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     js_to_json,
  14     merge_dicts,
  15     smuggle_url,
  16     unescapeHTML,
  17 )
  18 from .senateisvp import SenateISVPIE
  19 from .ustream import UstreamIE
  20
  21
  22 class CSpanIE(InfoExtractor):
  23     _VALID_URL = r'https?://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
  24     IE_DESC = 'C-SPAN'
  25     _TESTS = [{
  26         'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
  27         'md5': '94b29a4f131ff03d23471dd6f60b6a1d',
  28         'info_dict': {
  29             'id': '315139',
  30             'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
  31         },
  32         'playlist_mincount': 2,
  33         'skip': 'Regularly fails on travis, for unknown reasons',
  34     }, {
  35         'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models',
  36         # md5 is unstable
  37         'info_dict': {
  38             'id': 'c4486943',
  39             'ext': 'mp4',
  40             'title': 'CSPAN - International Health Care Models',
  41             'description': 'md5:7a985a2d595dba00af3d9c9f0783c967',
  42         }
  43     }, {
  44         'url': 'http://www.c-span.org/video/?318608-1/gm-ignition-switch-recall',
  45         'info_dict': {
  46             'id': '342759',
  47             'title': 'General Motors Ignition Switch Recall',
  48         },
  49         'playlist_mincount': 6,
  50     }, {
  51         # Video from senate.gov
  52         'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
  53         'info_dict': {
  54             'id': 'judiciary031715',
  55             'ext': 'mp4',
  56             'title': 'Immigration Reforms Needed to Protect Skilled American Workers',
  57         },
  58         'params': {
  59             'skip_download': True,  # m3u8 downloads
  60         }
  61     }, {
  62         # Ustream embedded video
  63         'url': 'https://www.c-span.org/video/?114917-1/armed-services',
  64         'info_dict': {
  65             'id': '58428542',
  66             'ext': 'flv',
  67             'title': 'USHR07 Armed Services Committee',
  68             'description': 'hsas00-2118-20150204-1000et-07\n\n\nUSHR07 Armed Services Committee',
  69             'timestamp': 1423060374,
  70             'upload_date': '20150204',
  71             'uploader': 'HouseCommittee',
  72             'uploader_id': '12987475',
  73         },
  74     }, {
  75         # Audio Only
  76         'url': 'https://www.c-span.org/video/?437336-1/judiciary-antitrust-competition-policy-consumer-rights',
  77         'only_matching': True,
  78     }]
  79     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
  80
  81     def _real_extract(self, url):
  82         video_id = self._match_id(url)
  83         video_type = None
  84         webpage = self._download_webpage(url, video_id)
  85
  86         ustream_url = UstreamIE._extract_url(webpage)
  87         if ustream_url:
  88             return self.url_result(ustream_url, UstreamIE.ie_key())
  89
  90         if '&vod' not in url:
  91             bc = self._search_regex(
  92                 r"(<[^>]+id='brightcove-player-embed'[^>]+>)",
  93                 webpage, 'brightcove embed', default=None)
  94             if bc:
  95                 bc_attr = extract_attributes(bc)
  96                 bc_url = self.BRIGHTCOVE_URL_TEMPLATE % (
  97                     bc_attr.get('data-bcaccountid', '3162030207001'),
  98                     bc_attr.get('data-noprebcplayerid', 'SyGGpuJy3g'),
  99                     bc_attr.get('data-newbcplayerid', 'default'),
 100                     bc_attr['data-bcid'])
 101                 return self.url_result(smuggle_url(bc_url, {'source_url': url}))
 102
 103         def add_referer(formats):
 104             for f in formats:
 105                 f.setdefault('http_headers', {})['Referer'] = url
 106
 107         # As of 01.12.2020 this path looks to cover all cases making the rest
 108         # of the code unnecessary
 109         jwsetup = self._parse_json(
 110             self._search_regex(
 111                 r'(?s)jwsetup\s*=\s*({.+?})\s*;', webpage, 'jwsetup',
 112                 default='{}'),
 113             video_id, transform_source=js_to_json, fatal=False)
 114         if jwsetup:
 115             info = self._parse_jwplayer_data(
 116                 jwsetup, video_id, require_title=False, m3u8_id='hls',
 117                 base_url=url)
 118             add_referer(info['formats'])
 119             ld_info = self._search_json_ld(webpage, video_id, default={})
 120             return merge_dicts(info, ld_info)
 121
 122         # Obsolete
 123         # We first look for clipid, because clipprog always appears before
