Merge branch 'master' of github.com:rg3/youtube-dl into weibo
[ytdl] / youtube_dl / extractor / itv.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import uuid
5 import xml.etree.ElementTree as etree
6 import json
7 import re
8
9 from .common import InfoExtractor
10 from ..compat import (
11     compat_str,
12     compat_etree_register_namespace,
13 )
14 from ..utils import (
15     extract_attributes,
16     xpath_with_ns,
17     xpath_element,
18     xpath_text,
19     int_or_none,
20     parse_duration,
21     ExtractorError,
22     determine_ext,
23 )
24
25
26 class ITVIE(InfoExtractor):
27     _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
28     _GEO_COUNTRIES = ['GB']
29     _TESTS = [{
30         'url': 'http://www.itv.com/hub/mr-bean-animated-series/2a2936a0053',
31         'info_dict': {
32             'id': '2a2936a0053',
33             'ext': 'flv',
34             'title': 'Home Movie',
35         },
36         'params': {
37             # rtmp download
38             'skip_download': True,
39         },
40     }, {
41         # unavailable via data-playlist-url
42         'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
43         'only_matching': True,
44     }]
45
46     def _real_extract(self, url):
47         video_id = self._match_id(url)
48         webpage = self._download_webpage(url, video_id)
49         params = extract_attributes(self._search_regex(
50             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
51
52         ns_map = {
53             'soapenv': 'http://schemas.xmlsoap.org/soap/envelope/',
54             'tem': 'http://tempuri.org/',
55             'itv': 'http://schemas.datacontract.org/2004/07/Itv.BB.Mercury.Common.Types',
56             'com': 'http://schemas.itv.com/2009/05/Common',
57         }
58         for ns, full_ns in ns_map.items():
59             compat_etree_register_namespace(ns, full_ns)
60
61         def _add_ns(name):
62             return xpath_with_ns(name, ns_map)
63
64         def _add_sub_element(element, name):
65             return etree.SubElement(element, _add_ns(name))
66
67         production_id = (
68             params.get('data-video-autoplay-id') or
69             '%s#001' % (
70                 params.get('data-video-episode-id') or
71                 video_id.replace('a', '/')))
72
73         req_env = etree.Element(_add_ns('soapenv:Envelope'))
74         _add_sub_element(req_env, 'soapenv:Header')
75         body = _add_sub_element(req_env, 'soapenv:Body')
76         get_playlist = _add_sub_element(body, ('tem:GetPlaylist'))
77         request = _add_sub_element(get_playlist, 'tem:request')
78         _add_sub_element(request, 'itv:ProductionId').text = production_id
79         _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper()
80         vodcrid = _add_sub_element(request, 'itv:Vodcrid')
81         _add_sub_element(vodcrid, 'com:Id')
82         _add_sub_element(request, 'itv:Partition')
83         user_info = _add_sub_element(get_playlist, 'tem:userInfo')
84         _add_sub_element(user_info, 'itv:Broadcaster').text = 'Itv'
85         _add_sub_element(user_info, 'itv:DM')
86         _add_sub_element(user_info, 'itv:RevenueScienceValue')
87         _add_sub_element(user_info, 'itv:SessionId')
88         _add_sub_element(user_info, 'itv:SsoToken')
89         _add_sub_element(user_info, 'itv:UserToken')
90         site_info = _add_sub_element(get_playlist, 'tem:siteInfo')
91         _add_sub_element(site_info, 'itv:AdvertisingRestriction').text = 'None'
92         _add_sub_element(site_info, 'itv:AdvertisingSite').text = 'ITV'
93         _add_sub_element(site_info, 'itv:AdvertisingType').text = 'Any'
94         _add_sub_element(site_info, 'itv:Area').text = 'ITVPLAYER.VIDEO'
95         _add_sub_element(site_info, 'itv:Category')
96         _add_sub_element(site_info, 'itv:Platform').text = 'DotCom'
97         _add_sub_element(site_info, 'itv:Site').text = 'ItvCom'
98         device_info = _add_sub_element(get_playlist, 'tem:deviceInfo')
99         _add_sub_element(device_info, 'itv:ScreenSize').text = 'Big'
100         player_info = _add_sub_element(get_playlist, 'tem:playerInfo')
101         _add_sub_element(player_info, 'itv:Version').text = '2'
102
103         headers = self.geo_verification_headers()
104         headers.update({
105             'Content-Type': 'text/xml; charset=utf-8',
106             'SOAPAction': 'http://tempuri.org/PlaylistService/GetPlaylist',
107         })
108
109         info = self._search_json_ld(webpage, video_id, default={})
110         formats = []
111         subtitles = {}
112
113         def extract_subtitle(sub_url):
114             ext = determine_ext(sub_url, 'ttml')
115             subtitles.setdefault('en', []).append({
116                 'url': sub_url,
117                 'ext': 'ttml' if ext == 'xml' else ext,
118             })
119
120         resp_env = self._download_xml(
121             params['data-playlist-url'], video_id,
122             headers=headers, data=etree.tostring(req_env))
123         playlist = xpath_element(resp_env, './/Playlist')
124         if playlist is None:
125             fault_code = xpath_text(resp_env, './/faultcode')
126             fault_string = xpath_text(resp_env, './/faultstring')
127             if fault_code == 'InvalidGeoRegion':
128                 self.raise_geo_restricted(
129                     msg=fault_string, countries=self._GEO_COUNTRIES)
130             elif fault_code != 'InvalidEntity':
131                 raise ExtractorError(
132                     '%s said: %s' % (self.IE_NAME, fault_string), expected=True)
133             info.update({
