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_attribute,
  12     get_element_by_class,
  13     int_or_none,
  14     js_to_json,
  15     merge_dicts,
  16     parse_iso8601,
  17     smuggle_url,
  18     str_to_int,
  19     unescapeHTML,
  20 )
  21 from .senateisvp import SenateISVPIE
  22 from .ustream import UstreamIE
  23
  24
  25 class CSpanIE(InfoExtractor):
  26     _VALID_URL = r'https?://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
  27     IE_DESC = 'C-SPAN'
  28     _TESTS = [{
  29         'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
  30         'md5': '94b29a4f131ff03d23471dd6f60b6a1d',
  31         'info_dict': {
  32             'id': '315139',
  33             'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
  34         },
  35         'playlist_mincount': 2,
  36         'skip': 'Regularly fails on travis, for unknown reasons',
  37     }, {
  38         'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models',
  39         # md5 is unstable
  40         'info_dict': {
  41             'id': 'c4486943',
  42             'ext': 'mp4',
  43             'title': 'CSPAN - International Health Care Models',
  44             'description': 'md5:7a985a2d595dba00af3d9c9f0783c967',
  45         }
  46     }, {
  47         'url': 'http://www.c-span.org/video/?318608-1/gm-ignition-switch-recall',
  48         'info_dict': {
  49             'id': '342759',
  50             'title': 'General Motors Ignition Switch Recall',
  51         },
  52         'playlist_mincount': 6,
  53     }, {
  54         # Video from senate.gov
  55         'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
  56         'info_dict': {
  57             'id': 'judiciary031715',
  58             'ext': 'mp4',
  59             'title': 'Immigration Reforms Needed to Protect Skilled American Workers',
  60         },
  61         'params': {
  62             'skip_download': True,  # m3u8 downloads
  63         }
  64     }, {
  65         # Ustream embedded video
  66         'url': 'https://www.c-span.org/video/?114917-1/armed-services',
  67         'info_dict': {
  68             'id': '58428542',
  69             'ext': 'flv',
  70             'title': 'USHR07 Armed Services Committee',
  71             'description': 'hsas00-2118-20150204-1000et-07\n\n\nUSHR07 Armed Services Committee',
  72             'timestamp': 1423060374,
  73             'upload_date': '20150204',
  74             'uploader': 'HouseCommittee',
  75             'uploader_id': '12987475',
  76         },
  77     }, {
  78         # Audio Only
  79         'url': 'https://www.c-span.org/video/?437336-1/judiciary-antitrust-competition-policy-consumer-rights',
  80         'only_matching': True,
  81     }]
  82     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
  83
  84     def _real_extract(self, url):
  85         video_id = self._match_id(url)
  86         video_type = None
  87         webpage = self._download_webpage(url, video_id)
  88
  89         ustream_url = UstreamIE._extract_url(webpage)
  90         if ustream_url:
  91             return self.url_result(ustream_url, UstreamIE.ie_key())
  92
  93         if '&vod' not in url:
  94             bc = self._search_regex(
  95                 r"(<[^>]+id='brightcove-player-embed'[^>]+>)",
  96                 webpage, 'brightcove embed', default=None)
  97             if bc:
  98                 bc_attr = extract_attributes(bc)
  99                 bc_url = self.BRIGHTCOVE_URL_TEMPLATE % (
 100                     bc_attr.get('data-bcaccountid', '3162030207001'),
 101                     bc_attr.get('data-noprebcplayerid', 'SyGGpuJy3g'),
 102                     bc_attr.get('data-newbcplayerid', 'default'),
 103                     bc_attr['data-bcid'])
 104                 return self.url_result(smuggle_url(bc_url, {'source_url': url}))
 105
 106         def add_referer(formats):
 107             for f in formats:
 108                 f.setdefault('http_headers', {})['Referer'] = url
 109
 110         # As of 01.12.2020 this path looks to cover all cases making the rest
 111         # of the code unnecessary
 112         jwsetup = self._parse_json(
 113             self._search_regex(
 114                 r'(?s)jwsetup\s*=\s*({.+?})\s*;', webpage, 'jwsetup',
 115                 default='{}'),
 116             video_id, transform_source=js_to_json, fatal=False)
 117         if jwsetup:
 118             info = self._parse_jwplayer_data(
 119                 jwsetup, video_id, require_title=False, m3u8_id='hls',
 120                 base_url=url)
 121             add_referer(info['formats'])
 122             for subtitles in info['subtitles'].values():
 123                 for subtitle in subtitles:
 124                     ext = determine_ext(subtitle['url'])
 125                     if ext == 'php':
 126                         ext = 'vtt'
 127                     subtitle['ext'] = ext
 128             ld_info = self._search_json_ld(webpage, video_id, default={})
 129             title = get_element_by_class('video-page-title', webpage) or \
 130                 self._og_search_title(webpage)
 131             description = get_element_by_attribute('itemprop', 'description', webpage) or \
 132                 self._html_search_meta(['og:description', 'description'], webpage)
 133             return merge_dicts(info, ld_info, {
 134                 'title': title,
 135                 'thumbnail': get_element_by_attribute('itemprop', 'thumbnailUrl', webpage),
 136                 'description': description,
