yt_dlp/extractor/cspan.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_HTMLParseError
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     extract_attributes,
  11     find_xpath_attr,
  12     get_element_by_attribute,
  13     get_element_by_class,
  14     int_or_none,
  15     join_nonempty,
  16     js_to_json,
  17     merge_dicts,
  18     parse_iso8601,
  19     parse_qs,
  20     smuggle_url,
  21     str_to_int,
  22     unescapeHTML,
  23 )
  24 from .senategov import SenateISVPIE
  25 from .ustream import UstreamIE
  26
  27
  28 class CSpanIE(InfoExtractor):
  29     _VALID_URL = r'https?://(?:www\.)?c-span\.org/video/\?(?P<id>[0-9a-f]+)'
  30     IE_DESC = 'C-SPAN'
  31     _TESTS = [{
  32         'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
  33         'md5': '94b29a4f131ff03d23471dd6f60b6a1d',
  34         'info_dict': {
  35             'id': '315139',
  36             'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
  37         },
  38         'playlist_mincount': 2,
  39         'skip': 'Regularly fails on travis, for unknown reasons',
  40     }, {
  41         'url': 'http://www.c-span.org/video/?c4486943/cspan-international-health-care-models',
  42         # md5 is unstable
  43         'info_dict': {
  44             'id': 'c4486943',
  45             'ext': 'mp4',
  46             'title': 'CSPAN - International Health Care Models',
  47             'description': 'md5:7a985a2d595dba00af3d9c9f0783c967',
  48         }
  49     }, {
  50         'url': 'http://www.c-span.org/video/?318608-1/gm-ignition-switch-recall',
  51         'info_dict': {
  52             'id': '342759',
  53             'title': 'General Motors Ignition Switch Recall',
  54         },
  55         'playlist_mincount': 6,
  56     }, {
  57         # Video from senate.gov
  58         'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
  59         'info_dict': {
  60             'id': 'judiciary031715',
  61             'ext': 'mp4',
  62             'title': 'Immigration Reforms Needed to Protect Skilled American Workers',
  63         },
  64         'params': {
  65             'skip_download': True,  # m3u8 downloads
  66         }
  67     }, {
  68         # Ustream embedded video
  69         'url': 'https://www.c-span.org/video/?114917-1/armed-services',
  70         'info_dict': {
  71             'id': '58428542',
  72             'ext': 'flv',
  73             'title': 'USHR07 Armed Services Committee',
  74             'description': 'hsas00-2118-20150204-1000et-07\n\n\nUSHR07 Armed Services Committee',
  75             'timestamp': 1423060374,
  76             'upload_date': '20150204',
  77             'uploader': 'HouseCommittee',
  78             'uploader_id': '12987475',
  79         },
  80     }, {
  81         # Audio Only
  82         'url': 'https://www.c-span.org/video/?437336-1/judiciary-antitrust-competition-policy-consumer-rights',
  83         'only_matching': True,
  84     }]
  85     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
  86
  87     def _real_extract(self, url):
  88         video_id = self._match_id(url)
  89         video_type = None
  90         webpage = self._download_webpage(url, video_id)
  91
  92         ustream_url = UstreamIE._extract_url(webpage)
  93         if ustream_url:
  94             return self.url_result(ustream_url, UstreamIE.ie_key())
  95
  96         if '&vod' not in url:
  97             bc = self._search_regex(
  98                 r"(<[^>]+id='brightcove-player-embed'[^>]+>)",
  99                 webpage, 'brightcove embed', default=None)
 100             if bc:
 101                 bc_attr = extract_attributes(bc)
 102                 bc_url = self.BRIGHTCOVE_URL_TEMPLATE % (
 103                     bc_attr.get('data-bcaccountid', '3162030207001'),
 104                     bc_attr.get('data-noprebcplayerid', 'SyGGpuJy3g'),
 105                     bc_attr.get('data-newbcplayerid', 'default'),
 106                     bc_attr['data-bcid'])
 107                 return self.url_result(smuggle_url(bc_url, {'source_url': url}))
 108
 109         def add_referer(formats):
 110             for f in formats:
 111                 f.setdefault('http_headers', {})['Referer'] = url
 112
 113         # As of 01.12.2020 this path looks to cover all cases making the rest
 114         # of the code unnecessary
 115         jwsetup = self._parse_json(
 116             self._search_regex(
 117                 r'(?s)jwsetup\s*=\s*({.+?})\s*;', webpage, 'jwsetup',
