtitle_re = re.compile('^\/subtitle-(\d+)\.html$')
+    episode_id_re = re.compile('^episode-\d+\.html$')
+    subtitle_re = re.compile('^\/subtitle-\d+\.html$')
+    link_re = re.compile('^(?P<series>.+) \(\d{4}-\d{4}\)$')
 
     def initialize(self):
         self.session = requests.Session()
@@ -77,6 +78,7 @@ class TVsubtitlesProvider(Provider):
         :param string url: part of the URL to reach with the leading slash
         :param dict params: params of the request
         :param dict data: data of the request
+        :param string method: method of the request
         :return: the response
         :rtype: :class:`bs4.BeautifulSoup`
         :raise: :class:`~subliminal.exceptions.ProviderNotAvailable`
@@ -106,6 +108,13 @@ class TVsubtitlesProvider(Provider):
         if not links:
             logger.info('Series %r not found', series)
             return None
+        for link in links:
+            match = self.link_re.match(link.string)
+            if not match:
+                logger.warning('Could not parse %r', link.string)
+                continue
+            if match.group('series').lower().replace('.', ' ').strip() == series.lower():
+                return int(link['href'][8:-5])
         return int(links[0]['href'][8:-5])
 
     @region.cache_on_arguments()

subliminal/score.py

@@ -6,7 +6,7 @@ from sympy import Eq, symbols, solve
 
 # Symbols
 release_group, resolution, video_codec, audio_codec = symbols('release_group resolution video_codec audio_codec')
-imdb_id, hash, title, series, tvdb_id, season, episode = symbols('imdb_id hash title series tvdb_id season episode')
+imdb_id, hash, title, series, tvdb_id, season, episode = symbols('imdb_id hash title series tvdb_id season episode')  # @ReservedAssignment
 year = symbols('year')
 
 

subliminal/subtitle.py

@@ -73,7 +73,7 @@ class Subtitle(object):
         return score
 
     def __repr__(self):
-        return '<%s [%r]>' % (self.__class__.__name__, self.language)
+        return '<%s [%s]>' % (self.__class__.__name__, self.language)
 
 
 def get_subtitle_path(video_path, language=None):
@@ -90,7 +90,7 @@ def get_subtitle_path(video_path, language=None):
     if language is not None:
         try:
             return subtitle_path + '.%s.%s' % (language.alpha2, 'srt')
-        except babelfish.ConvertError:
+        except babelfish.LanguageConvertError:
             return subtitle_path + '.%s.%s' % (language.alpha3, 'srt')
     return subtitle_path + '.srt'
 
@@ -105,8 +105,12 @@ def is_valid_subtitle(subtitle_text):
     try:
         pysrt.from_string(subtitle_text, error_handling=pysrt.ERROR_RAISE)
         return True
-    except pysrt.Error:
-        return False
+    except pysrt.Error as e:
+        if e.args[0] > 80:
+            return True
+    except:
+        logger.exception('Unexpected error when validating subtitle')
+    return False
 
 
 def compute_guess_matches(video, guess):

subliminal/tests/__init__.py

@@ -6,7 +6,7 @@ from subliminal import cache_region
 from . import test_providers, test_subliminal
 
 
-cache_region.configure('dogpile.cache.memory', expiration_time=60 * 30)
+cache_region.configure('dogpile.cache.memory', expiration_time=60 * 30)  # @UndefinedVariable
 suite = TestSuite([test_providers.suite(), test_subliminal.suite()])
 
 

subliminal/tests/test_providers.py

@@ -25,17 +25,18 @@ class Addic7edProviderTestCase(ProviderTestCase):
     def test_find_show_id(self):
         with self.Provider() as provider:
             show_id = provider.find_show_id('The Big Bang')
-        self.assertTrue(show_id == 126)
+        self.assertEqual(show_id, 126)
 
     def test_find_show_id_error(self):
         with self.Provider() as provider:
             show_id = provider.find_show_id('the big how i met your mother')
-        self.assertTrue(show_id is None)
+        self.assertIsNone(show_id)
 
