Merge remote-tracking branch 'origin/master'

CHANGELOG.md

@@ -1,3 +1,38 @@
+2.6.5.3109
+
+subscene, addic7ed and titlovi
+- either of those providers might impose a reCAPTCHA verification. In order to use those providers, please create an account at an AntiCaptcha service ([anti-captcha.com](http://getcaptchasolution.com/kkvviom7nh) or [deathbycaptcha.com](http://deathbycaptcha.com)), add funds, then supply your credentials/apikey in the configuration
+
+Changelog
+- providers: add Napisy24 (polish)
+- providers: subscene: reduce provider load by possibly half
+- providers: subscene: support logging in (username/password are now required)
+- providers: subscene: fallback to non year results if none found with year
+
+
+2.6.5.3099
+
+subscene, addic7ed and titlovi
+- either of those providers might impose a reCAPTCHA verification. In order to use those providers, please create an account at an AntiCaptcha service ([anti-captcha.com](http://getcaptchasolution.com/kkvviom7nh) or [deathbycaptcha.com](http://deathbycaptcha.com)), add funds, then supply your credentials/apikey in the configuration
+
+Changelog
+- core: allow system DNS again by putting "system" as the DNS
+- providers: subscene: fix again (subscene, contact us please, so we can end this)
+
+
+2.6.5.3092
+
+subscene, addic7ed and titlovi
+- either of those providers might impose a reCAPTCHA verification. In order to use those providers, please create an account at an AntiCaptcha service ([anti-captcha.com](http://getcaptchasolution.com/kkvviom7nh) or [deathbycaptcha.com](http://deathbycaptcha.com)), add funds, then supply your credentials/apikey in the configuration
+
+Changelog
+- providers: subscene: fix endpoint (hopefully for longer now)
+- providers: subscene: don't search for season packs (broken for now; relieves 50% of server load on provider)
+- providers: subscene: don't calculate video fn for now
+- providers: argenteam: backport fixes from bazarr
+- subtitle: try decoding with utf-16 by default as well (zho/farsi)
+- submod: HI: remove music tags by default
+- core: compat (bazarr): add env var SZ_KEEP_ENCODING to keep encoding of subtitles
 
 
 2.6.5.3074

Contents/Code/__init__.py

@@ -119,19 +119,23 @@ def agent_extract_embedded(video_part_map):
             for plexapi_part in get_all_parts(plexapi_item):
                 item_count = item_count + 1
                 used_one_unknown_stream = False
+                used_one_known_stream = False
                 for requested_language in config.lang_list:
+                    skip_unknown = used_one_unknown_stream or used_one_known_stream
                     embedded_subs = stored_subs.get_by_provider(plexapi_part.id, requested_language, "embedded")
                     current = stored_subs.get_any(plexapi_part.id, requested_language) or \
                         requested_language in scanned_video.external_subtitle_languages
 
                     if not embedded_subs:
                         stream_data = get_embedded_subtitle_streams(plexapi_part, requested_language=requested_language,
-                                                                    skip_unknown=used_one_unknown_stream)
+                                                                    skip_unknown=skip_unknown)
 
                         if stream_data:
                             stream = stream_data[0]["stream"]
                             if stream_data[0]["is_unknown"]:
                                 used_one_unknown_stream = True
+                            else:
+                                used_one_known_stream = True
 
                             to_extract.append(({scanned_video: part_info}, plexapi_part, str(stream.index),
                                                str(requested_language), not current))

Contents/Code/support/config.py

@@ -761,6 +761,7 @@ class Config(object):
         return {'opensubtitles': cast_bool(Prefs['provider.opensubtitles.enabled']),
                      # 'thesubdb': Prefs['provider.thesubdb.enabled'],
                      'podnapisi': cast_bool(Prefs['provider.podnapisi.enabled']),
+                     'napisy24': cast_bool(Prefs['provider.napisy24.enabled']),
                      'titlovi': cast_bool(Prefs['provider.titlovi.enabled']),
                      'addic7ed': cast_bool(Prefs['provider.addic7ed.enabled']) and self.has_anticaptcha,
                      'tvsubtitles': cast_bool(Prefs['provider.tvsubtitles.enabled']),
@@ -801,6 +802,7 @@ class Config(object):
             providers["argenteam"] = False
             providers["assrt"] = False
             providers["subscene"] = False
+            providers["napisy24"] = False
             providers_forced_off = dict(providers)
 
         if not self.unrar and providers["legendastv"]:
@@ -864,8 +866,14 @@ class Config(object):
                                  'only_foreign': self.forced_only,
                                  'also_foreign': self.forced_also,
                              },
+                             'napisy24': {
+                                 'username': Prefs['provider.napisy24.username'],
+                                 'password': Prefs['provider.napisy24.password'],
+                             },
                              'subscene': {
                                  'only_foreign': self.forced_only,
+                                 'username': Prefs['provider.subscene.username'],
+                                 'password': Prefs['provider.subscene.password'],
                              },
                              'legendastv': {'username': Prefs['provider.legendastv.username'],
                                             'password': Prefs['provider.legendastv.password'],

