X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/4e090bc3ceacc4e3cd464d12ea97700e3acad37d..a57dac80d518b579d7a7da90aa74ce401f138d0f:/youtube_dl/extractor/itv.py diff --git a/youtube_dl/extractor/itv.py b/youtube_dl/extractor/itv.py index 021c6b2..ad2f4ec 100644 --- a/youtube_dl/extractor/itv.py +++ b/youtube_dl/extractor/itv.py @@ -4,28 +4,33 @@ from __future__ import unicode_literals import uuid import xml.etree.ElementTree as etree import json +import re from .common import InfoExtractor +from .brightcove import BrightcoveNewIE from ..compat import ( compat_str, compat_etree_register_namespace, ) from ..utils import ( + determine_ext, + ExtractorError, extract_attributes, + int_or_none, + merge_dicts, + parse_duration, + smuggle_url, + url_or_none, xpath_with_ns, xpath_element, xpath_text, - int_or_none, - parse_duration, - ExtractorError, - determine_ext, ) class ITVIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P[0-9a-zA-Z]+)' _GEO_COUNTRIES = ['GB'] - _TEST = { + _TESTS = [{ 'url': 'http://www.itv.com/hub/mr-bean-animated-series/2a2936a0053', 'info_dict': { 'id': '2a2936a0053', @@ -36,7 +41,19 @@ class ITVIE(InfoExtractor): # rtmp download 'skip_download': True, }, - } + }, { + # unavailable via data-playlist-url + 'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033', + 'only_matching': True, + }, { + # InvalidVodcrid + 'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034', + 'only_matching': True, + }, { + # ContentUnavailable + 'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -59,12 +76,18 @@ class ITVIE(InfoExtractor): def _add_sub_element(element, name): return etree.SubElement(element, _add_ns(name)) + production_id = ( + params.get('data-video-autoplay-id') + or '%s#001' % ( + params.get('data-video-episode-id') + or video_id.replace('a', '/'))) + req_env = etree.Element(_add_ns('soapenv:Envelope')) _add_sub_element(req_env, 'soapenv:Header') body = _add_sub_element(req_env, 'soapenv:Body') get_playlist = _add_sub_element(body, ('tem:GetPlaylist')) request = _add_sub_element(get_playlist, 'tem:request') - _add_sub_element(request, 'itv:ProductionId').text = params['data-video-id'] + _add_sub_element(request, 'itv:ProductionId').text = production_id _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper() vodcrid = _add_sub_element(request, 'itv:Vodcrid') _add_sub_element(vodcrid, 'com:Id') @@ -94,39 +117,83 @@ class ITVIE(InfoExtractor): 'Content-Type': 'text/xml; charset=utf-8', 'SOAPAction': 'http://tempuri.org/PlaylistService/GetPlaylist', }) - resp_env = self._download_xml( - params['data-playlist-url'], video_id, - headers=headers, data=etree.tostring(req_env)) - playlist = xpath_element(resp_env, './/Playlist') - if playlist is None: - fault_code = xpath_text(resp_env, './/faultcode') - fault_string = xpath_text(resp_env, './/faultstring') - if fault_code == 'InvalidGeoRegion': - self.raise_geo_restricted( - msg=fault_string, countries=self._GEO_COUNTRIES) - raise ExtractorError('%s said: %s' % (self.IE_NAME, fault_string)) - title = xpath_text(playlist, 'EpisodeTitle', fatal=True) - video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True) - media_files = xpath_element(video_element, 'MediaFiles', fatal=True) - rtmp_url = media_files.attrib['base'] + info = self._search_json_ld(webpage, video_id, default={}) formats = [] - for media_file in media_files.findall('MediaFile'): - play_path = xpath_text(media_file, 'URL') - if not play_path: - continue - tbr = int_or_none(media_file.get('bitrate'), 1000) - formats.append({ - 'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''), - 'url': rtmp_url, - 'play_path': play_path, - 'tbr': tbr, - 'ext': 'flv', + subtitles = {} + + def extract_subtitle(sub_url): + ext = determine_ext(sub_url, 'ttml') + subtitles.setdefault('en', []).append({ + 'url': sub_url, + 'ext': 'ttml' if ext == 'xml' else ext, }) - ios_playlist_url = params.get('data-video-playlist') + resp_env = self._download_xml( + params['data-playlist-url'], video_id, + headers=headers, data=etree.tostring(req_env), fatal=False) + if resp_env: + playlist = xpath_element(resp_env, './/Playlist') + if playlist is None: + fault_code = xpath_text(resp_env, './/faultcode') + fault_string = xpath_text(resp_env, './/faultstring') + if fault_code == 'InvalidGeoRegion': + self.raise_geo_restricted( + msg=fault_string, countries=self._GEO_COUNTRIES) + elif fault_code not in ( + 'InvalidEntity', 'InvalidVodcrid', 'ContentUnavailable'): + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, fault_string), expected=True) + info.update({ + 'title': self._og_search_title(webpage), + 'episode_title': params.get('data-video-episode'), + 'series': params.get('data-video-title'), + }) + else: + title = xpath_text(playlist, 'EpisodeTitle', default=None) + info.update({ + 'title': title, + 'episode_title': title, + 'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')), + 'series': xpath_text(playlist, 