     def test_get_show_ids(self):
         with self.Provider() as provider:
             show_ids = provider.get_show_ids()
-        self.assertTrue('the big bang theory' in show_ids and show_ids['the big bang theory'] == 126)
+        self.assertIn('the big bang theory', show_ids)
+        self.assertEqual(show_ids['the big bang theory'], 126)
 
     def test_query_episode_0(self):
         video = EPISODES[0]
@@ -50,8 +51,8 @@ class Addic7edProviderTestCase(ProviderTestCase):
                    frozenset(['series', 'season'])}
         with self.Provider() as provider:
             subtitles = provider.query(video.series, video.season)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_episode_1(self):
         video = EPISODES[1]
@@ -68,8 +69,8 @@ class Addic7edProviderTestCase(ProviderTestCase):
                    frozenset(['series', 'season'])}
         with self.Provider() as provider:
             subtitles = provider.query(video.series, video.season)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_list_subtitles(self):
         video = EPISODES[0]
@@ -78,8 +79,8 @@ class Addic7edProviderTestCase(ProviderTestCase):
                    frozenset(['series', 'episode', 'season', 'title'])}
         with self.Provider() as provider:
             subtitles = provider.list_subtitles(video, languages)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_download_subtitle(self):
         video = EPISODES[0]
@@ -96,12 +97,12 @@ class BierDopjeProviderTestCase(ProviderTestCase):
     def test_find_show_id(self):
         with self.Provider() as provider:
             show_id = provider.find_show_id('The Big Bang')
-        self.assertTrue(show_id == 9203)
+        self.assertEqual(show_id, 9203)
 
     def test_find_show_id_error(self):
         with self.Provider() as provider:
             show_id = provider.find_show_id('the big how i met your mother')
-        self.assertTrue(show_id is None)
+        self.assertIsNone(show_id)
 
     def test_query_episode_0(self):
         video = EPISODES[0]
@@ -111,8 +112,8 @@ class BierDopjeProviderTestCase(ProviderTestCase):
                    frozenset(['episode', 'video_codec', 'season', 'series', 'resolution', 'release_group'])}
         with self.Provider() as provider:
             subtitles = provider.query(language, video.season, video.episode, series=video.series)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == {language})
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, {language})
 
     def test_query_episode_1(self):
         video = EPISODES[1]
@@ -124,8 +125,8 @@ class BierDopjeProviderTestCase(ProviderTestCase):
                    frozenset(['episode', 'video_codec', 'season', 'series', 'resolution', 'release_group'])}
         with self.Provider() as provider:
             subtitles = provider.query(language, video.season, video.episode, series=video.series)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == {language})
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, {language})
 
     def test_query_episode_0_tvdb_id(self):
         video = EPISODES[0]
@@ -135,8 +136,8 @@ class BierDopjeProviderTestCase(ProviderTestCase):
                    frozenset(['episode', 'series', 'video_codec', 'tvdb_id', 'resolution', 'season'])}
         with self.Provider() as provider:
             subtitles = provider.query(language, video.season, video.episode, tvdb_id=video.tvdb_id)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == {language})
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, {language})
 