Contents/Code/support/helpers.py

@@ -394,7 +394,7 @@ def get_language_from_stream(lang_code):
             return Language.fromietf(lang)
         elif lang:
             try:
-                return language_from_stream(lang)
+                return language_from_stream(lang_code)
             except LanguageError:
                 pass
 

Contents/Code/support/plex_media.py

@@ -177,6 +177,7 @@ def get_all_parts(plex_item):
 def get_embedded_subtitle_streams(part, requested_language=None, skip_duplicate_unknown=True, skip_unknown=False):
     streams = []
     streams_unknown = []
+    all_streams = []
     has_unknown = False
     found_requested_language = False
     for stream in part.streams:
@@ -189,27 +190,40 @@ def get_embedded_subtitle_streams(part, requested_language=None, skip_duplicate_
 
             is_unknown = False
             found_requested_language = requested_language and requested_language == language
+            stream_data = None
 
-            if not language and config.treat_und_as_first:
+            if not language:
                 # only consider first unknown subtitle stream
-                if has_unknown and skip_duplicate_unknown:
-                    continue
+                if requested_language and config.treat_und_as_first:
+                    if has_unknown and skip_duplicate_unknown:
+                        Log.Debug("skipping duplicate unknown")
+                        continue
 
-                language = Language.rebuild(list(config.lang_list)[0], forced=is_forced)
+                    language = Language.rebuild(list(config.lang_list)[0], forced=is_forced)
+                else:
+                    language = Language("unk")
                 is_unknown = True
                 has_unknown = True
-                streams_unknown.append({"stream": stream, "is_unknown": is_unknown, "language": language,
-                                        "is_forced": is_forced})
+                stream_data = {"stream": stream, "is_unknown": is_unknown, "language": language,
+                               "is_forced": is_forced}
+                streams_unknown.append(stream_data)
 
             if not requested_language or found_requested_language:
-                streams.append({"stream": stream, "is_unknown": is_unknown, "language": language,
-                                "is_forced": is_forced})
+                stream_data = {"stream": stream, "is_unknown": is_unknown, "language": language,
+                               "is_forced": is_forced}
+                streams.append(stream_data)
 
                 if found_requested_language:
                     break
 
-    if streams_unknown and not found_requested_language and not skip_unknown:
-        streams = streams_unknown
+            if stream_data:
+                all_streams.append(stream_data)
+
+    if requested_language:
+        if streams_unknown and not found_requested_language and not skip_unknown:
+            streams = streams_unknown
+    else:
+        streams = all_streams
 
     return streams
 

Contents/DefaultPrefs.json

@@ -335,6 +335,26 @@
     "type": "bool",
     "default": "true"
   },
+  {
+    "id": "provider.napisy24.enabled",
+    "label": "Provider: Enable Napisy24 (pl)",
+    "type": "bool",
+    "default": "false"
+  },
+  {
+    "id": "provider.napisy24.username",
+    "label": "Napisy24 Username",
+    "type": "text",
+    "default": ""
+  },
+  {
+    "id": "provider.napisy24.password",
+    "label": "Napisy24 Password",
+    "type": "text",
+    "option": "hidden",
+    "default": "",
+    "secure": "true"
+  },
   {
     "id": "provider.addic7ed.enabled",
     "label": "Provider: Enable Addic7ed (needs AntiCaptcha)",
@@ -431,6 +451,20 @@
     "type": "bool",
     "default": "true"
   },
+  {
+    "id": "provider.subscene.username",
+    "label": "SubScene Username",
+    "type": "text",
+    "default": ""
+  },
+  {
+    "id": "provider.subscene.password",
+    "label": "SubScene Password",
+    "type": "text",
+    "option": "hidden",
+    "default": "",
+    "secure": "true"
+  },
   {
     "id": "provider.supersubtitles.enabled",
     "label": "Provider: Enable feliratok.info (Hungarian)",

Contents/Info.plist

@@ -13,7 +13,7 @@
         <key>CFBundleSignature</key>
         <string>????</string>
         <key>CFBundleVersion</key>
-        <string>2.6.5.3099</string>
+        <string>2.6.5.3124</string>
         <key>PlexFrameworkVersion</key>
         <string>2</string>
         <key>PlexPluginClass</key>
@@ -32,7 +32,7 @@
 
 &lt;h1&gt;Sub-Zero for Plex&lt;/h1&gt;&lt;i&gt;Subtitles done right&lt;/i&gt;
 
-Version 2.6.5.3099
+Version 2.6.5.3124
 
 Originally based on @bramwalet's awesome &lt;a href=&quot;https://github.com/bramwalet/Subliminal.bundle&quot;&gt;Subliminal.bundle&lt;/a&gt;
 

