]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/brightcove.py
Merge tag 'upstream/2015.02.06'
[youtubedl] / youtube_dl / extractor / brightcove.py
index 71e3c7883338154eea0c3d369a0fdd0bee828e26..ea0969d4d259a99653bebbcabcebb0e1f87719f3 100644 (file)
+# encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 import json
 import xml.etree.ElementTree
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
+    compat_parse_qs,
+    compat_str,
     compat_urllib_parse,
-    find_xpath_attr,
+    compat_urllib_parse_urlparse,
+    compat_urllib_request,
     compat_urlparse,
 )
+from ..utils import (
+    determine_ext,
+    ExtractorError,
+    find_xpath_attr,
+    fix_xml_ampersands,
+    unescapeHTML,
+    unsmuggle_url,
+)
+
 
 class BrightcoveIE(InfoExtractor):
-    _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
+    _VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
-    _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'
-    
-    # There is a test for Brigtcove in GenericIE, that way we test both the download
-    # and the detection of videos, and we don't have to find an URL that is always valid
+
+    _TESTS = [
+        {
+            # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
+            'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
+            'md5': '5423e113865d26e40624dce2e4b45d95',
+            'note': 'Test Brightcove downloads and detection in GenericIE',
+            'info_dict': {
+                'id': '2371591881001',
+                'ext': 'mp4',
+                'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
+                'uploader': '8TV',
+                'description': 'md5:a950cc4285c43e44d763d036710cd9cd',
+            }
+        },
+        {
+            # From http://medianetwork.oracle.com/video/player/1785452137001
+            'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
+            'info_dict': {
+                'id': '1785452137001',
+                'ext': 'flv',
+                'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
+                'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.',
+                'uploader': 'Oracle',
+            },
+        },
+        {
+            # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/
+            'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
+            'info_dict': {
+                'id': '2750934548001',
+                'ext': 'mp4',
+                'title': 'This Bracelet Acts as a Personal Thermostat',
+                'description': 'md5:547b78c64f4112766ccf4e151c20b6a0',
+                'uploader': 'Mashable',
+            },
+        },
+        {
+            # test that the default referer works
+            # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/
+            'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
+            'info_dict': {
+                'id': '2878862109001',
+                'ext': 'mp4',
+                'title': 'Lost in Motion II',
+                'description': 'md5:363109c02998fee92ec02211bd8000df',
+                'uploader': 'National Ballet of Canada',
+            },
+        },
+        {
+            # test flv videos served by akamaihd.net
+            # From http://www.redbull.com/en/bike/stories/1331655643987/replay-uci-dh-world-cup-2014-from-fort-william
+            'url': 'http://c.brightcove.com/services/viewer/htmlFederated?%40videoPlayer=ref%3ABC2996102916001&linkBaseURL=http%3A%2F%2Fwww.redbull.com%2Fen%2Fbike%2Fvideos%2F1331655630249%2Freplay-uci-fort-william-2014-dh&playerKey=AQ%7E%7E%2CAAAApYJ7UqE%7E%2Cxqr_zXk0I-zzNndy8NlHogrCb5QdyZRf&playerID=1398061561001#__youtubedl_smuggle=%7B%22Referer%22%3A+%22http%3A%2F%2Fwww.redbull.com%2Fen%2Fbike%2Fstories%2F1331655643987%2Freplay-uci-dh-world-cup-2014-from-fort-william%22%7D',
+            # The md5 checksum changes on each download
+            'info_dict': {
+                'id': '2996102916001',
+                'ext': 'flv',
+                'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
+                'uploader': 'Red Bull TV',
+                'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
+            },
+        },
+        {
+            # playlist test
+            # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players
+            'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL',
+            'info_dict': {
+                'title': 'Sealife',
+            },
+            'playlist_mincount': 7,
+        },
+    ]
 