     def test_list_subtitles(self):
         video = EPISODES[1]
@@ -148,8 +149,8 @@ class BierDopjeProviderTestCase(ProviderTestCase):
                    frozenset(['episode', 'video_codec', 'season', 'series', 'tvdb_id', 'release_group'])}
         with self.Provider() as provider:
             subtitles = provider.list_subtitles(video, languages)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_download_subtitle(self):
         video = EPISODES[0]
@@ -173,8 +174,8 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['imdb_id', 'title', 'year', 'video_codec', 'resolution', 'release_group'])}
         with self.Provider() as provider:
             subtitles = provider.query(languages, query=video.title)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_episode_0_query(self):
         video = EPISODES[0]
@@ -184,8 +185,8 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['episode', 'title', 'series', 'imdb_id', 'video_codec', 'season'])}
         with self.Provider() as provider:
             subtitles = provider.query(languages, query=video.name.split(os.sep)[-1])
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_episode_1_query(self):
         video = EPISODES[1]
@@ -198,8 +199,8 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['series', 'episode', 'season', 'imdb_id'])}
         with self.Provider() as provider:
             subtitles = provider.query(languages, query=video.name.split(os.sep)[-1])
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_movie_0_imdb_id(self):
         video = MOVIES[0]
@@ -211,8 +212,8 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['imdb_id', 'resolution', 'title', 'year'])}
         with self.Provider() as provider:
             subtitles = provider.query(languages, imdb_id=video.imdb_id)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_episode_0_imdb_id(self):
         video = EPISODES[0]
@@ -223,8 +224,8 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['episode', 'title', 'series', 'imdb_id', 'video_codec', 'season'])}
         with self.Provider() as provider:
             subtitles = provider.query(languages, imdb_id=video.imdb_id)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_movie_0_hash(self):
         video = MOVIES[0]
@@ -232,12 +233,13 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
         matches = {frozenset(['hash', 'title', 'video_codec', 'year', 'resolution', 'imdb_id']),
                    frozenset(['hash', 'title', 'video_codec', 'year', 'resolution', 'release_group', 'imdb_id']),
                    frozenset(['year', 'video_codec', 'imdb_id', 'hash', 'title']),
+                   frozenset([]),
                    frozenset(['year', 'resolution', 'imdb_id', 'hash', 'title']),
                    frozenset(['year', 'imdb_id', 'hash', 'title'])}
         with self.Provider() as provider:
             subtitles = provider.query(languages, hash=video.hashes['opensubtitles'], size=video.size)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_episode_0_hash(self):
         video = EPISODES[0]
@@ -248,8 +250,8 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['series', 'resolution', 'hash', 'video_codec'])}
         with self.Provider() as provider:
             subtitles = provider.query(languages, hash=video.hashes['opensubtitles'], size=video.size)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_list_subtitles(self):
         video = MOVIES[0]
@@ -259,6 +261,7 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['year', 'video_codec', 'imdb_id', 'resolution', 'title']),
                    frozenset(['hash', 'title', 'video_codec', 'year', 'resolution', 'release_group', 'imdb_id']),
                    frozenset(['year', 'video_codec', 'imdb_id', 'hash', 'title']),
+                   frozenset([]),
                    frozenset(['year', 'resolution', 'imdb_id', 'hash', 'title']),
                    frozenset(['hash', 'title', 'video_codec', 'year', 'resolution', 'imdb_id']),
                    frozenset(['year', 'imdb_id', 'hash', 'title']),
@@ -266,8 +269,58 @@ class OpenSubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['year', 'imdb_id', 'resolution', 'title'])}
         with self.Provider() as provider:
             subtitles = provider.list_subtitles(video, languages)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
+
+    def test_download_subtitle(self):
+        video = MOVIES[0]
+        languages = {Language('eng'), Language('fra')}
+        with self.Provider() as provider:
+            subtitles = provider.list_subtitles(video, languages)
+            subtitle_text = provider.download_subtitle(subtitles[0])
+        self.assertTrue(is_valid_subtitle(subtitle_text))
+
+
+class PodnapisiProviderTestCase(ProviderTestCase):
+    provider_name = 'podnapisi'
+
+    def test_query_movie_0(self):
+        video = MOVIES[0]
+        language = Language('eng')
+        matches = {frozenset(['video_codec', 'title', 'resolution', 'year']),
+                   frozenset(['title', 'resolution', 'year']),
+                   frozenset(['video_codec', 'title', 'year']),
+                   frozenset(['title', 'year']),
+                   frozenset(['video_codec', 'title', 'resolution', 'release_group', 'year']),
+                   frozenset(['video_codec', 'title', 'resolution', 'audio_codec', 'year'])}
+        with self.Provider() as provider:
+            subtitles = provider.query(language, title=video.title, year=video.year)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, {language})
+
+    def test_query_episode_0(self):
+        video = EPISODES[0]
+        language = Language('eng')
+        matches = {frozenset(['episode', 'series', 'season', 'video_codec', 'resolution', 'release_group']),
+                   frozenset(['season', 'video_codec', 'episode', 'resolution', 'series'])}
+        with self.Provider() as provider:
+            subtitles = provider.query(language, series=video.series, season=video.season, episode=video.episode)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, {language})
+
+    def test_list_subtitles(self):
+        video = MOVIES[0]
+        languages = {Language('eng'), Language('fra')}
+        matches = {frozenset(['video_codec', 'title', 'resolution', 'year']),
+                   frozenset(['title', 'resolution', 'year']),
+                   frozenset(['video_codec', 'title', 'year']),
+                   frozenset(['title', 'year']),
+                   frozenset(['video_codec', 'title', 'resolution', 'release_group', 'year']),
+                   frozenset(['video_codec', 'title', 'resolution', 'audio_codec', 'year'])}
+        with self.Provider() as provider:
+            subtitles = provider.list_subtitles(video, languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_download_subtitle(self):
         video = MOVIES[0]
@@ -287,8 +340,8 @@ class TheSubDBProviderTestCase(ProviderTestCase):
         matches = {frozenset(['hash'])}
         with self.Provider() as provider:
             subtitles = provider.query(video.hashes['thesubdb'])
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_episode_1(self):
         video = EPISODES[1]
@@ -296,8 +349,8 @@ class TheSubDBProviderTestCase(ProviderTestCase):
         matches = {frozenset(['hash'])}
         with self.Provider() as provider:
             subtitles = provider.query(video.hashes['thesubdb'])
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_list_subtitles(self):
         video = MOVIES[0]
@@ -305,8 +358,8 @@ class TheSubDBProviderTestCase(ProviderTestCase):
         matches = {frozenset(['hash'])}
         with self.Provider() as provider:
             subtitles = provider.list_subtitles(video, languages)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_download_subtitle(self):
         video = MOVIES[0]
@@ -323,17 +376,27 @@ class TVsubtitlesProviderTestCase(ProviderTestCase):
     def test_find_show_id(self):
         with self.Provider() as provider:
             show_id = provider.find_show_id('The Big Bang')
-        self.assertTrue(show_id == 154)
+        self.assertEqual(show_id, 154)
+
+    def test_find_show_id_ambiguous(self):
+        with self.Provider() as provider:
+            show_id = provider.find_show_id('New Girl')
+        self.assertEqual(show_id, 977)
+
+    def test_find_show_id_no_dots(self):
+        with self.Provider() as provider:
+            show_id = provider.find_show_id('Marvel\'s Agents of S H I E L D')
+        self.assertEqual(show_id, 1340)
 
