X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/d9d7cd0e85dc712461d9185db9df9d6c900a573b..fe3f5064d6bfc2fe705eef08b4f75309b95b55b2:/youtube_dl/extractor/itv.py diff --git a/youtube_dl/extractor/itv.py b/youtube_dl/extractor/itv.py index f315680..18a7d7f 100644 --- a/youtube_dl/extractor/itv.py +++ b/youtube_dl/extractor/itv.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import uuid import xml.etree.ElementTree as etree import json +import re from .common import InfoExtractor from ..compat import ( @@ -25,7 +26,7 @@ from ..utils import ( 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 +37,11 @@ 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, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -59,12 +64,18 @@ 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') @@ -94,6 +105,18 @@ class ITVIE(InfoExtractor): 'Content-Type': 'text/xml; charset=utf-8', 'SOAPAction': 'http://tempuri.org/PlaylistService/GetPlaylist', }) + + info = self._search_json_ld(webpage, video_id, default={}) + formats = [] + 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, + }) + resp_env = self._download_xml( params['data-playlist-url'], video_id, headers=headers, data=etree.tostring(req_env)) @@ -104,41 +127,59 @@ class ITVIE(InfoExtractor): 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'] + elif fault_code != 'InvalidEntity': + 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'] - 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) - 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?://[^/]+/(.+)$', rtmp_url, 'app', default=None) - if app: - f.update({ - 'url': rtmp_url.split('?', 1)[0], - 'app': app, - }) - else: - f['url'] = rtmp_url - formats.append(f) - - ios_playlist_url = params.get('data-video-playlist') + for media_file in media_files.findall('MediaFile'): + play_path = xpath_text(media_file, 'URL') + if not play_path: + continue + tbr = int_or_none(media_file.get('bitrate'), 1000) + f = { + 'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''), + 'play_path': play_path, + # Providing this swfVfy allows to avoid truncated downloads + 'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf', + 'page_url': url, + 'tbr': tbr, + 'ext': 'flv', + } + app = self._search_regex( + 'rtmpe?://[^/]+/(.+)$', 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', @@ -153,12 +194,12 @@ 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': { @@ -167,10 +208,10 @@ 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: @@ -191,27 +232,22 @@ class ITVIE(InfoExtractor): 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 = sub.get('Href') + if isinstance(href, compat_str): + 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) - info = self._search_json_ld(webpage, video_id, default={}) 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')), }) return info