Contents/Libraries/Shared/subliminal_patch/converters/subscene.py

@@ -12,6 +12,13 @@ from_subscene = {
         'Malay': 'msa', 'Pashto': 'pus', 'Punjabi': 'pan', 'Swahili': 'swa'
 }
 
+from_subscene_with_country = {
+    'Brazillian Portuguese': ('por', 'BR')
+}
+
+to_subscene_with_country = {val: key for key, val in from_subscene_with_country.items()}
+
+
 to_subscene = {v: k for k, v in from_subscene.items()}
 
 exact_languages_alpha3 = [
@@ -34,12 +41,12 @@ language_ids = {
         'mkd': 48, 'mal': 64, 'mni': 65, 'mon': 72, 'pus': 67, 'pol': 31,
         'por': 32, 'pan': 66, 'rus': 34, 'srp': 35, 'sin': 58, 'slk': 36,
         'slv': 37, 'som': 70, 'tgl': 53, 'tam': 59, 'tel': 63, 'tha': 40,
-        'tur': 41, 'ukr': 56, 'urd': 42, 'yor': 71
+        'tur': 41, 'ukr': 56, 'urd': 42, 'yor': 71, 'pt-BR': 4
 }
 
 # TODO: specify codes for unspecified_languages
 unspecified_languages = [
-        'Big 5 code', 'Brazillian Portuguese', 'Bulgarian/ English',
+        'Big 5 code', 'Bulgarian/ English',
         'Chinese BG code', 'Dutch/ English', 'English/ German',
         'Hungarian/ English', 'Rohingya'
 ]
@@ -50,6 +57,8 @@ alpha3_of_code = {l.name: l.alpha3 for l in supported_languages}
 
 supported_languages.update({Language(l) for l in to_subscene})
 
+supported_languages.update({Language(lang, cr) for lang, cr in to_subscene_with_country})
+
 
 class SubsceneConverter(LanguageReverseConverter):
     codes = {l.name for l in supported_languages}
@@ -61,9 +70,15 @@ class SubsceneConverter(LanguageReverseConverter):
         if alpha3 in to_subscene:
             return to_subscene[alpha3]
 
+        if (alpha3, country) in to_subscene_with_country:
+            return to_subscene_with_country[(alpha3, country)]
+
         raise ConfigurationError('Unsupported language for subscene: %s, %s, %s' % (alpha3, country, script))
 
     def reverse(self, code):
+        if code in from_subscene_with_country:
+            return from_subscene_with_country[code]
+
         if code in from_subscene:
             return (from_subscene[code],)
 

Contents/Libraries/Shared/subliminal_patch/core.py

@@ -473,7 +473,7 @@ if is_windows_special_path:
     SZAsyncProviderPool = SZProviderPool
 
 
-def scan_video(path, dont_use_actual_file=False, hints=None, providers=None, skip_hashing=False):
+def scan_video(path, dont_use_actual_file=False, hints=None, providers=None, skip_hashing=False, hash_from=None):
     """Scan a video from a `path`.
 
     patch:
@@ -538,28 +538,34 @@ def scan_video(path, dont_use_actual_file=False, hints=None, providers=None, ski
             video.alternative_titles.append(alt_guess["title"])
         logger.debug("Adding alternative title: %s", alt_guess["title"])
 
-    if dont_use_actual_file:
+    if dont_use_actual_file and not hash_from:
         return video
 
     # size and hashes
     if not skip_hashing:
-        video.size = os.path.getsize(path)
+        hash_path = hash_from or path
+        video.size = os.path.getsize(hash_path)
         if video.size > 10485760:
             logger.debug('Size is %d', video.size)
+            osub_hash = None
             if "opensubtitles" in providers:
-                video.hashes['opensubtitles'] = hash_opensubtitles(path)
+                video.hashes['opensubtitles'] = osub_hash = hash_opensubtitles(hash_path)
 
             if "shooter" in providers:
-                video.hashes['shooter'] = hash_shooter(path)
+                video.hashes['shooter'] = hash_shooter(hash_path)
 
             if "thesubdb" in providers:
-                video.hashes['thesubdb'] = hash_thesubdb(path)
+                video.hashes['thesubdb'] = hash_thesubdb(hash_path)
 
             if "napiprojekt" in providers:
                 try:
-                    video.hashes['napiprojekt'] = hash_napiprojekt(path)
+                    video.hashes['napiprojekt'] = hash_napiprojekt(hash_path)
                 except MemoryError:
-                    logger.warning(u"Couldn't compute napiprojekt hash for %s", path)
+                    logger.warning(u"Couldn't compute napiprojekt hash for %s", hash_path)
+
+            if "napisy24" in providers:
+                # Napisy24 uses the same hash as opensubtitles
+                video.hashes['napisy24'] = osub_hash or hash_opensubtitles(hash_path)
 
