]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/nfb.py
Initiate new release.
[youtubedl] / youtube_dl / extractor / nfb.py
index e88566c693c9139ae725604427fb8cd3894e7e3b..ea077254b4320fe18e59eb9b67461b13c146b873 100644 (file)
@@ -1,9 +1,7 @@
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_urllib_request,
     compat_urllib_parse,
 )
     compat_urllib_request,
     compat_urllib_parse,
 )
@@ -12,7 +10,7 @@ from ..utils import (
 class NFBIE(InfoExtractor):
     IE_NAME = 'nfb'
     IE_DESC = 'National Film Board of Canada'
 class NFBIE(InfoExtractor):
     IE_NAME = 'nfb'
     IE_DESC = 'National Film Board of Canada'
-    _VALID_URL = r'https?://(?:www\.)?(nfb|onf)\.ca/film/(?P<id>[\da-z_-]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:nfb|onf)\.ca/film/(?P<id>[\da-z_-]+)'
 
     _TEST = {
         'url': 'https://www.nfb.ca/film/qallunaat_why_white_people_are_funny',
 
     _TEST = {
         'url': 'https://www.nfb.ca/film/qallunaat_why_white_people_are_funny',
@@ -32,18 +30,18 @@ class NFBIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
-        page = self._download_webpage('https://www.nfb.ca/film/%s' % video_id, video_id, 'Downloading film page')
+        video_id = self._match_id(url)
+        page = self._download_webpage(
+            'https://www.nfb.ca/film/%s' % video_id, video_id,
+            'Downloading film page')
 
         uploader_id = self._html_search_regex(r'<a class="director-link" href="/explore-all-directors/([^/]+)/"',
 
         uploader_id = self._html_search_regex(r'<a class="director-link" href="/explore-all-directors/([^/]+)/"',
-            page, 'director id', fatal=False)
+                                              page, 'director id', fatal=False)
         uploader = self._html_search_regex(r'<em class="director-name" itemprop="name">([^<]+)</em>',
         uploader = self._html_search_regex(r'<em class="director-name" itemprop="name">([^<]+)</em>',
-            page, 'director name', fatal=False)
+                                           page, 'director name', fatal=False)
 
         request = compat_urllib_request.Request('https://www.nfb.ca/film/%s/player_config' % video_id,
 
         request = compat_urllib_request.Request('https://www.nfb.ca/film/%s/player_config' % video_id,
-            compat_urllib_parse.urlencode({'getConfig': 'true'}).encode('ascii'))
+                                                compat_urllib_parse.urlencode({'getConfig': 'true'}).encode('ascii'))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         request.add_header('X-NFB-Referer', 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf')
 
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         request.add_header('X-NFB-Referer', 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf')
 
@@ -73,14 +71,16 @@ class NFBIE(InfoExtractor):
                 title = media.find('title').text
                 description = media.find('description').text
                 # It seems assets always go from lower to better quality, so no need to sort
                 title = media.find('title').text
                 description = media.find('description').text
                 # It seems assets always go from lower to better quality, so no need to sort
-                formats = [{
-                    'url': x.find('default/streamerURI').text,
-                    'app': x.find('default/streamerURI').text.split('/', 3)[3],
-                    'play_path': x.find('default/url').text,
-                    'rtmp_live': False,
-                    'ext': 'mp4',
-                    'format_id': x.get('quality'),
-                } for x in media.findall('assets/asset')]
+                for asset in media.findall('assets/asset'):
+                    for x in asset:
+                        formats.append({
+                            'url': x.find('streamerURI').text,
+                            'app': x.find('streamerURI').text.split('/', 3)[3],
+                            'play_path': x.find('url').text,
+                            'rtmp_live': False,
+                            'ext': 'mp4',
+                            'format_id': '%s-%s' % (x.tag, asset.get('quality')),
+                        })
 
         return {
             'id': video_id,
 
         return {
             'id': video_id,
@@ -91,4 +91,4 @@ class NFBIE(InfoExtractor):
             'uploader': uploader,
             'uploader_id': uploader_id,
             'formats': formats,
             'uploader': uploader,
             'uploader_id': uploader_id,
             'formats': formats,
-        }
\ No newline at end of file
+        }