'ProgrammeTitle'), + 'duration': parse_duration(xpath_text(playlist, 'Duration')), + }) + video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True) + media_files = xpath_element(video_element, 'MediaFiles', fatal=True) + rtmp_url = media_files.attrib['base'] + + for media_file in media_files.findall('MediaFile'): + play_path = xpath_text(media_file, 'URL') + if not play_path: + continue + tbr = int_or_none(media_file.get('bitrate'), 1000) + f = { + 'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''), + 'play_path': play_path, + # Providing this swfVfy allows to avoid truncated downloads + 'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf', + 'page_url': url, + 'tbr': tbr, + 'ext': 'flv', + } + app = self._search_regex( + 'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None) + if app: + f.update({ + 'url': rtmp_url.split('?', 1)[0], + 'app': app, + }) + else: + f['url'] = rtmp_url + formats.append(f) + + for caption_url in video_element.findall('ClosedCaptioningURIs/URL'): + if caption_url.text: + extract_subtitle(caption_url.text) + + ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id') hmac = params.get('data-video-hmac') - if ios_playlist_url and hmac: + if ios_playlist_url and hmac and re.match(r'https?://', ios_playlist_url): headers = self.geo_verification_headers() headers.update({ 'Accept': 'application/vnd.itv.vod.playlist.v2+json', @@ -141,12 +208,12 @@ class ITVIE(InfoExtractor): 'token': '' }, 'device': { - 'manufacturer': 'Apple', - 'model': 'iPad', + 'manufacturer': 'Safari', + 'model': '5', 'os': { - 'name': 'iPhone OS', - 'version': '9.3', - 'type': 'ios' + 'name': 'Windows NT', + 'version': '6.1', + 'type': 'desktop' } }, 'client': { @@ -155,10 +222,10 @@ class ITVIE(InfoExtractor): }, 'variantAvailability': { 'featureset': { - 'min': ['hls', 'aes'], - 'max': ['hls', 'aes'] + 'min': ['hls', 'aes', 'outband-webvtt'], + 'max': ['hls', 'aes', 'outband-webvtt'] }, - 'platformTag': 'mobile' + 'platformTag': 'dotcom' } }).encode(), headers=headers, fatal=False) if ios_playlist: @@ -172,30 +239,74 @@ class ITVIE(InfoExtractor): href = ios_base_url + href ext = determine_ext(href) if ext == 'm3u8': - formats.extend(self._extract_m3u8_formats(href, video_id, 'mp4', m3u8_id='hls', fatal=False)) + formats.extend(self._extract_m3u8_formats( + href, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) else: formats.append({ 'url': href, }) - self._sort_formats(formats) + subs = video_data.get('Subtitles') + if isinstance(subs, list): + for sub in subs: + if not isinstance(sub, dict): + continue + href = url_or_none(sub.get('Href')) + if href: + extract_subtitle(href) + if not info.get('duration'): + info['duration'] = parse_duration(video_data.get('Duration')) - subtitles = {} - for caption_url in video_element.findall('ClosedCaptioningURIs/URL'): - if not caption_url.text: - continue - ext = determine_ext(caption_url.text, 'ttml') - subtitles.setdefault('en', []).append({ - 'url': caption_url.text, - 'ext': 'ttml' if ext == 'xml' else ext, - }) + self._sort_formats(formats) - return { + info.update({ 'id': video_id, - 'title': title, 'formats': formats, 'subtitles': subtitles, - 'episode_title': title, - 'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')), - 'series': xpath_text(playlist, 'ProgrammeTitle'), - 'duartion': parse_duration(xpath_text(playlist, 'Duration')), - } + }) + + webpage_info = self._search_json_ld(webpage, video_id, default={}) + if not webpage_info.get('title'): + webpage_info['title'] = self._html_search_regex( + r'(?s)]+\bclass=["\'][^>]*episode-title["\'][^>]*>([^<]+)<', + webpage, 'title', default=None) or self._og_search_title( + webpage, default=None) or self._html_search_meta( + 'twitter:title', webpage, 'title', + default=None) or webpage_info['episode'] + + return merge_dicts(info, webpage_info) + + +class ITVBTCCIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?itv\.com/btcc/(?:[^/]+/)*(?P[^/?#&]+)' + _TEST = { + 'url': 'http://www.itv.com/btcc/races/btcc-2018-all-the-action-from-brands-hatch', + 'info_dict': { + 'id': 'btcc-2018-all-the-action-from-brands-hatch', + 'title': 'BTCC 2018: All the action from Brands Hatch', + }, + 'playlist_mincount': 9, + } + BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1582188683001/HkiHLnNRx_default/index.html?videoId=%s' + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + entries = [ + self.url_result( + smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % video_id, { + # ITV does not like some GB IP ranges, so here are some + # IP blocks it accepts + 'geo_ip_blocks': [ + '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21' + ], + 'referrer': url, + }), + ie=BrightcoveNewIE.ie_key(), video_id=video_id) + for video_id in re.findall(r'data-video-id=["\'](\d+)', webpage)] + + title = self._og_search_title(webpage, fatal=False) + + return self.playlist_result(entries, playlist_id, title)