             logger.debug('Computed hashes %r', video.hashes)
         else:
@@ -575,7 +581,7 @@ def _search_external_subtitles(path, languages=None, only_one=False, scandir_gen
     subtitles = {}
     _scandir = _scandir_generic if scandir_generic else scandir
     for entry in _scandir(dirpath):
-        if not entry.name and not scandir_generic:
+        if (not entry.name or entry.name in ('\x0c', '$', ',', '\x7f')) and not scandir_generic:
             logger.debug('Could not determine the name of the file, retrying with scandir_generic')
             return _search_external_subtitles(path, languages, only_one, True)
         if not entry.is_file(follow_symlinks=False):

Contents/Libraries/Shared/subliminal_patch/http.py

@@ -354,7 +354,6 @@ def patch_create_connection():
                         return _orig_create_connection((ip, port), *args, **kwargs)
                     except dns.exception.DNSException:
                         logger.warning("DNS: Couldn't resolve %s with DNS: %s", host, custom_resolver.nameservers)
-                        raise
 
         logger.debug("DNS: Falling back to default DNS or IP on %s", host)
         return _orig_create_connection((host, port), *args, **kwargs)

Contents/Libraries/Shared/subliminal_patch/providers/napisy24.py

@@ -0,0 +1,124 @@
+import logging
+import os
+from io import BytesIO
+from zipfile import ZipFile
+
+from requests import Session
+
+from subliminal_patch.subtitle import Subtitle
+from subliminal_patch.providers import Provider
+from subliminal import __short_version__
+from subliminal.exceptions import AuthenticationError, ConfigurationError
+from subliminal.subtitle import fix_line_ending
+from subzero.language import Language
+
+logger = logging.getLogger(__name__)
+
+
+class Napisy24Subtitle(Subtitle):
+    '''Napisy24 Subtitle.'''
+    provider_name = 'napisy24'
+
+    def __init__(self, language, hash, imdb_id, napis_id):
+        super(Napisy24Subtitle, self).__init__(language)
+        self.hash = hash
+        self.imdb_id = imdb_id
+        self.napis_id = napis_id
+
+    @property
+    def id(self):
+        return self.hash
+
+    def get_matches(self, video):
+        matches = set()
+
+        # hash
+        if 'napisy24' in video.hashes and video.hashes['napisy24'] == self.hash:
+            matches.add('hash')
+
+        # imdb_id
+        if video.imdb_id and self.imdb_id == video.imdb_id:
+            matches.add('imdb_id')
+
+        return matches
+
+
+class Napisy24Provider(Provider):
+    '''Napisy24 Provider.'''
+    languages = {Language(l) for l in ['pol']}
+    required_hash = 'napisy24'
+    api_url = 'http://napisy24.pl/run/CheckSubAgent.php'
+
+    def __init__(self, username=None, password=None):
+        if all((username, password)):
+            self.username = username
+            self.password = password
+        else:
+            self.username = 'subliminal'
+            self.password = 'lanimilbus'
+
+        self.session = None
+
+    def initialize(self):
+        self.session = Session()
+        self.session.headers['User-Agent'] = 'Subliminal/%s' % __short_version__
+        self.session.headers['Content-Type'] = 'application/x-www-form-urlencoded'
+
+    def terminate(self):
+        self.session.close()
+
+    def query(self, language, size, name, hash):
+        params = {
+            'postAction': 'CheckSub',
+            'ua': self.username,
+            'ap': self.password,
+            'fs': size,
+            'fh': hash,
+            'fn': os.path.basename(name),
+            'n24pref': 1
+        }
+
+        response = self.session.post(self.api_url, data=params, timeout=10)
+        response.raise_for_status()
+
+        response_content = response.content.split(b'||', 1)
+        n24_data = response_content[0].decode()
+
+        if n24_data[:2] != 'OK':
+            if n24_data[:11] == 'login error':
+                raise AuthenticationError('Login failed')
+            logger.error('Unknown response: %s', response.content)
+            return None
+
+        n24_status = n24_data[:4]
+        if n24_status == 'OK-0':
+            logger.info('No subtitles found')
+            return None
+
+        subtitle_info = dict(p.split(':', 1) for p in n24_data.split('|')[1:])
+        logger.debug('Subtitle info: %s', subtitle_info)
+
+        if n24_status == 'OK-1':
+            logger.info('No subtitles found but got video info')
+            return None
+        elif n24_status == 'OK-2':
+            logger.info('Found subtitles')
+        elif n24_status == 'OK-3':
+            logger.info('Found subtitles but not from Napisy24 database')
+            return None
+        
+        subtitle_content = response_content[1]
+
+        subtitle = Napisy24Subtitle(language, hash, 'tt%s' % subtitle_info['imdb'].zfill(7), subtitle_info['napisId'])
+        with ZipFile(BytesIO(subtitle_content)) as zf:
+            subtitle.content = fix_line_ending(zf.open(zf.namelist()[0]).read())
+
+        return subtitle
+
+    def list_subtitles(self, video, languages):
+        subtitles = [self.query(l, video.size, video.name, video.hashes['napisy24']) for l in languages]
+        return [s for s in subtitles if s is not None]
+
+    def download_subtitle(self, subtitle):
+        # there is no download step, content is already filled from listing subtitles
+        pass

