]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/nfb.py
d/p/remove-autoupdate-mechanism.patch: Update patch with metadata at the top.
[youtubedl] / youtube_dl / extractor / nfb.py
index e88566c693c9139ae725604427fb8cd3894e7e3b..adcc636bc32c062fec74074044783e452a3725d9 100644 (file)
@@ -1,29 +1,31 @@
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
 from .common import InfoExtractor
 from ..utils import (
-    compat_urllib_request,
-    compat_urllib_parse,
+    clean_html,
+    determine_ext,
+    int_or_none,
+    qualities,
+    urlencode_postdata,
+    xpath_text,
 )
 
 
 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',
         'info_dict': {
             'id': 'qallunaat_why_white_people_are_funny',
 
     _TEST = {
         'url': 'https://www.nfb.ca/film/qallunaat_why_white_people_are_funny',
         'info_dict': {
             'id': 'qallunaat_why_white_people_are_funny',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Qallunaat! Why White People Are Funny ',
             'title': 'Qallunaat! Why White People Are Funny ',
-            'description': 'md5:836d8aff55e087d04d9f6df554d4e038',
+            'description': 'md5:6b8e32dde3abf91e58857b174916620c',
             'duration': 3128,
             'duration': 3128,
+            'creator': 'Mark Sandiford',
             'uploader': 'Mark Sandiford',
             'uploader': 'Mark Sandiford',
-            'uploader_id': 'mark-sandiford',
         },
         'params': {
             # rtmp download
         },
         'params': {
             # rtmp download
@@ -32,63 +34,79 @@ 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')
-
-        uploader_id = self._html_search_regex(r'<a class="director-link" href="/explore-all-directors/([^/]+)/"',
-            page, 'director id', fatal=False)
-        uploader = self._html_search_regex(r'<em class="director-name" itemprop="name">([^<]+)</em>',
-            page, 'director name', fatal=False)
+        video_id = self._match_id(url)
 
 
-        request = compat_urllib_request.Request('https://www.nfb.ca/film/%s/player_config' % video_id,
-            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')
+        config = self._download_xml(
+            'https://www.nfb.ca/film/%s/player_config' % video_id,
+            video_id, 'Downloading player config XML',
+            data=urlencode_postdata({'getConfig': 'true'}),
+            headers={
+                'Content-Type': 'application/x-www-form-urlencoded',
+                'X-NFB-Referer': 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf'
+            })
 
 
-        config = self._download_xml(request, video_id, 'Downloading player config XML')
-
-        title = None
-        description = None
-        thumbnail = None
-        duration = None
-        formats = []
-
-        def extract_thumbnail(media):
-            thumbnails = {}
-            for asset in media.findall('assets/asset'):
-                thumbnails[asset.get('quality')] = asset.find('default/url').text
-            if not thumbnails:
-                return None
-            if 'high' in thumbnails:
-                return thumbnails['high']
-            return list(thumbnails.values())[0]
+        title, description, thumbnail, duration, uploader, author = [None] * 6
+        thumbnails, formats = [[]] * 2
+        subtitles = {}
 
         for media in config.findall('./player/stream/media'):
             if media.get('type') == 'posterImage':
 
         for media in config.findall('./player/stream/media'):
             if media.get('type') == 'posterImage':
-                thumbnail = extract_thumbnail(media)
+                quality_key = qualities(('low', 'high'))
+                thumbnails = []
+                for asset in media.findall('assets/asset'):
+                    asset_url = xpath_text(asset, 'default/url', default=None)
+                    if not asset_url:
+                        continue
+                    quality = asset.get('quality')
+                    thumbnails.append({
+                        'url': asset_url,
+                        'id': quality,
+                        'preference': quality_key(quality),
+                    })
             elif media.get('type') == 'video':
             elif media.get('type') == 'video':
-                duration = int(media.get('duration'))
-                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')]
+                title = xpath_text(media, 'title', fatal=True)
+                for asset in media.findall('assets/asset'):
+                    quality = asset.get('quality')
+                    height = int_or_none(self._search_regex(
+                        r'^(\d+)[pP]$', quality or '', 'height', default=None))
+                    for node in asset:
+                        streamer = xpath_text(node, 'streamerURI', default=None)
+                        if not streamer:
+                            continue
+                        play_path = xpath_text(node, 'url', default=None)
+                        if not play_path:
+                            continue
+                        formats.append({
+                            'url': streamer,
+                            'app': streamer.split('/', 3)[3],
+                            'play_path': play_path,
+                            'rtmp_live': False,
+                            'ext': 'flv',
+                            'format_id': '%s-%s' % (node.tag, quality) if quality else node.tag,
+                            'height': height,
+                        })
+                self._sort_formats(formats)
+                description = clean_html(xpath_text(media, 'description'))
+                uploader = xpath_text(media, 'author')
+                duration = int_or_none(media.get('duration'))
+                for subtitle in media.findall('./subtitles/subtitle'):
+                    subtitle_url = xpath_text(subtitle, 'url', default=None)
+                    if not subtitle_url:
+                        continue
+                    lang = xpath_text(subtitle, 'lang', default='en')
+                    subtitles.setdefault(lang, []).append({
+                        'url': subtitle_url,
+                        'ext': (subtitle.get('format') or determine_ext(subtitle_url)).lower(),
+                    })
 
         return {
             'id': video_id,
             'title': title,
             'description': description,
 
         return {
             'id': video_id,
             'title': title,
             'description': description,
-            'thumbnail': thumbnail,
+            'thumbnails': thumbnails,
             'duration': duration,
             'duration': duration,
+            'creator': uploader,
             'uploader': uploader,
             'uploader': uploader,
-            'uploader_id': uploader_id,
             'formats': formats,
             'formats': formats,
-        }
\ No newline at end of file
+            'subtitles': subtitles,
+        }