[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / cspan.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7     determine_ext,
8     ExtractorError,
9     extract_attributes,
10     find_xpath_attr,
11     get_element_by_class,
12     int_or_none,
13     smuggle_url,
14     unescapeHTML,
15 )
16 from .senateisvp import SenateISVPIE
17 from .ustream import UstreamIE
18
19
20 class CSpanIE(InfoExtractor):
21     _VALID_URL = r'https?://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
22     IE_DESC = 'C-SPAN'
23     _TESTS = [{
24         'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
25         'md5': '94b29a4f131ff03d23471dd6f60b6a1d',
26         'info_dict': {
27             'id': '315139',
28             'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
29         },
30         'playlist_mincount': 2,
31         'skip': 'Regularly fails on travis, for unknown reasons',
32     }, {
33         'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models',
34         # md5 is unstable
35         'info_dict': {
36             'id': 'c4486943',
37             'ext': 'mp4',
38             'title': 'CSPAN - International Health Care Models',
39             'description': 'md5:7a985a2d595dba00af3d9c9f0783c967',
40         }
41     }, {
42         'url': 'http://www.c-span.org/video/?318608-1/gm-ignition-switch-recall',
43         'info_dict': {
44             'id': '342759',
45             'title': 'General Motors Ignition Switch Recall',
46         },
47         'playlist_mincount': 6,
48     }, {
49         # Video from senate.gov
50         'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
51         'info_dict': {
52             'id': 'judiciary031715',
53             'ext': 'mp4',
54             'title': 'Immigration Reforms Needed to Protect Skilled American Workers',
55         },
56         'params': {
57             'skip_download': True,  # m3u8 downloads
58         }
59     }, {
60         # Ustream embedded video
61         'url': 'https://www.c-span.org/video/?114917-1/armed-services',
62         'info_dict': {
63             'id': '58428542',
64             'ext': 'flv',
65             'title': 'USHR07 Armed Services Committee',
66             'description': 'hsas00-2118-20150204-1000et-07\n\n\nUSHR07 Armed Services Committee',
67             'timestamp': 1423060374,
68             'upload_date': '20150204',
69             'uploader': 'HouseCommittee',
70             'uploader_id': '12987475',
71         },
72     }, {
73         # Audio Only
74         'url': 'https://www.c-span.org/video/?437336-1/judiciary-antitrust-competition-policy-consumer-rights',
75         'only_matching': True,
76     }]
77     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
78
79     def _real_extract(self, url):
80         video_id = self._match_id(url)
81         video_type = None
82         webpage = self._download_webpage(url, video_id)
83
84         ustream_url = UstreamIE._extract_url(webpage)
85         if ustream_url:
86             return self.url_result(ustream_url, UstreamIE.ie_key())
87
88         if '&vod' not in url:
89             bc = self._search_regex(
90                 r"(<[^>]+id='brightcove-player-embed'[^>]+>)",
91                 webpage, 'brightcove embed', default=None)
92             if bc:
93                 bc_attr = extract_attributes(bc)
94                 bc_url = self.BRIGHTCOVE_URL_TEMPLATE % (
95                     bc_attr.get('data-bcaccountid', '3162030207001'),
96                     bc_attr.get('data-noprebcplayerid', 'SyGGpuJy3g'),
97                     bc_attr.get('data-newbcplayerid', 'default'),
98                     bc_attr['data-bcid'])
99                 return self.url_result(smuggle_url(bc_url, {'source_url': url}))
100
101         # We first look for clipid, because clipprog always appears before
102         patterns = [r'id=\'clip(%s)\'\s*value=\'([0-9]+)\'' % t for t in ('id', 'prog')]
103         results = list(filter(None, (re.search(p, webpage) for p in patterns)))
104         if results:
105             matches = results[0]
106             video_type, video_id = matches.groups()
107             video_type = 'clip' if video_type == 'id' else 'program'
108         else:
109             m = re.search(r'data-(?P<type>clip|prog)id=["\'](?P<id>\d+)', webpage)
110             if m:
111                 video_id = m.group('id')
112                 video_type = 'program' if m.group('type') == 'prog' else 'clip'
113             else:
114                 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
115                 if senate_isvp_url:
116                     title = self._og_search_title(webpage)
117                     surl = smuggle_url(senate_isvp_url, {'force_title': title})
118                     return self.url_result(surl, 'SenateISVP', video_id, title)
119                 video_id = self._search_regex(
120                     r'jwsetup\.clipprog\s*=\s*(\d+);',
121                     webpage, 'jwsetup program id', default=None)
122                 if video_id:
123                     video_type = 'program'
124         if video_type is None or video_id is None:
125             error_message = get_element_by_class('VLplayer-error-message', webpage)
126             if error_message:
127                 raise ExtractorError(error_message)
128             raise ExtractorError('unable to find video id and type')
129
130         def get_text_attr(d, attr):
131             return d.get(attr, {}).get('#text')
132
133         data = self._download_json(
134             'http://www.c-span.org/assets/player/ajax-player.php?os=android&html5=%s&id=%s' % (video_type, video_id),
135             video_id)['video']
136         if data['@status'] != 'Success':
137             raise ExtractorError('%s said: %s' % (self.IE_NAME, get_text_attr(data, 'error')), expected=True)
138
139         doc = self._download_xml(
140             'http://www.c-span.org/common/services/flashXml.php?%sid=%s' % (video_type, video_id),
141             video_id)
142
143         description = self._html_search_meta('description', webpage)
144
145         title = find_xpath_attr(doc, './/string', 'name', 'title').text
146         thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text
147
148         files = data['files']
149         capfile = get_text_attr(data, 'capfile')
150
151         entries = []
152         for partnum, f in enumerate(files):
153             formats = []
154             for quality in f.get('qualities', []):
155                 formats.append({
156                     'format_id': '%s-%sp' % (get_text_attr(quality, 'bitrate'), get_text_attr(quality, 'height')),
157                     'url': unescapeHTML(get_text_attr(quality, 'file')),
158                     'height': int_or_none(get_text_attr(quality, 'height')),
159                     'tbr': int_or_none(get_text_attr(quality, 'bitrate')),
160                 })
161             if not formats:
162                 path = unescapeHTML(get_text_attr(f, 'path'))
163                 if not path:
164                     continue
165                 formats = self._extract_m3u8_formats(
166                     path, video_id, 'mp4', entry_protocol='m3u8_native',
167                     m3u8_id='hls') if determine_ext(path) == 'm3u8' else [{'url': path, }]
168             self._sort_formats(formats)
169             entries.append({
170                 'id': '%s_%d' % (video_id, partnum + 1),
171                 'title': (
172                     title if len(files) == 1 else
173                     '%s part %d' % (title, partnum + 1)),
174                 'formats': formats,
175                 'description': description,
176                 'thumbnail': thumbnail,
177                 'duration': int_or_none(get_text_attr(f, 'length')),
178                 'subtitles': {
179                     'en': [{
180                         'url': capfile,
181                         'ext': determine_ext(capfile, 'dfxp')
182                     }],
183                 } if capfile else None,
184             })
185
186         if len(entries) == 1:
187             entry = dict(entries[0])
188             entry['id'] = 'c' + video_id if video_type == 'clip' else video_id
189             return entry
190         else:
191             return {
192                 '_type': 'playlist',
193                 'entries': entries,
194                 'title': title,
195                 'id': 'c' + video_id if video_type == 'clip' else video_id,
196             }