 137                 'timestamp': parse_iso8601(get_element_by_attribute('itemprop', 'uploadDate', webpage)),
 138                 'location': get_element_by_attribute('itemprop', 'contentLocation', webpage),
 139                 'duration': int_or_none(self._search_regex(
 140                     r'jwsetup\.seclength\s*=\s*(\d+);',
 141                     webpage, 'duration', fatal=False)),
 142                 'view_count': str_to_int(self._search_regex(
 143                     r"<span[^>]+class='views'[^>]*>([\d,]+)\s+Views</span>",
 144                     webpage, 'views', fatal=False)),
 145             })
 146
 147         # Obsolete
 148         # We first look for clipid, because clipprog always appears before
 149         patterns = [r'id=\'clip(%s)\'\s*value=\'([0-9]+)\'' % t for t in ('id', 'prog')]
 150         results = list(filter(None, (re.search(p, webpage) for p in patterns)))
 151         if results:
 152             matches = results[0]
 153             video_type, video_id = matches.groups()
 154             video_type = 'clip' if video_type == 'id' else 'program'
 155         else:
 156             m = re.search(r'data-(?P<type>clip|prog)id=["\'](?P<id>\d+)', webpage)
 157             if m:
 158                 video_id = m.group('id')
 159                 video_type = 'program' if m.group('type') == 'prog' else 'clip'
 160             else:
 161                 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
 162                 if senate_isvp_url:
 163                     title = self._og_search_title(webpage)
 164                     surl = smuggle_url(senate_isvp_url, {'force_title': title})
 165                     return self.url_result(surl, 'SenateISVP', video_id, title)
 166                 video_id = self._search_regex(
 167                     r'jwsetup\.clipprog\s*=\s*(\d+);',
 168                     webpage, 'jwsetup program id', default=None)
 169                 if video_id:
 170                     video_type = 'program'
 171         if video_type is None or video_id is None:
 172             error_message = get_element_by_class('VLplayer-error-message', webpage)
 173             if error_message:
 174                 raise ExtractorError(error_message)
 175             raise ExtractorError('unable to find video id and type')
 176
 177         def get_text_attr(d, attr):
 178             return d.get(attr, {}).get('#text')
 179
 180         data = self._download_json(
 181             'http://www.c-span.org/assets/player/ajax-player.php?os=android&html5=%s&id=%s' % (video_type, video_id),
 182             video_id)['video']
 183         if data['@status'] != 'Success':
 184             raise ExtractorError('%s said: %s' % (self.IE_NAME, get_text_attr(data, 'error')), expected=True)
 185
 186         doc = self._download_xml(
 187             'http://www.c-span.org/common/services/flashXml.php?%sid=%s' % (video_type, video_id),
 188             video_id)
 189
 190         description = self._html_search_meta('description', webpage)
 191
 192         title = find_xpath_attr(doc, './/string', 'name', 'title').text
 193         thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text
 194
 195         files = data['files']
 196         capfile = get_text_attr(data, 'capfile')
 197
 198         entries = []
 199         for partnum, f in enumerate(files):
 200             formats = []
 201             for quality in f.get('qualities', []):
 202                 formats.append({
 203                     'format_id': '%s-%sp' % (get_text_attr(quality, 'bitrate'), get_text_attr(quality, 'height')),
 204                     'url': unescapeHTML(get_text_attr(quality, 'file')),
 205                     'height': int_or_none(get_text_attr(quality, 'height')),
 206                     'tbr': int_or_none(get_text_attr(quality, 'bitrate')),
 207                 })
 208             if not formats:
 209                 path = unescapeHTML(get_text_attr(f, 'path'))
 210                 if not path:
 211                     continue
 212                 formats = self._extract_m3u8_formats(
 213                     path, video_id, 'mp4', entry_protocol='m3u8_native',
 214                     m3u8_id='hls') if determine_ext(path) == 'm3u8' else [{'url': path, }]
 215             add_referer(formats)
 216             self._sort_formats(formats)
 217             entries.append({
 218                 'id': '%s_%d' % (video_id, partnum + 1),
 219                 'title': (
 220                     title if len(files) == 1 else
 221                     '%s part %d' % (title, partnum + 1)),
 222                 'formats': formats,
 223                 'description': description,
 224                 'thumbnail': thumbnail,
 225                 'duration': int_or_none(get_text_attr(f, 'length')),
 226                 'subtitles': {
 227                     'en': [{
 228                         'url': capfile,
 229                         'ext': determine_ext(capfile, 'dfxp')
 230                     }],
 231                 } if capfile else None,
 232             })
 233
 234         if len(entries) == 1:
 235             entry = dict(entries[0])
 236             entry['id'] = 'c' + video_id if video_type == 'clip' else video_id
 237             return entry
 238         else:
 239             return {
 240                 '_type': 'playlist',
 241                 'entries': entries,
 242                 'title': title,
 243                 'id': 'c' + video_id if video_type == 'clip' else video_id,
 244             }