Library update

libs/subliminal/services/subtitulos.py

@@ -18,8 +18,10 @@
 from . import ServiceBase
 from ..subtitles import get_subtitle_path, ResultSubtitle
 from ..videos import Episode
+from bs4 import BeautifulSoup
+from guessit.language import lang_set
 from subliminal.utils import get_keywords, split_keyword
-import BeautifulSoup
+import guessit
 import logging
 import re
 import unicodedata
@@ -32,19 +34,19 @@ logger = logging.getLogger(__name__)
 class Subtitulos(ServiceBase):
     server_url = 'http://www.subtitulos.es'
     api_based = False
-    languages = {u'English (US)': 'en', u'English (UK)': 'en', u'English': 'en', u'French': 'fr', u'Brazilian': 'po',
-                 u'Portuguese': 'pt', u'Español (Latinoamérica)': 'es', u'Español (España)': 'es', u'Español': 'es',
-                 u'Italian': 'it', u'Català': 'ca'}
-    reverted_languages = True
+    languages = lang_set([u'English (US)', u'English (UK)', u'English', u'French', u'Brazilian',
+                          u'Portuguese', u'Español (Latinoamérica)', u'Español (España)', u'Español',
+                          u'Italian', u'Català'], strict=True)
     videos = [Episode]
     require_video = False
-    release_pattern = re.compile('Versión (.+) ([0-9]+).([0-9])+ megabytes')
+    required_features = ['permissive']
+    # the '.+' in the pattern for Version allows us to match both 'ó'
+    # and the 'ó' char directly. This is because now BS4 converts the html
+    # code chars into their equivalent unicode char
+    release_pattern = re.compile('Versi.+n (.+) ([0-9]+).([0-9])+ megabytes')
 
-    def list(self, video, languages):
-        if not self.check_validity(video, languages):
-            return []
-        results = self.query(video.path or video.release, languages, get_keywords(video.guess), video.series, video.season, video.episode)
-        return results
+    def list_checked(self, video, languages):
+        return self.query(video.path or video.release, languages, get_keywords(video.guess), video.series, video.season, video.episode)
 
     def query(self, filepath, languages, keywords, series, season, episode):
         request_series = series.lower().replace(' ', '_')
@@ -58,7 +60,7 @@ class Subtitulos(ServiceBase):
         if r.status_code != 200:
             logger.error(u'Request %s returned status code %d' % (r.url, r.status_code))
             return []
-        soup = BeautifulSoup.BeautifulSoup(r.content)
+        soup = BeautifulSoup(r.content, self.required_features)
         subtitles = []
         for sub in soup('div', {'id': 'version'}):
             sub_keywords = split_keyword(self.release_pattern.search(sub.find('p', {'class': 'title-sub'}).contents[1]).group(1).lower())
@@ -66,8 +68,8 @@ class Subtitulos(ServiceBase):
                 logger.debug(u'None of subtitle keywords %r in %r' % (sub_keywords, keywords))
                 continue
             for html_language in sub.findAllNext('ul', {'class': 'sslist'}):
-                language = self.get_revert_language(html_language.findNext('li', {'class': 'li-idioma'}).find('strong').contents[0].string.strip())
-                if not language in languages:
+                language = guessit.Language(html_language.findNext('li', {'class': 'li-idioma'}).find('strong').contents[0].string.strip())
+                if language not in languages:
                     logger.debug(u'Language %r not in wanted languages %r' % (language, languages))
                     continue
                 html_status = html_language.findNext('li', {'class': 'li-estado green'})
@@ -80,4 +82,5 @@ class Subtitulos(ServiceBase):
                 subtitles.append(subtitle)
         return subtitles
 
+
 Service = Subtitulos