 124         patterns = [r'id=\'clip(%s)\'\s*value=\'([0-9]+)\'' % t for t in ('id', 'prog')]
 125         results = list(filter(None, (re.search(p, webpage) for p in patterns)))
 126         if results:
 127             matches = results[0]
 128             video_type, video_id = matches.groups()
 129             video_type = 'clip' if video_type == 'id' else 'program'
 130         else:
 131             m = re.search(r'data-(?P<type>clip|prog)id=["\'](?P<id>\d+)', webpage)
 132             if m:
 133                 video_id = m.group('id')
 134                 video_type = 'program' if m.group('type') == 'prog' else 'clip'
 135             else:
 136                 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
 137                 if senate_isvp_url:
 138                     title = self._og_search_title(webpage)
 139                     surl = smuggle_url(senate_isvp_url, {'force_title': title})
 140                     return self.url_result(surl, 'SenateISVP', video_id, title)
 141                 video_id = self._search_regex(
 142                     r'jwsetup\.clipprog\s*=\s*(\d+);',
 143                     webpage, 'jwsetup program id', default=None)
 144                 if video_id:
 145                     video_type = 'program'
 146         if video_type is None or video_id is None:
 147             error_message = get_element_by_class('VLplayer-error-message', webpage)
 148             if error_message:
 149                 raise ExtractorError(error_message)
 150             raise ExtractorError('unable to find video id and type')
 151
 152         def get_text_attr(d, attr):
 153             return d.get(attr, {}).get('#text')
 154
 155         data = self._download_json(
 156             'http://www.c-span.org/assets/player/ajax-player.php?os=android&html5=%s&id=%s' % (video_type, video_id),
 157             video_id)['video']
 158         if data['@status'] != 'Success':
 159             raise ExtractorError('%s said: %s' % (self.IE_NAME, get_text_attr(data, 'error')), expected=True)
 160
 161         doc = self._download_xml(
 162             'http://www.c-span.org/common/services/flashXml.php?%sid=%s' % (video_type, video_id),
 163             video_id)
 164
 165         description = self._html_search_meta('description', webpage)
 166
 167         title = find_xpath_attr(doc, './/string', 'name', 'title').text
 168         thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text
 169
 170         files = data['files']
 171         capfile = get_text_attr(data, 'capfile')
 172
 173         entries = []
 174         for partnum, f in enumerate(files):
 175             formats = []
 176             for quality in f.get('qualities', []):
 177                 formats.append({
 178                     'format_id': '%s-%sp' % (get_text_attr(quality, 'bitrate'), get_text_attr(quality, 'height')),
 179                     'url': unescapeHTML(get_text_attr(quality, 'file')),
 180                     'height': int_or_none(get_text_attr(quality, 'height')),
 181                     'tbr': int_or_none(get_text_attr(quality, 'bitrate')),
 182                 })
 183             if not formats:
 184                 path = unescapeHTML(get_text_attr(f, 'path'))
 185                 if not path:
 186                     continue
 187                 formats = self._extract_m3u8_formats(
 188                     path, video_id, 'mp4', entry_protocol='m3u8_native',
 189                     m3u8_id='hls') if determine_ext(path) == 'm3u8' else [{'url': path, }]
 190             add_referer(formats)
 191             self._sort_formats(formats)
 192             entries.append({
 193                 'id': '%s_%d' % (video_id, partnum + 1),
 194                 'title': (
 195                     title if len(files) == 1 else
 196                     '%s part %d' % (title, partnum + 1)),
 197                 'formats': formats,
 198                 'description': description,
 199                 'thumbnail': thumbnail,
 200                 'duration': int_or_none(get_text_attr(f, 'length')),
 201                 'subtitles': {
 202                     'en': [{
 203                         'url': capfile,
 204                         'ext': determine_ext(capfile, 'dfxp')
 205                     }],
 206                 } if capfile else None,
 207             })
 208
 209         if len(entries) == 1:
 210             entry = dict(entries[0])
 211             entry['id'] = 'c' + video_id if video_type == 'clip' else video_id
 212             return entry
 213         else:
 214             return {
 215                 '_type': 'playlist',
 216                 'entries': entries,
 217                 'title': title,
 218                 'id': 'c' + video_id if video_type == 'clip' else video_id,
 219             }