     def test_find_show_id_error(self):
         with self.Provider() as provider:
             show_id = provider.find_show_id('the big gaming')
-        self.assertTrue(show_id is None)
+        self.assertIsNone(show_id)
 
     def test_find_episode_ids(self):
         with self.Provider() as provider:
             episode_ids = provider.find_episode_ids(154, 5)
-        self.assertTrue(set(episode_ids.keys()) == set(range(1, 25)))
+        self.assertEqual(set(episode_ids.keys()), set(range(1, 25)))
 
     def test_query_episode_0(self):
         video = EPISODES[0]
@@ -342,8 +405,8 @@ class TVsubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['series', 'episode', 'season'])}
         with self.Provider() as provider:
             subtitles = provider.query(video.series, video.season, video.episode)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_query_episode_1(self):
         video = EPISODES[1]
@@ -354,8 +417,8 @@ class TVsubtitlesProviderTestCase(ProviderTestCase):
                    frozenset(['series', 'episode', 'season'])}
         with self.Provider() as provider:
             subtitles = provider.query(video.series, video.season, video.episode)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_list_subtitles(self):
         video = EPISODES[0]
@@ -363,8 +426,8 @@ class TVsubtitlesProviderTestCase(ProviderTestCase):
         matches = {frozenset(['series', 'episode', 'season'])}
         with self.Provider() as provider:
             subtitles = provider.list_subtitles(video, languages)
-        self.assertTrue({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles} == matches)
-        self.assertTrue({subtitle.language for subtitle in subtitles} == languages)
+        self.assertEqual({frozenset(subtitle.compute_matches(video)) for subtitle in subtitles}, matches)
+        self.assertEqual({subtitle.language for subtitle in subtitles}, languages)
 
