logo

youtube-dl

[mirror] Download/Watch videos from video hostersgit clone https://hacktivis.me/git/mirror/youtube-dl.git

itv.py (16222B)

    

  1. # coding: utf-8
  2. from __future__ import unicode_literals
  4. import json
  5. import re
  6. import sys
  8. from .common import InfoExtractor
  9. from .brightcove import BrightcoveNewIE
  10. from ..compat import (
  11.     compat_HTTPError,
  12.     compat_integer_types,
  13.     compat_kwargs,
  14.     compat_urlparse,
  15. )
  16. from ..utils import (
  17.     clean_html,
  18.     determine_ext,
  19.     error_to_compat_str,
  20.     extract_attributes,
  21.     ExtractorError,
  22.     get_element_by_attribute,
  23.     int_or_none,
  24.     merge_dicts,
  25.     parse_duration,
  26.     parse_iso8601,
  27.     remove_start,
  28.     smuggle_url,
  29.     strip_or_none,
  30.     traverse_obj,
  31.     url_or_none,
  32.     urljoin,
  33. )
  36. class ITVBaseIE(InfoExtractor):
  38.     def _search_nextjs_data(self, webpage, video_id, **kw):
  39.         transform_source = kw.pop('transform_source', None)
  40.         fatal = kw.pop('fatal', True)
  41.         return self._parse_json(
  42.             self._search_regex(
  43.                 r'''<script\b[^>]+\bid=('|")__NEXT_DATA__\1[^>]*>(?P<js>[^<]+)</script>''',
  44.                 webpage, 'next.js data', group='js', fatal=fatal, **kw),
  45.             video_id, transform_source=transform_source, fatal=fatal)
  47.     def __handle_request_webpage_error(self, err, video_id=None, errnote=None, fatal=True):
  48.         if errnote is False:
  49.             return False
  50.         if errnote is None:
  51.             errnote = 'Unable to download webpage'
  53.         errmsg = '%s: %s' % (errnote, error_to_compat_str(err))
  54.         if fatal:
  55.             raise ExtractorError(errmsg, sys.exc_info()[2], cause=err, video_id=video_id)
  56.         else:
  57.             self._downloader.report_warning(errmsg)
  58.             return False
  60.     @staticmethod
  61.     def _vanilla_ua_header():
  62.         return {'User-Agent': 'Mozilla/5.0'}
  64.     def _download_webpage_handle(self, url, video_id, *args, **kwargs):
  65.         # specialised to (a) use vanilla UA (b) detect geo-block
  66.         params = self._downloader.params
  67.         nkwargs = {}
  68.         if (
  69.                 'user_agent' not in params
  70.                 and not any(re.match(r'(?i)user-agent\s*:', h)
  71.                             for h in (params.get('headers') or []))
  72.                 and 'User-Agent' not in (kwargs.get('headers') or {})):
  74.             kwargs.setdefault('headers', {})
  75.             kwargs['headers'] = self._vanilla_ua_header()
  76.             nkwargs = kwargs
  77.         if kwargs.get('expected_status') is not None:
  78.             exp = kwargs['expected_status']
  79.             if isinstance(exp, compat_integer_types):
  80.                 exp = [exp]
  81.             if isinstance(exp, (list, tuple)) and 403 not in exp:
  82.                 kwargs['expected_status'] = [403]
  83.                 kwargs['expected_status'].extend(exp)
  84.                 nkwargs = kwargs
  85.         else:
  86.             kwargs['expected_status'] = 403
  87.             nkwargs = kwargs
  89.         if nkwargs:
  90.             kwargs = compat_kwargs(kwargs)
  92.         ret = super(ITVBaseIE, self)._download_webpage_handle(url, video_id, *args, **kwargs)
  93.         if ret is False:
  94.             return ret
  95.         webpage, urlh = ret
  97.         if urlh.getcode() == 403:
  98.             # geo-block error is like this, with an unnecessary 'Of':
  99.             # '{\n  "Message" : "Request Originated Outside Of Allowed Geographic Region",\
  100.             # \n  "TransactionId" : "oas-magni-475082-xbYF0W"\n}'
  101.             if '"Request Originated Outside Of Allowed Geographic Region"' in webpage:
  102.                 self.raise_geo_restricted(countries=['GB'])