 118                 default='{}'),
 119             video_id, transform_source=js_to_json, fatal=False)
 120         if jwsetup:
 121             info = self._parse_jwplayer_data(
 122                 jwsetup, video_id, require_title=False, m3u8_id='hls',
 123                 base_url=url)
 124             add_referer(info['formats'])
 125             for subtitles in info['subtitles'].values():
 126                 for subtitle in subtitles:
 127                     ext = determine_ext(subtitle['url'])
 128                     if ext == 'php':
 129                         ext = 'vtt'
 130                     subtitle['ext'] = ext
 131             ld_info = self._search_json_ld(webpage, video_id, default={})
 132             try:
 133                 title = get_element_by_class('video-page-title', webpage)
 134             except compat_HTMLParseError:
 135                 title = None
 136             if title is None:
 137                 title = self._og_search_title(webpage)
 138             description = get_element_by_attribute('itemprop', 'description', webpage) or \
 139                 self._html_search_meta(['og:description', 'description'], webpage)
 140             return merge_dicts(info, ld_info, {
 141                 'title': title,
 142                 'thumbnail': get_element_by_attribute('itemprop', 'thumbnailUrl', webpage),
 143                 'description': description,
 144                 'timestamp': parse_iso8601(get_element_by_attribute('itemprop', 'uploadDate', webpage)),
 145                 'location': get_element_by_attribute('itemprop', 'contentLocation', webpage),
 146                 'duration': int_or_none(self._search_regex(
 147                     r'jwsetup\.seclength\s*=\s*(\d+);',
 148                     webpage, 'duration', fatal=False)),
 149                 'view_count': str_to_int(self._search_regex(
 150                     r"<span[^>]+class='views'[^>]*>([\d,]+)\s+Views</span>",
 151                     webpage, 'views', fatal=False)),
 152             })
 153
 154         # Obsolete
 155         # We first look for clipid, because clipprog always appears before
 156         patterns = [r'id=\'clip(%s)\'\s*value=\'([0-9]+)\'' % t for t in ('id', 'prog')]
 157         results = list(filter(None, (re.search(p, webpage) for p in patterns)))
 158         if results:
 159             matches = results[0]
 160             video_type, video_id = matches.groups()
 161             video_type = 'clip' if video_type == 'id' else 'program'
 162         else:
 163             m = re.search(r'data-(?P<type>clip|prog)id=["\'](?P<id>\d+)', webpage)
 164             if m:
 165                 video_id = m.group('id')
 166                 video_type = 'program' if m.group('type') == 'prog' else 'clip'
 167             else:
 168                 senate_isvp_url = SenateISVPIE._search_iframe_url(webpage)
 169                 if senate_isvp_url:
 170                     title = self._og_search_title(webpage)
 171                     surl = smuggle_url(senate_isvp_url, {'force_title': title})
 172                     return self.url_result(surl, 'SenateISVP', video_id, title)
 173                 video_id = self._search_regex(
 174                     r'jwsetup\.clipprog\s*=\s*(\d+);',
 175                     webpage, 'jwsetup program id', default=None)
 176                 if video_id:
 177                     video_type = 'program'
 178         if video_type is None or video_id is None:
 179             error_message = get_element_by_class('VLplayer-error-message', webpage)
 180             if error_message:
 181                 raise ExtractorError(error_message)
 182             raise ExtractorError('unable to find video id and type')
 183
 184         def get_text_attr(d, attr):
 185             return d.get(attr, {}).get('#text')
 186
 187         data = self._download_json(
 188             'http://www.c-span.org/assets/player/ajax-player.php?os=android&html5=%s&id=%s' % (video_type, video_id),
 189             video_id)['video']
 190         if data['@status'] != 'Success':
 191             raise ExtractorError('%s said: %s' % (self.IE_NAME, get_text_attr(data, 'error')), expected=True)
 192
 193         doc = self._download_xml(
 194             'http://www.c-span.org/common/services/flashXml.php?%sid=%s' % (video_type, video_id),
 195             video_id)
 196
 197         description = self._html_search_meta('description', webpage)
 198