     def test_download_subtitle(self):
         video = EPISODES[0]
@@ -380,6 +443,7 @@ def suite():
     suite.addTest(TestLoader().loadTestsFromTestCase(Addic7edProviderTestCase))
     suite.addTest(TestLoader().loadTestsFromTestCase(BierDopjeProviderTestCase))
     suite.addTest(TestLoader().loadTestsFromTestCase(OpenSubtitlesProviderTestCase))
+    suite.addTest(TestLoader().loadTestsFromTestCase(PodnapisiProviderTestCase))
     suite.addTest(TestLoader().loadTestsFromTestCase(TheSubDBProviderTestCase))
     suite.addTest(TestLoader().loadTestsFromTestCase(TVsubtitlesProviderTestCase))
     return suite

subliminal/tests/test_subliminal.py

@@ -5,7 +5,7 @@ import os
 import shutil
 from unittest import TestCase, TestSuite, TestLoader, TextTestRunner
 from babelfish import Language
-from subliminal import list_subtitles, download_subtitles, download_best_subtitles, scan_video, scan_videos
+from subliminal import list_subtitles, download_subtitles, download_best_subtitles, scan_video
 from subliminal.tests.common import MOVIES, EPISODES
 
 
@@ -23,22 +23,22 @@ class ApiTestCase(TestCase):
         videos = [MOVIES[0]]
         languages = {Language('eng')}
         subtitles = list_subtitles(videos, languages)
-        self.assertTrue(len(subtitles) == len(videos))
-        self.assertTrue(len(subtitles[videos[0]]) > 0)
+        self.assertEqual(len(subtitles), len(videos))
+        self.assertGreater(len(subtitles[videos[0]]), 0)
 
     def test_list_subtitles_movie_0_por_br(self):
         videos = [MOVIES[0]]
         languages = {Language('por', 'BR')}
         subtitles = list_subtitles(videos, languages)
-        self.assertTrue(len(subtitles) == len(videos))
-        self.assertTrue(len(subtitles[videos[0]]) > 0)
+        self.assertEqual(len(subtitles), len(videos))
+        self.assertGreater(len(subtitles[videos[0]]), 0)
 
     def test_list_subtitles_episodes(self):
         videos = [EPISODES[0], EPISODES[1]]
         languages = {Language('eng'), Language('fra')}
         subtitles = list_subtitles(videos, languages)
-        self.assertTrue(len(subtitles) == len(videos))
-        self.assertTrue(len(subtitles[videos[0]]) > 0)
+        self.assertEqual(len(subtitles), len(videos))
+        self.assertGreater(len(subtitles[videos[0]]), 0)
 
     def test_download_subtitles(self):
         videos = [EPISODES[0], EPISODES[1]]
@@ -68,7 +68,7 @@ class ApiTestCase(TestCase):
         languages = {Language('eng'), Language('fra')}
         subtitles = download_best_subtitles(videos, languages)
         for video in videos:
-            self.assertTrue(video in subtitles and len(subtitles[video]) == 2)
+            self.assertEqual(video in subtitles and len(subtitles[video]), 2)
             self.assertTrue(os.path.exists(os.path.splitext(video.name)[0] + '.en.srt'))
             self.assertTrue(os.path.exists(os.path.splitext(video.name)[0] + '.fr.srt'))
 
@@ -79,7 +79,8 @@ class ApiTestCase(TestCase):
         languages = {Language('eng'), Language('fra')}
         subtitles = download_best_subtitles(videos, languages, single=True)
         for video in videos:
-            self.assertTrue(video in subtitles and len(subtitles[video]) == 1)
+            self.assertIn(video, subtitles)
+            self.assertEqual(len(subtitles[video]), 1)
             self.assertTrue(os.path.exists(os.path.splitext(video.name)[0] + '.srt'))
 
     def test_download_best_subtitles_min_score(self):
@@ -88,7 +89,7 @@ class ApiTestCase(TestCase):
             video.name = os.path.join(TEST_DIR, video.name.split(os.sep)[-1])
         languages = {Language('eng'), Language('fra')}
         subtitles = download_best_subtitles(videos, languages, min_score=1000)
-        self.assertTrue(len(subtitles) == 0)
+        self.assertEqual(len(subtitles), 0)
 