134                 'title': self._og_search_title(webpage),
135                 'episode_title': params.get('data-video-episode'),
136                 'series': params.get('data-video-title'),
137             })
138         else:
139             title = xpath_text(playlist, 'EpisodeTitle', default=None)
140             info.update({
141                 'title': title,
142                 'episode_title': title,
143                 'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')),
144                 'series': xpath_text(playlist, 'ProgrammeTitle'),
145                 'duration': parse_duration(xpath_text(playlist, 'Duration')),
146             })
147             video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True)
148             media_files = xpath_element(video_element, 'MediaFiles', fatal=True)
149             rtmp_url = media_files.attrib['base']
150
151             for media_file in media_files.findall('MediaFile'):
152                 play_path = xpath_text(media_file, 'URL')
153                 if not play_path:
154                     continue
155                 tbr = int_or_none(media_file.get('bitrate'), 1000)
156                 f = {
157                     'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''),
158                     'play_path': play_path,
159                     # Providing this swfVfy allows to avoid truncated downloads
160                     'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf',
161                     'page_url': url,
162                     'tbr': tbr,
163                     'ext': 'flv',
164                 }
165                 app = self._search_regex(
166                     'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None)
167                 if app:
168                     f.update({
169                         'url': rtmp_url.split('?', 1)[0],
170                         'app': app,
171                     })
172                 else:
173                     f['url'] = rtmp_url
174                 formats.append(f)
175
176             for caption_url in video_element.findall('ClosedCaptioningURIs/URL'):
177                 if caption_url.text:
178                     extract_subtitle(caption_url.text)
179
180         ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id')
181         hmac = params.get('data-video-hmac')
182         if ios_playlist_url and hmac and re.match(r'https?://', ios_playlist_url):
183             headers = self.geo_verification_headers()
184             headers.update({
185                 'Accept': 'application/vnd.itv.vod.playlist.v2+json',
186                 'Content-Type': 'application/json',
187                 'hmac': hmac.upper(),
188             })
189             ios_playlist = self._download_json(
190                 ios_playlist_url, video_id, data=json.dumps({
191                     'user': {
192                         'itvUserId': '',
193                         'entitlements': [],
194                         'token': ''
195                     },
196                     'device': {
197                         'manufacturer': 'Safari',
198                         'model': '5',
199                         'os': {
200                             'name': 'Windows NT',
201                             'version': '6.1',
202                             'type': 'desktop'
203                         }
204                     },
205                     'client': {
206                         'version': '4.1',
207                         'id': 'browser'
208                     },
209                     'variantAvailability': {
210                         'featureset': {
211                             'min': ['hls', 'aes', 'outband-webvtt'],
212                             'max': ['hls', 'aes', 'outband-webvtt']
213                         },
214                         'platformTag': 'dotcom'
215                     }
216                 }).encode(), headers=headers, fatal=False)
217             if ios_playlist:
218                 video_data = ios_playlist.get('Playlist', {}).get('Video', {})
219                 ios_base_url = video_data.get('Base')
220                 for media_file in video_data.get('MediaFiles', []):
221                     href = media_file.get('Href')
222                     if not href:
223                         continue
224                     if ios_base_url:
225                         href = ios_base_url + href
226                     ext = determine_ext(href)
227                     if ext == 'm3u8':
228                         formats.extend(self._extract_m3u8_formats(
229                             href, video_id, 'mp4', entry_protocol='m3u8_native',
230                             m3u8_id='hls', fatal=False))
231                     else:
232                         formats.append({
233                             'url': href,
234                         })
235                 subs = video_data.get('Subtitles')
236                 if isinstance(subs, list):
237                     for sub in subs:
238                         if not isinstance(sub, dict):
239                             continue
240                         href = sub.get('Href')
241                         if isinstance(href, compat_str):
242                             extract_subtitle(href)
243                 if not info.get('duration'):
244                     info['duration'] = parse_duration(video_data.get('Duration'))
245
246         self._sort_formats(formats)
247
248         info.update({
249             'id': video_id,
250             'formats': formats,
251             'subtitles': subtitles,
252         })
253         return info