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
  16
  17 class CSpanIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
  19     IE_DESC = 'C-SPAN'
  20     _TESTS = [{
  21         'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
  22         'md5': '94b29a4f131ff03d23471dd6f60b6a1d',
  23         'info_dict': {
  24             'id': '315139',
  25             'ext': 'mp4',
  26             'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
  27             'description': 'Attorney General Eric Holder speaks to reporters following the Supreme Court decision in [Shelby County v. Holder], in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced.',
  28         },
  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': '8e5fbfabe6ad0f89f3012a7943c1287b',
  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         'md5': '2ae5051559169baadba13fc35345ae74',
  42         'info_dict': {
  43             'id': '342759',
  44             'ext': 'mp4',
  45             'title': 'General Motors Ignition Switch Recall',
  46             'duration': 14848,
  47             'description': 'md5:118081aedd24bf1d3b68b3803344e7f3'
  48         },
  49     }, {
  50         # Video from senate.gov
  51         'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
  52         'info_dict': {
  53             'id': 'judiciary031715',
  54             'ext': 'mp4',
  55             'title': 'Immigration Reforms Needed to Protect Skilled American Workers',
  56         },
  57         'params': {
  58             'skip_download': True,  # m3u8 downloads
  59         }
  60     }]
  61
  62     def _real_extract(self, url):
  63         video_id = self._match_id(url)
  64         video_type = None
  65         webpage = self._download_webpage(url, video_id)
  66         # We first look for clipid, because clipprog always appears before
  67         patterns = [r'id=\'clip(%s)\'\s*value=\'([0-9]+)\'' % t for t in ('id', 'prog')]
  68         results = list(filter(None, (re.search(p, webpage) for p in patterns)))
  69         if results:
  70             matches = results[0]
  71             video_type, video_id = matches.groups()
  72             video_type = 'clip' if video_type == 'id' else 'program'
  73         else:
  74             m = re.search(r'data-(?P<type>clip|prog)id=["\'](?P<id>\d+)', webpage)
  75             if m:
  76                 video_id = m.group('id')
  77                 video_type = 'program' if m.group('type') == 'prog' else 'clip'
  78             else:
  79                 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
  80                 if senate_isvp_url:
  81                     title = self._og_search_title(webpage)
  82                     surl = smuggle_url(senate_isvp_url, {'force_title': title})
  83                     return self.url_result(surl, 'SenateISVP', video_id, title)
  84         if video_type is None or video_id is None:
  85             raise ExtractorError('unable to find video id and type')
  86
  87         def get_text_attr(d, attr):
  88             return d.get(attr, {}).get('#text')
  89
  90         data = self._download_json(
  91             'http://www.c-span.org/assets/player/ajax-player.php?os=android&html5=%s&id=%s' % (video_type, video_id),
  92             video_id)['video']
  93         if data['@status'] != 'Success':
  94             raise ExtractorError('%s said: %s' % (self.IE_NAME, get_text_attr(data, 'error')), expected=True)
  95
  96         doc = self._download_xml(
  97             'http://www.c-span.org/common/services/flashXml.php?%sid=%s' % (video_type, video_id),
  98             video_id)
  99
 100         description = self._html_search_meta('description', webpage)
 101
 102         title = find_xpath_attr(doc, './/string', 'name', 'title').text
 103         thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text
 104
 105         files = data['files']
 106         capfile = get_text_attr(data, 'capfile')
 107
 108         entries = []
 109         for partnum, f in enumerate(files):
 110             formats = []
 111             for quality in f['qualities']:
 112                 formats.append({
 113                     'format_id': '%s-%sp' % (get_text_attr(quality, 'bitrate'), get_text_attr(quality, 'height')),
 114                     'url': unescapeHTML(get_text_attr(quality, 'file')),
 115                     'height': int_or_none(get_text_attr(quality, 'height')),
 116                     'tbr': int_or_none(get_text_attr(quality, 'bitrate')),
 117                 })
 118             if not formats:
 119                 path = unescapeHTML(get_text_attr(f, 'path'))
 120                 if not path:
 121                     continue
 122                 formats = self._extract_m3u8_formats(
 123                     path, video_id, 'mp4', entry_protocol='m3u8_native',
 124                     m3u8_id='hls') if determine_ext(path) == 'm3u8' else [{'url': path, }]
 125             self._sort_formats(formats)
 126             entries.append({
 127                 'id': '%s_%d' % (video_id, partnum + 1),
 128                 'title': (
 129                     title if len(files) == 1 else
 130                     '%s part %d' % (title, partnum + 1)),
 131                 'formats': formats,
 132                 'description': description,
 133                 'thumbnail': thumbnail,
 134                 'duration': int_or_none(get_text_attr(f, 'length')),
 135                 'subtitles': {
 136                     'en': [{
 137                         'url': capfile,
 138                         'ext': determine_ext(capfile, 'dfxp')
 139                     }],
 140                 } if capfile else None,
 141             })
 142
 143         if len(entries) == 1:
 144             entry = dict(entries[0])
 145             entry['id'] = 'c' + video_id if video_type == 'clip' else video_id
 146             return entry
 147         else:
 148             return {
 149                 '_type': 'playlist',
 150                 'entries': entries,
 151                 'title': title,
 152                 'id': 'c' + video_id if video_type == 'clip' else video_id,
 153             }