Prepare for upload.

[youtubedl] / youtube_dl / extractor / brightcove.py

diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py
index 558b3d0093975eff273d38a7a2bad9f1d4e36355..d8c35465a34fa4c4d4ca822d499892504a51ce62 100644 (file)
--- a/youtube_dl/extractor/brightcove.py
+++ b/youtube_dl/extractor/brightcove.py
@@ -9,10 +9,13 @@ from ..utils import (
     compat_urllib_parse,
     find_xpath_attr,
     compat_urlparse,
     compat_urllib_parse,
     find_xpath_attr,
     compat_urlparse,

+    compat_str,
+    compat_urllib_request,

 
     ExtractorError,
 )
 
 
     ExtractorError,
 )
 

+

 class BrightcoveIE(InfoExtractor):
     _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
 class BrightcoveIE(InfoExtractor):
     _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'


@@ -23,7 +26,7 @@ class BrightcoveIE(InfoExtractor):
             # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
             u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
             u'file': u'2371591881001.mp4',
             # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
             u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
             u'file': u'2371591881001.mp4',

-            u'md5': u'9e80619e0a94663f0bdc849b4566af19',
+            u'md5': u'8eccab865181d29ec2958f32a6a754f5',

             u'note': u'Test Brightcove downloads and detection in GenericIE',
             u'info_dict': {
                 u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
             u'note': u'Test Brightcove downloads and detection in GenericIE',
             u'info_dict': {
                 u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',


@@ -41,6 +44,17 @@ class BrightcoveIE(InfoExtractor):
                 u'uploader': u'Oracle',
             },
         },
                 u'uploader': u'Oracle',
             },
         },

+        {
+            # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/
+            u'url': u'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
+            u'info_dict': {
+                u'id': u'2750934548001',
+                u'ext': u'mp4',
+                u'title': u'This Bracelet Acts as a Personal Thermostat',
+                u'description': u'md5:547b78c64f4112766ccf4e151c20b6a0',
+                u'uploader': u'Mashable',
+            },
+        },

     ]
 
     @classmethod
     ]
 
     @classmethod


@@ -49,6 +63,13 @@ class BrightcoveIE(InfoExtractor):
         Build a Brightcove url from a xml string containing
         <object class="BrightcoveExperience">{params}</object>
         """
         Build a Brightcove url from a xml string containing
         <object class="BrightcoveExperience">{params}</object>
         """

+
+        # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553
+        object_str = re.sub(r'(<param name="[^"]+" value="[^"]+")>',
+                            lambda m: m.group(1) + '/>', object_str)
+        # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
+        object_str = object_str.replace(u'<--', u'<!--')
+

         object_doc = xml.etree.ElementTree.fromstring(object_str)
         assert u'BrightcoveExperience' in object_doc.attrib['class']
         params = {'flashID': object_doc.attrib['id'],
         object_doc = xml.etree.ElementTree.fromstring(object_str)
         assert u'BrightcoveExperience' in object_doc.attrib['class']
         params = {'flashID': object_doc.attrib['id'],


@@ -61,24 +82,48 @@ class BrightcoveIE(InfoExtractor):
         videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer')
         if videoPlayer is not None:
             params['@videoPlayer'] = videoPlayer.attrib['value']
         videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer')
         if videoPlayer is not None:
             params['@videoPlayer'] = videoPlayer.attrib['value']

+        linkBase = find_xpath_attr(object_doc, './param', 'name', 'linkBaseURL')
+        if linkBase is not None:
+            params['linkBaseURL'] = linkBase.attrib['value']

         data = compat_urllib_parse.urlencode(params)
         return cls._FEDERATED_URL_TEMPLATE % data
 
         data = compat_urllib_parse.urlencode(params)
         return cls._FEDERATED_URL_TEMPLATE % data
 

+    @classmethod
+    def _extract_brightcove_url(cls, webpage):
+        """Try to extract the brightcove url from the wepbage, returns None
+        if it can't be found
+        """
+        m_brightcove = re.search(
+            r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>',
+            webpage, re.DOTALL)
+        if m_brightcove is not None:
+            return cls._build_brighcove_url(m_brightcove.group())
+        else:
+            return None
+

     def _real_extract(self, url):
     def _real_extract(self, url):