     @classmethod
     def _build_brighcove_url(cls, object_str):
@@ -23,64 +106,211 @@ class BrightcoveIE(InfoExtractor):
         Build a Brightcove url from a xml string containing
         <object class="BrightcoveExperience">{params}</object>
         """
-        object_doc = xml.etree.ElementTree.fromstring(object_str)
-        assert u'BrightcoveExperience' in object_doc.attrib['class']
-        params = {'flashID': object_doc.attrib['id'],
-                  'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
-                  }
-        playerKey = find_xpath_attr(object_doc, './param', 'name', 'playerKey')
+
+        # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553
+        object_str = re.sub(r'(<param(?:\s+[a-zA-Z0-9_]+="[^"]*")*)>',
+                            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('<--', '<!--')
+        # remove namespace to simplify extraction
+        object_str = re.sub(r'(<object[^>]*)(xmlns=".*?")', r'\1', object_str)
+        object_str = fix_xml_ampersands(object_str)
+
+        object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
+
+        fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')
+        if fv_el is not None:
+            flashvars = dict(
+                (k, v[0])
+                for k, v in compat_parse_qs(fv_el.attrib['value']).items())
+        else:
+            flashvars = {}
+
+        def find_param(name):
+            if name in flashvars:
+                return flashvars[name]
+            node = find_xpath_attr(object_doc, './param', 'name', name)
+            if node is not None:
+                return node.attrib['value']
+            return None
+
+        params = {}
+
+        playerID = find_param('playerID')
+        if playerID is None:
+            raise ExtractorError('Cannot find player ID')
+        params['playerID'] = playerID
+
+        playerKey = find_param('playerKey')
         # Not all pages define this value
         if playerKey is not None:
-            params['playerKey'] = playerKey.attrib['value']
-        videoPlayer = find_xpath_attr(object_doc, './param', 'name', '@videoPlayer')
+            params['playerKey'] = playerKey
+        # The three fields hold the id of the video
+        videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID')
         if videoPlayer is not None:
-            params['@videoPlayer'] = videoPlayer.attrib['value']
+            params['@videoPlayer'] = videoPlayer
+        linkBase = find_param('linkBaseURL')
+        if linkBase is not None:
+            params['linkBaseURL'] = linkBase
         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 webpage, returns None
+        if it can't be found
+        """
+        urls = cls._extract_brightcove_urls(webpage)
+        return urls[0] if urls else None
+
+    @classmethod
+    def _extract_brightcove_urls(cls, webpage):
+        """Return a list of all Brightcove URLs from the webpage """
+
+        url_m = re.search(
+            r'<meta\s+property="og:video"\s+content="(https?://(?:secure|c)\.brightcove.com/[^"]+)"',
+            webpage)
+        if url_m:
+            url = unescapeHTML(url_m.group(1))
+            # Some sites don't add it, we can't download with this url, for example:
+            # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
+            if 'playerKey' in url or 'videoId' in url:
+                return [url]
+
+        matches = re.findall(
+            r'''(?sx)<object
+            (?:
+                [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] |
+                [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
+            ).+?</object>''',
+            webpage)
+        return [cls._build_brighcove_url(m) for m in matches]
+
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(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:
-            return self._get_video_info(videoPlayer[0], query_str)
-        else:
+            # We set the original url as the default 'Referer' header
+            referer = smuggled_data.get('Referer', url)
+            return self._get_video_info(
+                videoPlayer[0], query_str, query, referer=referer)
+        elif 'playerKey' in query:
             player_key = query['playerKey']
             return self._get_playlist_info(player_key[0])
+        else:
+            raise ExtractorError(
+                'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?',
+                expected=True)
 
-    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, referer=None):
+        request_url = self._FEDERATED_URL_TEMPLATE % query_str
+        req = compat_urllib_request.Request(request_url)
+        linkBase = query.get('linkBaseURL')
+        if linkBase is not None:
+            referer = linkBase[0]
+        if referer is not None:
+            req.add_header('Referer', referer)
+        webpage = self._download_webpage(req, video_id)
+
+        error_msg = self._html_search_regex(
+            r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage,
+            'error message', default=None)
+        if error_msg is not None:
+            raise ExtractorError(
+                'brightcove said: %s' % error_msg, expected=True)
 
         self.report_extraction(video_id)
-        info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
+        info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json')
         info = json.loads(info)['data']
         video_info = info['programmedContent']['videoPlayer']['mediaDTO']
+        video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
 
         return self._extract_video_info(video_info)
 
     def _get_playlist_info(self, player_key):
-        playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
-                                               player_key, u'Downloading playlist information')
+        info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key
+        playlist_info = self._download_webpage(
+            info_url, player_key, 'Downloading playlist information')
 
-        playlist_info = json.loads(playlist_info)['videoList']
+        json_data = json.loads(playlist_info)
+        if 'videoList' not in json_data:
+            raise ExtractorError('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'],
                                     playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
 
     def _extract_video_info(self, video_info):
-        renditions = video_info['renditions']
-        renditions = sorted(renditions, key=lambda r: r['size'])
-        best_format = renditions[-1]
+        info = {
+            'id': compat_str(video_info['id']),
+            'title': video_info['displayName'].strip(),
+            'description': video_info.get('shortDescription'),
+            'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
+            'uploader': video_info.get('publisherName'),
+        }
 
-        return {'id': video_info['id'],
-                'title': video_info['displayName'],
-                'url': best_format['defaultURL'], 
-                'ext': 'mp4',
-                'description': video_info.get('shortDescription'),
-                'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
-                'uploader': video_info.get('publisherName'),
+        renditions = video_info.get('renditions')
+        if renditions:
+            formats = []
+            for rend in renditions:
+                url = rend['defaultURL']
+                if not url:
+                    continue
+                ext = None
+                if rend['remote']:
+                    url_comp = compat_urllib_parse_urlparse(url)
+                    if url_comp.path.endswith('.m3u8'):
+                        formats.extend(
+                            self._extract_m3u8_formats(url, info['id'], 'mp4'))
+                        continue
+                    elif 'akamaihd.net' in url_comp.netloc:
+                        # This type of renditions are served through
+                        # akamaihd.net, but they don't use f4m manifests
+                        url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB'
+                        ext = 'flv'
+                if ext is None:
+                    ext = determine_ext(url)
+                size = rend.get('size')
+                formats.append({
+                    'url': url,
+                    'ext': ext,
+                    'height': rend.get('frameHeight'),
+                    'width': rend.get('frameWidth'),
+                    'filesize': size if size != 0 else None,
+                })
+            self._sort_formats(formats)
+            info['formats'] = formats
+        elif video_info.get('FLVFullLengthURL') is not None:
+            info.update({
+                'url': video_info['FLVFullLengthURL'],
+            })
+
+        if self._downloader.params.get('include_ads', False):
+            adServerURL = video_info.get('_youtubedl_adServerURL')
+            if adServerURL:
+                ad_info = {
+                    '_type': 'url',
+                    'url': adServerURL,
                 }
+                if 'url' in info:
+                    return {
+                        '_type': 'playlist',
+                        'title': info['title'],
+                        'entries': [ad_info, info],
+                    }
+                else:
+                    return ad_info
+
+        if 'url' not in info and not info.get('formats'):
+            raise ExtractorError('Unable to extract video url for %s' % info['id'])
+        return info