     def test_download_best_subtitles_hearing_impaired(self):
         videos = [MOVIES[0]]
@@ -96,7 +97,7 @@ class ApiTestCase(TestCase):
             video.name = os.path.join(TEST_DIR, video.name.split(os.sep)[-1])
         languages = {Language('eng')}
         subtitles = download_best_subtitles(videos, languages, hearing_impaired=True)
-        self.assertTrue(subtitles[videos[0]][0].hearing_impaired == True)
+        self.assertTrue(subtitles[videos[0]][0].hearing_impaired)
 
 
 class VideoTestCase(TestCase):
@@ -111,47 +112,47 @@ class VideoTestCase(TestCase):
     def test_scan_video_movie(self):
         video = MOVIES[0]
         scanned_video = scan_video(os.path.join(TEST_DIR, os.path.split(video.name)[1]))
-        self.assertTrue(scanned_video.name == os.path.join(TEST_DIR, os.path.split(video.name)[1]))
-        self.assertTrue(scanned_video.title.lower() == video.title.lower())
-        self.assertTrue(scanned_video.year == video.year)
-        self.assertTrue(scanned_video.video_codec == video.video_codec)
-        self.assertTrue(scanned_video.resolution == video.resolution)
-        self.assertTrue(scanned_video.release_group == video.release_group)
-        self.assertTrue(scanned_video.subtitle_languages == set())
-        self.assertTrue(scanned_video.hashes == {})
-        self.assertTrue(scanned_video.audio_codec is None)
-        self.assertTrue(scanned_video.imdb_id is None)
-        self.assertTrue(scanned_video.size == 0)
+        self.assertEqual(scanned_video.name, os.path.join(TEST_DIR, os.path.split(video.name)[1]))
+        self.assertEqual(scanned_video.title.lower(), video.title.lower())
+        self.assertEqual(scanned_video.year, video.year)
+        self.assertEqual(scanned_video.video_codec, video.video_codec)
+        self.assertEqual(scanned_video.resolution, video.resolution)
+        self.assertEqual(scanned_video.release_group, video.release_group)
+        self.assertEqual(scanned_video.subtitle_languages, set())
+        self.assertEqual(scanned_video.hashes, {})
+        self.assertIsNone(scanned_video.audio_codec)
+        self.assertIsNone(scanned_video.imdb_id)
+        self.assertEqual(scanned_video.size, 0)
 
     def test_scan_video_episode(self):
         video = EPISODES[0]
         scanned_video = scan_video(os.path.join(TEST_DIR, os.path.split(video.name)[1]))
-        self.assertTrue(scanned_video.name == os.path.join(TEST_DIR, os.path.split(video.name)[1]))
-        self.assertTrue(scanned_video.series == video.series)
-        self.assertTrue(scanned_video.season == video.season)
-        self.assertTrue(scanned_video.episode == video.episode)
-        self.assertTrue(scanned_video.video_codec == video.video_codec)
-        self.assertTrue(scanned_video.resolution == video.resolution)
-        self.assertTrue(scanned_video.release_group == video.release_group)
-        self.assertTrue(scanned_video.subtitle_languages == set())
-        self.assertTrue(scanned_video.hashes == {})
-        self.assertTrue(scanned_video.title is None)
-        self.assertTrue(scanned_video.tvdb_id is None)
-        self.assertTrue(scanned_video.imdb_id is None)
-        self.assertTrue(scanned_video.audio_codec is None)
-        self.assertTrue(scanned_video.size == 0)
+        self.assertEqual(scanned_video.name, os.path.join(TEST_DIR, os.path.split(video.name)[1]))
+        self.assertEqual(scanned_video.series, video.series)
+        self.assertEqual(scanned_video.season, video.season)
+        self.assertEqual(scanned_video.episode, video.episode)
+        self.assertEqual(scanned_video.video_codec, video.video_codec)
+        self.assertEqual(scanned_video.resolution, video.resolution)
+        self.assertEqual(scanned_video.release_group, video.release_group)
+        self.assertEqual(scanned_video.subtitle_languages, set())
+        self.assertEqual(scanned_video.hashes, {})
+        self.assertIsNone(scanned_video.title)
+        self.assertIsNone(scanned_video.tvdb_id)
+        self.assertIsNone(scanned_video.imdb_id)
+        self.assertIsNone(scanned_video.audio_codec)
+        self.assertEqual(scanned_video.size, 0)
 
