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