  103.             ret = self.__handle_request_webpage_error(
  104.                 compat_HTTPError(urlh.geturl(), 403, 'HTTP Error 403: Forbidden', urlh.headers, urlh),
  105.                 fatal=kwargs.get('fatal'))
  107.         return ret
  110. class ITVIE(ITVBaseIE):
  111.     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:(?P<w>watch)|hub)/[^/]+/(?(w)[\w-]+/)(?P<id>\w+)'
  112.     _IE_DESC = 'ITVX'
  113.     _TESTS = [{
  114.         'note': 'Hub URLs redirect to ITVX',
  115.         'url': 'https://www.itv.com/hub/liar/2a4547a0012',
  116.         'only_matching': True,
  117.     }, {
  118.         'note': 'Hub page unavailable via data-playlist-url (404 now)',
  119.         'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
  120.         'only_matching': True,
  121.     }, {
  122.         'note': 'Hub page with InvalidVodcrid (404 now)',
  123.         'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
  124.         'only_matching': True,
  125.     }, {
  126.         'note': 'Hub page with ContentUnavailable (404 now)',
  127.         'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
  128.         'only_matching': True,
  129.     }, {
  130.         'note': 'ITVX, or itvX, show',
  131.         'url': 'https://www.itv.com/watch/vera/1a7314/1a7314a0014',
  132.         'md5': 'bd0ad666b2c058fffe7d036785880064',
  133.         'info_dict': {
  134.             'id': '1a7314a0014',
  135.             'ext': 'mp4',
  136.             'title': 'Vera - Series 3 - Episode 4 - Prodigal Son',
  137.             'description': 'Vera and her team investigate the fatal stabbing of an ex-Met police officer outside a busy Newcastle nightclub - but there aren\'t many clues.',
  138.             'timestamp': 1653591600,
  139.             'upload_date': '20220526',
  140.             'uploader': 'ITVX',
  141.             'thumbnail': r're:https://\w+\.itv\.com/images/(?:\w+/)+\d+x\d+\?',
  142.             'duration': 5340.8,
  143.             'age_limit': 16,
  144.             'series': 'Vera',
  145.             'series_number': 3,
  146.             'episode': 'Prodigal Son',
  147.             'episode_number': 4,
  148.             'channel': 'ITV3',
  149.             'categories': list,
  150.         },
  151.         'params': {
  152.             # m3u8 download
  153.             # 'skip_download': True,
  154.         },
  155.         'skip': 'only available in UK',
  156.     }, {
  157.         'note': 'Latest ITV news bulletin: details change daily',
  158.         'url': 'https://www.itv.com/watch/news/varies-but-is-not-checked/6js5d0f',
  159.         'info_dict': {
  160.             'id': '6js5d0f',
  161.             'ext': 'mp4',
  162.             'title': r're:The latest ITV News headlines - \S.+',
  163.             'description': r'''re:.* today's top stories from the ITV News team.$''',
  164.             'timestamp': int,
  165.             'upload_date': r're:2\d\d\d(?:0[1-9]|1[0-2])(?:[012][1-9]|3[01])',
  166.             'uploader': 'ITVX',
  167.             'thumbnail': r're:https://images\.ctfassets\.net/(?:\w+/)+[\w.]+\.(?:jpg|png)',
  168.             'duration': float,
  169.             'age_limit': None,
  170.         },
  171.         'params': {
  172.             # variable download
  173.             # 'skip_download': True,
  174.         },
  175.         'skip': 'only available in UK',
  176.     }
  177.     ]
  179.     def _og_extract(self, webpage, require_title=False):
  180.         return {
  181.             'title': self._og_search_title(webpage, fatal=require_title),
  182.             'description': self._og_search_description(webpage, default=None),
  183.             'thumbnail': self._og_search_thumbnail(webpage, default=None),
  184.             'uploader': self._og_search_property('site_name', webpage, default=None),
  185.         }
  187.     def _real_extract(self, url):
  188.         video_id = self._match_id(url)
  190.         webpage = self._download_webpage(url, video_id)
  192.         # now quite different params!