Contents/Libraries/Shared/subliminal_patch/providers/subscene.py

@@ -4,24 +4,34 @@ import io
 import logging
 import os
 import time
+import traceback
+
+import requests
+
 import inflect
+import re
+import json
+import HTMLParser
+import urlparse
 
 from zipfile import ZipFile
 from babelfish import language_converters
 from guessit import guessit
+from dogpile.cache.api import NO_VALUE
 from subliminal import Episode, ProviderError
+from subliminal.exceptions import ConfigurationError, ServiceUnavailable
 from subliminal.utils import sanitize_release_group
+from subliminal.cache import region
 from subliminal_patch.http import RetryingCFSession
 from subliminal_patch.providers import Provider
 from subliminal_patch.providers.mixins import ProviderSubtitleArchiveMixin
 from subliminal_patch.subtitle import Subtitle, guess_matches
 from subliminal_patch.converters.subscene import language_ids, supported_languages
-from subscene_api.subscene import search, Subtitle as APISubtitle
+from subscene_api.subscene import search, Subtitle as APISubtitle, SITE_DOMAIN
 from subzero.language import Language
 
 p = inflect.engine()
 
-
 language_converters.register('subscene = subliminal_patch.converters.subscene:SubsceneConverter')
 logger = logging.getLogger(__name__)
 
@@ -112,28 +122,106 @@ class SubsceneProvider(Provider, ProviderSubtitleArchiveMixin):
     skip_wrong_fps = False
     hearing_impaired_verifiable = True
     only_foreign = False
+    username = None
+    password = None
 
-    search_throttle = 2  # seconds
+    search_throttle = 8  # seconds
+
+    def __init__(self, only_foreign=False, username=None, password=None):
+        if not all((username, password)):
+            raise ConfigurationError('Username and password must be specified')
 
-    def __init__(self, only_foreign=False):
         self.only_foreign = only_foreign
+        self.username = username
+        self.password = password
 
     def initialize(self):
         logger.info("Creating session")
         self.session = RetryingCFSession()
 
+        prev_cookies = region.get("subscene_cookies2")
+        if prev_cookies != NO_VALUE:
+            logger.debug("Re-using old subscene cookies: %r", prev_cookies)
+            self.session.cookies.update(prev_cookies)
+
+        else:
+            logger.debug("Logging in")
+            self.login()
+
+    def login(self):
+        r = self.session.get("https://subscene.com/account/login")
+        if "Server Error" in r.content:
+            logger.error("Login unavailable; Maintenance?")
+            raise ServiceUnavailable("Login unavailable; Maintenance?")
+
+        match = re.search(r"<script id='modelJson' type='application/json'>\s*(.+)\s*</script>", r.content)
+
+        if match:
+            h = HTMLParser.HTMLParser()
+            data = json.loads(h.unescape(match.group(1)))
+            login_url = urlparse.urljoin(data["siteUrl"], data["loginUrl"])
+            time.sleep(1.0)
+
+            r = self.session.post(login_url,
+                                  {
+                                      "username": self.username,
+                                      "password": self.password,
+                                      data["antiForgery"]["name"]: data["antiForgery"]["value"]
+                                  })
+            pep_content = re.search(r"<form method=\"post\" action=\"https://subscene\.com/\">"
+                                    r".+name=\"id_token\".+?value=\"(?P<id_token>.+?)\".*?"
+                                    r"access_token\".+?value=\"(?P<access_token>.+?)\".+?"
+                                    r"token_type.+?value=\"(?P<token_type>.+?)\".+?"
+                                    r"expires_in.+?value=\"(?P<expires_in>.+?)\".+?"
+                                    r"scope.+?value=\"(?P<scope>.+?)\".+?"
+                                    r"state.+?value=\"(?P<state>.+?)\".+?"
+                                    r"session_state.+?value=\"(?P<session_state>.+?)\"",
+                                    r.content, re.MULTILINE | re.DOTALL)
+
+            if pep_content:
+                r = self.session.post(SITE_DOMAIN, pep_content.groupdict())
+                try:
+                    r.raise_for_status()
+                except Exception:
+                    raise ProviderError("Something went wrong when trying to log in: %s", traceback.format_exc())
+                else:
+                    cj = self.session.cookies.copy()
+                    store_cks = ("scene", "idsrv", "idsrv.xsrf", "idsvr.clients", "idsvr.session", "idsvr.username")
+                    for cn in self.session.cookies.iterkeys():
+                        if cn not in store_cks:
+                            del cj[cn]
+
+                    logger.debug("Storing cookies: %r", cj)
+                    region.set("subscene_cookies2", cj)
+                    return
+        raise ProviderError("Something went wrong when trying to log in #1")
+
     def terminate(self):
         logger.info("Closing session")
         self.session.close()
 