 199         title = find_xpath_attr(doc, './/string', 'name', 'title').text
 200         thumbnail = find_xpath_attr(doc, './/string', 'name', 'poster').text
 201
 202         files = data['files']
 203         capfile = get_text_attr(data, 'capfile')
 204
 205         entries = []
 206         for partnum, f in enumerate(files):
 207             formats = []
 208             for quality in f.get('qualities', []):
 209                 formats.append({
 210                     'format_id': '%s-%sp' % (get_text_attr(quality, 'bitrate'), get_text_attr(quality, 'height')),
 211                     'url': unescapeHTML(get_text_attr(quality, 'file')),
 212                     'height': int_or_none(get_text_attr(quality, 'height')),
 213                     'tbr': int_or_none(get_text_attr(quality, 'bitrate')),
 214                 })
 215             if not formats:
 216                 path = unescapeHTML(get_text_attr(f, 'path'))
 217                 if not path:
 218                     continue
 219                 formats = self._extract_m3u8_formats(
 220                     path, video_id, 'mp4', entry_protocol='m3u8_native',
 221                     m3u8_id='hls') if determine_ext(path) == 'm3u8' else [{'url': path, }]
 222             add_referer(formats)
 223             self._sort_formats(formats)
 224             entries.append({
 225                 'id': '%s_%d' % (video_id, partnum + 1),
 226                 'title': (
 227                     title if len(files) == 1 else
 228                     '%s part %d' % (title, partnum + 1)),
 229                 'formats': formats,
 230                 'description': description,
 231                 'thumbnail': thumbnail,
 232                 'duration': int_or_none(get_text_attr(f, 'length')),
 233                 'subtitles': {
 234                     'en': [{
 235                         'url': capfile,
 236                         'ext': determine_ext(capfile, 'dfxp')
 237                     }],
 238                 } if capfile else None,
 239             })
 240
 241         if len(entries) == 1:
 242             entry = dict(entries[0])
 243             entry['id'] = 'c' + video_id if video_type == 'clip' else video_id
 244             return entry
 245         else:
 246             return {
 247                 '_type': 'playlist',
 248                 'entries': entries,
 249                 'title': title,
 250                 'id': 'c' + video_id if video_type == 'clip' else video_id,
 251             }
 252
 253
 254 class CSpanCongressIE(InfoExtractor):
 255     _VALID_URL = r'https?://(?:www\.)?c-span\.org/congress/'
 256     _TESTS = [{
 257         'url': 'https://www.c-span.org/congress/?chamber=house&date=2017-12-13&t=1513208380',
 258         'info_dict': {
 259             'id': 'house_2017-12-13',
 260             'title': 'Congressional Chronicle - Members of Congress, Hearings and More',
 261             'description': 'md5:54c264b7a8f219937987610243305a84',
 262             'thumbnail': r're:https://ximage.c-spanvideo.org/.+',
 263             'ext': 'mp4'
 264         }
 265     }]
 266
 267     def _real_extract(self, url):
 268         query = parse_qs(url)
 269         video_date = query.get('date', [None])[0]
 270         video_id = join_nonempty(query.get('chamber', ['senate'])[0], video_date, delim='_')
 271         webpage = self._download_webpage(url, video_id)
 272         if not video_date:
 273             jwp_date = re.search(r'jwsetup.clipprogdate = \'(?P<date>\d{4}-\d{2}-\d{2})\';', webpage)
 274             if jwp_date:
 275                 video_id = f'{video_id}_{jwp_date.group("date")}'
 276         jwplayer_data = self._parse_json(
 277             self._search_regex(r'jwsetup\s*=\s*({(?:.|\n)[^;]+});', webpage, 'player config'),
 278             video_id, transform_source=js_to_json)
 279
 280         title = (self._og_search_title(webpage, default=None)
 281                  or self._html_search_regex(r'(?s)<title>(.*?)</title>', webpage, 'video title'))
 282         description = (self._og_search_description(webpage, default=None)
 283                        or self._html_search_meta('description', webpage, 'description', default=None))
 284
 285         return {
 286             **self._parse_jwplayer_data(jwplayer_data, video_id, False),
 287             'title': re.sub(r'\s+', ' ', title.split('|')[0]).strip(),
 288             'description': description,
 289             'http_headers': {'Referer': 'https://www.c-span.org/'},
 290         }