  193.         params = extract_attributes(self._search_regex(
  194.             r'''(<[^>]+\b(?:class|data-testid)\s*=\s*("|')genie-container\2[^>]*>)''',
  195.             webpage, 'params'))
  197.         ios_playlist_url = traverse_obj(
  198.             params, 'data-video-id', 'data-video-playlist',
  199.             get_all=False, expected_type=url_or_none)
  201.         headers = self.geo_verification_headers()
  202.         headers.update({
  203.             'Accept': 'application/vnd.itv.vod.playlist.v2+json',
  204.             'Content-Type': 'application/json',
  205.         })
  206.         ios_playlist = self._download_json(
  207.             ios_playlist_url, video_id, data=json.dumps({
  208.                 'user': {
  209.                     'entitlements': [],
  210.                 },
  211.                 'device': {
  212.                     'manufacturer': 'Mobile Safari',
  213.                     'model': '5.1',
  214.                     'os': {
  215.                         'name': 'iOS',
  216.                         'version': '5.0',
  217.                         'type': ' mobile'
  218.                     }
  219.                 },
  220.                 'client': {
  221.                     'version': '4.1',
  222.                     'id': 'browser',
  223.                     'supportsAdPods': True,
  224.                     'service': 'itv.x',
  225.                     'appversion': '2.43.28',
  226.                 },
  227.                 'variantAvailability': {
  228.                     'player': 'hls',
  229.                     'featureset': {
  230.                         'min': ['hls', 'aes', 'outband-webvtt'],
  231.                         'max': ['hls', 'aes', 'outband-webvtt']
  232.                     },
  233.                     'platformTag': 'mobile'
  234.                 }
  235.             }).encode(), headers=headers)
  236.         video_data = ios_playlist['Playlist']['Video']
  237.         ios_base_url = traverse_obj(video_data, 'Base', expected_type=url_or_none)
  239.         media_url = (
  240.             (lambda u: url_or_none(urljoin(ios_base_url, u)))
  241.             if ios_base_url else url_or_none)
  243.         formats = []
  244.         for media_file in traverse_obj(video_data, 'MediaFiles', expected_type=list) or []:
  245.             href = traverse_obj(media_file, 'Href', expected_type=media_url)
  246.             if not href:
  247.                 continue
  248.             ext = determine_ext(href)
  249.             if ext == 'm3u8':
  250.                 formats.extend(self._extract_m3u8_formats(
  251.                     href, video_id, 'mp4', entry_protocol='m3u8',
  252.                     m3u8_id='hls', fatal=False))
  254.             else:
  255.                 formats.append({
  256.                     'url': href,
  257.                 })
  258.         self._sort_formats(formats)
  259.         for f in formats:
  260.             f.setdefault('http_headers', {})
  261.             f['http_headers'].update(self._vanilla_ua_header())
  263.         subtitles = {}
  264.         for sub in traverse_obj(video_data, 'Subtitles', expected_type=list) or []:
  265.             href = traverse_obj(sub, 'Href', expected_type=url_or_none)
  266.             if not href:
  267.                 continue
  268.             subtitles.setdefault('en', []).append({
  269.                 'url': href,
  270.                 'ext': determine_ext(href, 'vtt'),
  271.             })
  273.         next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default='{}')
  274.         video_data.update(traverse_obj(next_data, ('props', 'pageProps', ('title', 'episode')), expected_type=dict)[0] or {})
  275.         title = traverse_obj(video_data, 'headerTitle', 'episodeTitle')
  276.         info = self._og_extract(webpage, require_title=not title)
  277.         tn = info.pop('thumbnail', None)
  278.         if tn:
  279.             info['thumbnails'] = [{'url': tn}]
  281.         # num. episode title
  282.         num_ep_title = video_data.get('numberedEpisodeTitle')
  283.         if not num_ep_title:
  284.             num_ep_title = clean_html(get_element_by_attribute('data-testid', 'episode-hero-description-strong', webpage))
  285.             num_ep_title = num_ep_title and num_ep_title.rstrip(' -')
  286.         ep_title = strip_or_none(
  287.             video_data.get('episodeTitle')
  288.             or (num_ep_title.split('.', 1)[-1] if num_ep_title else None))
  289.         title = title or re.sub(r'\s+-\s+ITVX$', '', info['title'])
  290.         if ep_title and ep_title != title:
  291.             title = title + ' - ' + ep_title
  293.         def get_thumbnails():
  294.             tns = []
  295.             for w, x in (traverse_obj(video_data, ('imagePresets'), expected_type=dict) or {}).items():
  296.                 if isinstance(x, dict):
  297.                     for y, z in x.items():
  298.                         tns.append({'id': w + '_' + y, 'url': z})
  299.             return tns or None
  301.         video_str = lambda *x: traverse_obj(
  302.             video_data, *x, get_all=False, expected_type=strip_or_none)