     def _create_filters(self, languages):
         self.filters = dict(HearingImpaired="2")
+        acc_filters = self.filters.copy()
         if self.only_foreign:
             self.filters["ForeignOnly"] = "True"
+            acc_filters["ForeignOnly"] = self.filters["ForeignOnly"].lower()
             logger.info("Only searching for foreign/forced subtitles")
 
-        self.filters["LanguageFilter"] = ",".join((str(language_ids[l.alpha3]) for l in languages
-                                                   if l.alpha3 in language_ids))
+        selected_ids = []
+        for l in languages:
+            lid = language_ids.get(l.basename, language_ids.get(l.alpha3, None))
+            if lid:
+                selected_ids.append(str(lid))
+
+        acc_filters["SelectedIds"] = selected_ids
+        self.filters["LanguageFilter"] = ",".join(acc_filters["SelectedIds"])
+
+        last_filters = region.get("subscene_filters")
+        if last_filters != acc_filters:
+            region.set("subscene_filters", acc_filters)
+            logger.debug("Setting account filters to %r", acc_filters)
+            self.session.post("https://u.subscene.com/filter", acc_filters, allow_redirects=False)
 
         logger.debug("Filter created: '%s'" % self.filters)
 
@@ -176,7 +264,11 @@ class SubsceneProvider(Provider, ProviderSubtitleArchiveMixin):
     def parse_results(self, video, film):
         subtitles = []
         for s in film.subtitles:
-            subtitle = SubsceneSubtitle.from_api(s)
+            try:
+                subtitle = SubsceneSubtitle.from_api(s)
+            except NotImplementedError, e:
+                logger.info(e)
+                continue
             subtitle.asked_for_release_group = video.release_group
             if isinstance(video, Episode):
                 subtitle.asked_for_episode = video.episode
@@ -189,10 +281,16 @@ class SubsceneProvider(Provider, ProviderSubtitleArchiveMixin):
 
         return subtitles
 
+    def do_search(self, *args, **kwargs):
+        try:
+            return search(*args, **kwargs)
+        except requests.HTTPError:
+            region.delete("subscene_cookies2")
+
     def query(self, video):
-        #vfn = get_video_filename(video)
+        # vfn = get_video_filename(video)
         subtitles = []
-        #logger.debug(u"Searching for: %s", vfn)
+        # logger.debug(u"Searching for: %s", vfn)
         # film = search(vfn, session=self.session)
         #
         # if film and film.subtitles:
@@ -201,16 +299,17 @@ class SubsceneProvider(Provider, ProviderSubtitleArchiveMixin):
         # else:
         #     logger.debug('No release results found')
 
-        #time.sleep(self.search_throttle)
+        # time.sleep(self.search_throttle)
 
         # re-search for episodes without explicit release name
         if isinstance(video, Episode):
-            #term = u"%s S%02iE%02i" % (video.series, video.season, video.episode)
-            more_than_one = len([video.series] + video.alternative_series) > 1
-            for series in [video.series] + video.alternative_series:
+            titles = list(set([video.series] + video.alternative_series))[:2]
+            # term = u"%s S%02iE%02i" % (video.series, video.season, video.episode)
+            more_than_one = len(titles) > 1
+            for series in titles:
                 term = u"%s - %s Season" % (series, p.number_to_words("%sth" % video.season).capitalize())
                 logger.debug('Searching for alternative results: %s', term)
-                film = search(term, session=self.session, release=False, throttle=self.search_throttle)
+                film = self.do_search(term, session=self.session, release=False, throttle=self.search_throttle)
                 if film and film.subtitles:
                     logger.debug('Alternative results found: %s', len(film.subtitles))
                     subtitles += self.parse_results(video, film)
@@ -233,11 +332,12 @@ class SubsceneProvider(Provider, ProviderSubtitleArchiveMixin):
                 if more_than_one:
                     time.sleep(self.search_throttle)
         else:
-            more_than_one = len([video.title] + video.alternative_titles) > 1
-            for title in [video.title] + video.alternative_titles:
-                logger.debug('Searching for movie results: %s', title)
-                film = search(title, year=video.year, session=self.session, limit_to=None, release=False,
-                              throttle=self.search_throttle)
+            titles = list(set([video.title] + video.alternative_titles))[:2]
+            more_than_one = len(titles) > 1
+            for title in titles:
+                logger.debug('Searching for movie results: %r', title)
+                film = self.do_search(title, year=video.year, session=self.session, limit_to=None, release=False,
+                                      throttle=self.search_throttle)
                 if film and film.subtitles:
                     subtitles += self.parse_results(video, film)
                 if more_than_one:

Contents/Libraries/Shared/subliminal_patch/providers/titlovi.py

@@ -226,7 +226,7 @@ class TitloviProvider(Provider, ProviderSubtitleArchiveMixin):
                         page_link = self.server_url + sub.a.attrs['href']
                         # subtitle language
                         _lang = sub.select_one('.lang')
-                        match = lang_re.search(_lang.attrs.get('src', _lang.attrs.get('cfsrc', '')))
+                        match = lang_re.search(_lang.attrs.get('src', _lang.attrs.get('data-cfsrc', '')))
                         if match:
                             try:
                                 # decode language

Contents/Libraries/Shared/submod_test.py

@@ -21,7 +21,7 @@ if debug:
     logging.basicConfig(level=logging.DEBUG)
 
 #sub = Subtitle(Language.fromietf("eng:forced"), mods=["common", "remove_HI", "OCR_fixes", "fix_uppercase", "shift_offset(ms=-500)", "shift_offset(ms=500)", "shift_offset(s=2,ms=800)"])
-sub = Subtitle(Language.fromietf("eng:forced"), mods=["common", "remove_HI", "OCR_fixes", "fix_uppercase", "shift_offset(ms=0,s=1)"])
+sub = Subtitle(Language.fromietf("eng"), mods=["common", "remove_HI", "OCR_fixes", "fix_uppercase", "shift_offset(ms=0,s=1)"])
 sub.content = open(fn).read()
 sub.normalize()
 content = sub.get_modified_content(debug=True)

Contents/Libraries/Shared/subscene_api/subscene.py

@@ -125,7 +125,7 @@ class Subtitle(object):
         subtitles = []
 
         for row in rows:
-            if row.td.a is not None:
+            if row.td.a is not None and row.td.get("class", ["lazy"])[0] != "empty":
                 subtitles.append(cls.from_row(row))
 
         return subtitles
@@ -255,26 +255,39 @@ def get_first_film(soup, section, year=None, session=None):
                 url = SITE_DOMAIN + t.div.a.get("href")
                 break
         if not url:
-            return
+            # fallback to non-year results
+            logger.info("Falling back to non-year results as year wasn't found (%s)", year)
+            url = SITE_DOMAIN + tag.findNext("ul").find("li").div.a.get("href")
 
     return Film.from_url(url, session=session)
 
 
+def find_endpoint(session, content=None):
+    endpoint = region.get("subscene_endpoint2")
+    if endpoint is NO_VALUE:
+        if not content:
+            content = session.get(SITE_DOMAIN).text
+
+        m = ENDPOINT_RE.search(content)
+        if m:
+            endpoint = m.group(1).strip()
+            logger.debug("Switching main endpoint to %s", endpoint)
+            region.set("subscene_endpoint2", endpoint)
+    return endpoint
+
+
 def search(term, release=True, session=None, year=None, limit_to=SearchTypes.Exact, throttle=0):
     # note to subscene: if you actually start to randomize the endpoint, we'll have to query your server even more
 
     if release:
         endpoint = "release"
     else:
-        endpoint = region.get("subscene_endpoint2")
-        if endpoint is NO_VALUE:
-            ret = session.get(SITE_DOMAIN)
-            time.sleep(throttle)
-            m = ENDPOINT_RE.search(ret.text)
-            if m:
-                endpoint = m.group(1).strip()
-                logger.debug("Switching main endpoint to %s", endpoint)
-                region.set("subscene_endpoint2", endpoint)
+        endpoint = find_endpoint(session)
+        time.sleep(throttle)
+
+    if not endpoint:
+        logger.error("Couldn't find endpoint, exiting")
+        return
 
     soup = soup_for("%s/subtitles/%s" % (SITE_DOMAIN, endpoint), data={"query": term},
                     session=session)

Contents/Libraries/Shared/subzero/language.py

@@ -8,6 +8,25 @@ repl_map = {
     "dk": "da",
     "nld": "nl",
     "english": "en",
+    "alb": "sq",
+    "arm": "hy",
+    "baq": "eu",
+    "bur": "my",
+    "chi": "zh",
+    "cze": "cs",
+    "dut": "nl",
+    "fre": "fr",
+    "geo": "ka",
+    "ger": "de",
+    "gre": "el",
+    "ice": "is",
+    "mac": "mk",
+    "mao": "mi",
+    "may": "ms",
+    "per": "fa",
+    "rum": "ro",
+    "slo": "sk",
+    "tib": "bo",
 }
 
 