     def test_scan_video_subtitle_language_und(self):
         video = EPISODES[0]
         open(os.path.join(TEST_DIR, os.path.splitext(os.path.split(video.name)[1])[0]) + '.srt', 'w').close()
         scanned_video = scan_video(os.path.join(TEST_DIR, os.path.split(video.name)[1]))
-        self.assertTrue(scanned_video.subtitle_languages == {Language('und')})
+        self.assertEqual(scanned_video.subtitle_languages, {Language('und')})
 
     def test_scan_video_subtitles_language_eng(self):
         video = EPISODES[0]
         open(os.path.join(TEST_DIR, os.path.splitext(os.path.split(video.name)[1])[0]) + '.en.srt', 'w').close()
         scanned_video = scan_video(os.path.join(TEST_DIR, os.path.split(video.name)[1]))
-        self.assertTrue(scanned_video.subtitle_languages == {Language('eng')})
+        self.assertEqual(scanned_video.subtitle_languages, {Language('eng')})
 
     def test_scan_video_subtitles_languages(self):
         video = EPISODES[0]
@@ -159,7 +160,8 @@ class VideoTestCase(TestCase):
         open(os.path.join(TEST_DIR, os.path.splitext(os.path.split(video.name)[1])[0]) + '.fr.srt', 'w').close()
         open(os.path.join(TEST_DIR, os.path.splitext(os.path.split(video.name)[1])[0]) + '.srt', 'w').close()
         scanned_video = scan_video(os.path.join(TEST_DIR, os.path.split(video.name)[1]))
-        self.assertTrue(scanned_video.subtitle_languages == {Language('eng'), Language('fra'), Language('und')})
+        self.assertEqual(scanned_video.subtitle_languages, {Language('eng'), Language('fra'), Language('und')})
+
 
 def suite():
     suite = TestSuite()

subliminal/video.py

@@ -156,7 +156,7 @@ def scan_subtitle_languages(path):
     :rtype: set
 
     """
-    language_extensions = tuple('.' + c for c in babelfish.CONVERTERS['alpha2'].codes)
+    language_extensions = tuple('.' + c for c in babelfish.get_language_converter('alpha2').codes)
     dirpath, filename = os.path.split(path)
     subtitles = set()
     for p in os.listdir(dirpath):
@@ -239,14 +239,24 @@ def scan_video(path, subtitles=True, embedded_subtitles=True):
                 if embedded_subtitles:
                     embedded_subtitle_languages = set()
                     for st in mkv.subtitle_tracks:
-                        try:
-                            embedded_subtitle_languages.add(babelfish.Language.fromalpha3b(st.language or 'und'))
-                        except babelfish.Error:
-                            logger.error('Embedded subtitle language %r is not a valid language', st.language)
+                        if st.language:
+                            try:
+                                embedded_subtitle_languages.add(babelfish.Language.fromalpha3b(st.language))
+                            except babelfish.Error:
+                                logger.error('Embedded subtitle track language %r is not a valid language', st.language)
+                                embedded_subtitle_languages.add(babelfish.Language('und'))
+                        elif st.name:
+                            try:
+                                embedded_subtitle_languages.add(babelfish.Language.fromname(st.name))
+                            except babelfish.Error:
+                                logger.error('Embedded subtitle track name %r is not a valid language', st.name)
+                                embedded_subtitle_languages.add(babelfish.Language('und'))
+                        else:
+                            embedded_subtitle_languages.add(babelfish.Language('und'))
                     logger.debug('Found embedded subtitle %r with enzyme', embedded_subtitle_languages)
                     video.subtitle_languages |= embedded_subtitle_languages
             else:
-                logger.info('MKV has no subtitle track')
+                logger.debug('MKV has no subtitle track')
     except enzyme.Error:
         logger.error('Parsing video metadata with enzyme failed')
     return video