aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/brightcove.py
blob: 63577a3cb91f3fe660333a197063530e05c25f44 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
# encoding: utf-8

import re
import json
import xml.etree.ElementTree

from .common import InfoExtractor
from ..utils import (
    compat_urllib_parse,
    find_xpath_attr,
    compat_urlparse,
    compat_str,
    compat_urllib_request,

    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'
    _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'

    _TESTS = [
        {
            # 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'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'uploader': u'8TV',
                u'description': u'md5:a950cc4285c43e44d763d036710cd9cd',
            }
        },
        {
            # From http://medianetwork.oracle.com/video/player/1785452137001
            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
            u'file': u'1785452137001.flv',
            u'info_dict': {
                u'title': u'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
                u'description': u'John Rose speaks at the JVM Language Summit, August 1, 2012.',
                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',
            },
        },
        {
            # test that the default referer works
            # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/
            u'url': u'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
            u'info_dict': {
                u'id': u'2878862109001',
                u'ext': u'mp4',
                u'title': u'Lost in Motion II',
                u'description': u'md5:363109c02998fee92ec02211bd8000df',
                u'uploader': u'National Ballet of Canada',                
            },
        },
    ]

    @classmethod
    def _build_brighcove_url(cls, object_str):
        """
        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'],
                  'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
                  }
        def find_param(name):
            node = find_xpath_attr(object_doc, './param', 'name', name)
            if node is not None:
                return node.attrib['value']
            return None
        playerKey = find_param('playerKey')
        # Not all pages define this value
        if playerKey is not None:
            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
        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 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):
        # 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, query,
                # We set the original url as the default 'Referer' header
                referer=url)
        else:
            player_key = query['playerKey']
            return self._get_playlist_info(player_key[0])

    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)

        self.report_extraction(video_id)
        info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
        info = json.loads(info)['data']
        video_info = info['programmedContent']['videoPlayer']['mediaDTO']

        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')

        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'],
                                    playlist_title=playlist_info['mediaCollectionDTO']['displayName'])

    def _extract_video_info(self, video_info):
        info = {
            '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'),
            'uploader': video_info.get('publisherName'),
        }

        renditions = video_info.get('renditions')
        if renditions:
            renditions = sorted(renditions, key=lambda r: r['size'])
            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'],
            })
        else:
            raise ExtractorError(u'Unable to extract video url for %s' % info['id'])
        return info