Contents/Libraries/Shared/subzero/modification/dictionaries/make_data.py

@@ -36,6 +36,7 @@ SZ_FIX_DATA = {
             u" l ": u" I ",
             u"'sjust": u"'s just",
             u"'tjust": u"'t just",
+            u"\";": u"'s",
         },
         "WholeWords": {
             u"I'11": u"I'll",

Contents/Libraries/Shared/subzero/video.py

@@ -52,10 +52,10 @@ def set_existing_languages(video, video_info, external_subtitles=False, embedded
             video.subtitle_languages.add(language)
 
 
-def parse_video(fn, hints, skip_hashing=False, dry_run=False, providers=None):
+def parse_video(fn, hints, skip_hashing=False, dry_run=False, providers=None, hash_from=None):
     logger.debug("Parsing video: %s, hints: %s", os.path.basename(fn), hints)
     return scan_video(fn, hints=hints, dont_use_actual_file=dry_run, providers=providers,
-                      skip_hashing=skip_hashing)
+                      skip_hashing=skip_hashing, hash_from=hash_from)
 
 
 def refine_video(video, no_refining=False, refiner_settings=None):

README.md

@@ -90,38 +90,29 @@ the.vbm, mmgoodnow, Vertig0ne, thliu78, tattoomees, ostman, count_confucius, ehe
 
 ## Changelog
 
-2.6.5.3099
+
+2.6.5.3124
 
 subscene, addic7ed and titlovi
 - either of those providers might impose a reCAPTCHA verification. In order to use those providers, please create an account at an AntiCaptcha service ([anti-captcha.com](http://getcaptchasolution.com/kkvviom7nh) or [deathbycaptcha.com](http://deathbycaptcha.com)), add funds, then supply your credentials/apikey in the configuration
 
 Changelog
-- core: allow system DNS again by putting "system" as the DNS
-- providers: subscene: fix again (subscene, contact us please, so we can end this)
-
-
-2.6.5.3092
-
-subscene, addic7ed and titlovi
-- either of those providers might impose a reCAPTCHA verification. In order to use those providers, please create an account at an AntiCaptcha service ([anti-captcha.com](http://getcaptchasolution.com/kkvviom7nh) or [deathbycaptcha.com](http://deathbycaptcha.com)), add funds, then supply your credentials/apikey in the configuration
-
-Changelog
-- providers: subscene: fix endpoint (hopefully for longer now)
-- providers: subscene: don't search for season packs (broken for now; relieves 50% of server load on provider)
-- providers: subscene: don't calculate video fn for now
-- providers: argenteam: backport fixes from bazarr
-- subtitle: try decoding with utf-16 by default as well (zho/farsi)
-- submod: HI: remove music tags by default
-- core: compat (bazarr): add env var SZ_KEEP_ENCODING to keep encoding of subtitles
-
-
-
+- core: http: fallback to default DNS when normal resolving fails; fixes #657
+- core: extract embedded/menu: fix detection of unknown streams; don't use unknown streams if a known language was previously found
+- core: language: use replacement map from bazarr
+- providers: titlovi: fix matching
+- providers: subscene: fix unknown language code error when "empty" result is returned
+- providers: subscene: add support for pt-BR (based on Diaoul/subliminal@b22cf08)
+- providers: subscene: explicitly set account filters for languages
+- providers: subscene: limit alternative searches to 3; set throttle to 8
+- providers: subscene: move login/cookies to initialization sequence
+- submod: generic: en: fix ";='s
 
 
 [older changes](CHANGELOG.md)
 
 
-Subtitles provided by [OpenSubtitles.org](http://www.opensubtitles.org/), [Podnapisi.NET](https://www.podnapisi.net/), [TVSubtitles.net](http://www.tvsubtitles.net/), [Addic7ed.com](http://www.addic7ed.com/), [Legendas TV](http://legendas.tv/), [Napi Projekt](http://www.napiprojekt.pl/), [Shooter](http://shooter.cn/), [Titlovi](http://titlovi.com), [aRGENTeaM](http://argenteam.net), [SubScene](https://subscene.com/), [Hosszupuska](http://hosszupuskasub.com/)
+Subtitles provided by [OpenSubtitles.org](http://www.opensubtitles.org/), [Podnapisi.NET](https://www.podnapisi.net/), [TVSubtitles.net](http://www.tvsubtitles.net/), [Addic7ed.com](http://www.addic7ed.com/), [Legendas TV](http://legendas.tv/), [Napi Projekt](http://www.napiprojekt.pl/), [Shooter](http://shooter.cn/), [Titlovi](http://titlovi.com), [aRGENTeaM](http://argenteam.net), [SubScene](https://subscene.com/), [Hosszupuska](http://hosszupuskasub.com/), [Napisy24](https://napisy24.pl/)
 
 [3rd party licenses](https://github.com/pannal/Sub-Zero.bundle/tree/master/Licenses)