+        # Change the 'videoId' and others field to '@videoPlayer'
+        url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url)
+        # Change bckey (used by bcove.me urls) to playerKey
+        url = re.sub(r'(?<=[?&])bckey', 'playerKey', url)

         mobj = re.match(self._VALID_URL, url)
         query_str = mobj.group('query')
         query = compat_urlparse.parse_qs(query_str)
 
         videoPlayer = query.get('@videoPlayer')
         if videoPlayer:
         mobj = re.match(self._VALID_URL, url)
         query_str = mobj.group('query')
         query = compat_urlparse.parse_qs(query_str)
 
         videoPlayer = query.get('@videoPlayer')
         if videoPlayer:

-            return self._get_video_info(videoPlayer[0], query_str)
+            return self._get_video_info(videoPlayer[0], query_str, query)

         else:
             player_key = query['playerKey']
             return self._get_playlist_info(player_key[0])
 
         else:
             player_key = query['playerKey']
             return self._get_playlist_info(player_key[0])
 

-    def _get_video_info(self, video_id, query):
-        request_url = self._FEDERATED_URL_TEMPLATE % query
-        webpage = self._download_webpage(request_url, video_id)
+    def _get_video_info(self, video_id, query_str, query):
+        request_url = self._FEDERATED_URL_TEMPLATE % query_str
+        req = compat_urllib_request.Request(request_url)
+        linkBase = query.get('linkBaseURL')
+        if linkBase is not None:
+            req.add_header('Referer', linkBase[0])
+        webpage = self._download_webpage(req, video_id)

 
         self.report_extraction(video_id)
         info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
 
         self.report_extraction(video_id)
         info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')


@@ -91,7 +136,10 @@ class BrightcoveIE(InfoExtractor):
         playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
                                                player_key, u'Downloading playlist information')
 
         playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
                                                player_key, u'Downloading playlist information')
 

-        playlist_info = json.loads(playlist_info)['videoList']
+        json_data = json.loads(playlist_info)
+        if 'videoList' not in json_data:
+            raise ExtractorError(u'Empty playlist')
+        playlist_info = json_data['videoList']

         videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
 
         return self.playlist_result(videos, playlist_id=playlist_info['id'],
         videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
 
         return self.playlist_result(videos, playlist_id=playlist_info['id'],


@@ -99,7 +147,7 @@ class BrightcoveIE(InfoExtractor):
 
     def _extract_video_info(self, video_info):
         info = {
 
     def _extract_video_info(self, video_info):
         info = {

-            'id': video_info['id'],
+            'id': compat_str(video_info['id']),

             'title': video_info['displayName'],
             'description': video_info.get('shortDescription'),
             'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
             'title': video_info['displayName'],
             'description': video_info.get('shortDescription'),
             'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),


@@ -109,15 +157,14 @@ class BrightcoveIE(InfoExtractor):
         renditions = video_info.get('renditions')
         if renditions:
             renditions = sorted(renditions, key=lambda r: r['size'])
         renditions = video_info.get('renditions')
         if renditions:
             renditions = sorted(renditions, key=lambda r: r['size'])

-            best_format = renditions[-1]
-            info.update({
-                'url': best_format['defaultURL'],
-                'ext': 'mp4',
-            })
+            info['formats'] = [{
+                'url': rend['defaultURL'],
+                'height': rend.get('frameHeight'),
+                'width': rend.get('frameWidth'),
+            } for rend in renditions]

         elif video_info.get('FLVFullLengthURL') is not None:
             info.update({
                 'url': video_info['FLVFullLengthURL'],
         elif video_info.get('FLVFullLengthURL') is not None:
             info.update({
                 'url': video_info['FLVFullLengthURL'],

-                'ext': 'flv',

             })
         else:
             raise ExtractorError(u'Unable to extract video url for %s' % info['id'])
             })
         else:
             raise ExtractorError(u'Unable to extract video url for %s' % info['id'])