  304.         return merge_dicts({
  305.             'id': video_id,
  306.             'title': title,
  307.             'formats': formats,
  308.             'subtitles': subtitles,
  309.             # parsing hh:mm:ss:nnn not yet patched
  310.             'duration': parse_duration(re.sub(r'(\d{2})(:)(\d{3}$)', r'\1.\3', video_data.get('Duration') or '')),
  311.             'description': video_str('synopsis'),
  312.             'timestamp': traverse_obj(video_data, 'broadcastDateTime', 'dateTime', expected_type=parse_iso8601),
  313.             'thumbnails': get_thumbnails(),
  314.             'series': video_str('showTitle', 'programmeTitle'),
  315.             'series_number': int_or_none(video_data.get('seriesNumber')),
  316.             'episode': ep_title,
  317.             'episode_number': int_or_none((num_ep_title or '').split('.')[0]),
  318.             'channel': video_str('channel'),
  319.             'categories': traverse_obj(video_data, ('categories', 'formatted'), expected_type=list),
  320.             'age_limit': {False: 16, True: 0}.get(video_data.get('isChildrenCategory')),
  321.         }, info)
  324. class ITVBTCCIE(ITVBaseIE):
  325.     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?!(?:watch|hub)/)(?:[^/]+/)+(?P<id>[^/?#&]+)'
  326.     _IE_DESC = 'ITV articles: News, British Touring Car Championship'
  327.     _TESTS = [{
  328.         'note': 'British Touring Car Championship',
  329.         'url': 'https://www.itv.com/btcc/articles/btcc-2018-all-the-action-from-brands-hatch',
  330.         'info_dict': {
  331.             'id': 'btcc-2018-all-the-action-from-brands-hatch',
  332.             'title': 'BTCC 2018: All the action from Brands Hatch',
  333.         },
  334.         'playlist_mincount': 9,
  335.     }, {
  336.         'note': 'redirects to /btcc/articles/...',
  337.         'url': 'http://www.itv.com/btcc/races/btcc-2018-all-the-action-from-brands-hatch',
  338.         'only_matching': True,
  339.     }, {
  340.         'note': 'news article',
  341.         'url': 'https://www.itv.com/news/wales/2020-07-23/sean-fletcher-shows-off-wales-coastline-in-new-itv-series-as-british-tourists-opt-for-staycations',
  342.         'info_dict': {
  343.             'id': 'sean-fletcher-shows-off-wales-coastline-in-new-itv-series-as-british-tourists-opt-for-staycations',
  344.             'title': '''Sean Fletcher on why Wales' coastline should be your 'staycation' destination | ITV News''',
  345.         },
  346.         'playlist_mincount': 1,
  347.     }]
  349.     # should really be a class var of the BC IE
  350.     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'
  351.     BRIGHTCOVE_ACCOUNT = '1582188683001'
  352.     BRIGHTCOVE_PLAYER = 'HkiHLnNRx'
  354.     def _real_extract(self, url):
  355.         playlist_id = self._match_id(url)
  357.         webpage, urlh = self._download_webpage_handle(url, playlist_id)
  358.         link = compat_urlparse.urlparse(urlh.geturl()).path.strip('/')
  360.         next_data = self._search_nextjs_data(webpage, playlist_id, fatal=False, default='{}')
  361.         path_prefix = compat_urlparse.urlparse(next_data.get('assetPrefix') or '').path.strip('/')
  362.         link = remove_start(link, path_prefix).strip('/')
  364.         content = traverse_obj(
  365.             next_data, ('props', 'pageProps', Ellipsis),
  366.             expected_type=lambda x: x if x['link'] == link else None,
  367.             get_all=False, default={})
  368.         content = traverse_obj(
  369.             content, ('body', 'content', Ellipsis, 'data'),
  370.             expected_type=lambda x: x if x.get('name') == 'Brightcove' or x.get('type') == 'Brightcove' else None)
  372.         contraband = {
  373.             # ITV does not like some GB IP ranges, so here are some
  374.             # IP blocks it accepts
  375.             'geo_ip_blocks': [
  376.                 '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
  377.             ],
  378.             'referrer': urlh.geturl(),
  379.         }
  381.         def entries():
  383.             for data in content or []:
  384.                 video_id = data.get('id')
  385.                 if not video_id:
  386.                     continue
  387.                 account = data.get('accountId') or self.BRIGHTCOVE_ACCOUNT
  388.                 player = data.get('playerId') or self.BRIGHTCOVE_PLAYER
  389.                 yield self.url_result(
  390.                     smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % (account, player, video_id), contraband),
  391.                     ie=BrightcoveNewIE.ie_key(), video_id=video_id)
  393.             # obsolete ?
  394.             for video_id in re.findall(r'''data-video-id=["'](\d+)''', webpage):
  395.                 yield self.url_result(
  396.                     smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % (self.BRIGHTCOVE_ACCOUNT, self.BRIGHTCOVE_PLAYER, video_id), contraband),
  397.                     ie=BrightcoveNewIE.ie_key(), video_id=video_id)
  399.         title = self._og_search_title(webpage, fatal=False)
  401.         return self.playlist_result(entries(), playlist_id, title)