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