[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / arte.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import compat_str
8 from ..utils import (
9     ExtractorError,
10     int_or_none,
11     qualities,
12     try_get,
13     unified_strdate,
14 )
15
16 # There are different sources of video in arte.tv, the extraction process
17 # is different for each one. The videos usually expire in 7 days, so we can't
18 # add tests.
19
20
21 class ArteTVBaseIE(InfoExtractor):
22     def _extract_from_json_url(self, json_url, video_id, lang, title=None):
23         info = self._download_json(json_url, video_id)
24         player_info = info['videoJsonPlayer']
25
26         vsr = try_get(player_info, lambda x: x['VSR'], dict)
27         if not vsr:
28             error = None
29             if try_get(player_info, lambda x: x['custom_msg']['type']) == 'error':
30                 error = try_get(
31                     player_info, lambda x: x['custom_msg']['msg'], compat_str)
32             if not error:
33                 error = 'Video %s is not available' % player_info.get('VID') or video_id
34             raise ExtractorError(error, expected=True)
35
36         upload_date_str = player_info.get('shootingDate')
37         if not upload_date_str:
38             upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
39
40         title = (player_info.get('VTI') or title or player_info['VID']).strip()
41         subtitle = player_info.get('VSU', '').strip()
42         if subtitle:
43             title += ' - %s' % subtitle
44
45         info_dict = {
46             'id': player_info['VID'],
47             'title': title,
48             'description': player_info.get('VDE'),
49             'upload_date': unified_strdate(upload_date_str),
50             'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
51         }
52         qfunc = qualities(['MQ', 'HQ', 'EQ', 'SQ'])
53
54         LANGS = {
55             'fr': 'F',
56             'de': 'A',
57             'en': 'E[ANG]',
58             'es': 'E[ESP]',
59             'it': 'E[ITA]',
60             'pl': 'E[POL]',
61         }
62
63         langcode = LANGS.get(lang, lang)
64
65         formats = []
66         for format_id, format_dict in vsr.items():
67             f = dict(format_dict)
68             versionCode = f.get('versionCode')
69             l = re.escape(langcode)
70
71             # Language preference from most to least priority
72             # Reference: section 6.8 of
73             # https://www.arte.tv/sites/en/corporate/files/complete-technical-guidelines-arte-geie-v1-07-1.pdf
74             PREFERENCES = (
75                 # original version in requested language, without subtitles
76                 r'VO{0}$'.format(l),
77                 # original version in requested language, with partial subtitles in requested language
78                 r'VO{0}-ST{0}$'.format(l),
79                 # original version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
80                 r'VO{0}-STM{0}$'.format(l),
81                 # non-original (dubbed) version in requested language, without subtitles
82                 r'V{0}$'.format(l),
83                 # non-original (dubbed) version in requested language, with subtitles partial subtitles in requested language
84                 r'V{0}-ST{0}$'.format(l),
85                 # non-original (dubbed) version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
86                 r'V{0}-STM{0}$'.format(l),
87                 # original version in requested language, with partial subtitles in different language
88                 r'VO{0}-ST(?!{0}).+?$'.format(l),
89                 # original version in requested language, with subtitles for the deaf and hard-of-hearing in different language
90                 r'VO{0}-STM(?!{0}).+?$'.format(l),
91                 # original version in different language, with partial subtitles in requested language
92                 r'VO(?:(?!{0}).+?)?-ST{0}$'.format(l),
93                 # original version in different language, with subtitles for the deaf and hard-of-hearing in requested language
94                 r'VO(?:(?!{0}).+?)?-STM{0}$'.format(l),
95                 # original version in different language, without subtitles
96                 r'VO(?:(?!{0}))?$'.format(l),
97                 # original version in different language, with partial subtitles in different language
98                 r'VO(?:(?!{0}).+?)?-ST(?!{0}).+?$'.format(l),
99                 # original version in different language, with subtitles for the deaf and hard-of-hearing in different language
100                 r'VO(?:(?!{0}).+?)?-STM(?!{0}).+?$'.format(l),
101             )
102
103             for pref, p in enumerate(PREFERENCES):
104                 if re.match(p, versionCode):
105                     lang_pref = len(PREFERENCES) - pref
106                     break
107             else:
108                 lang_pref = -1
109
110             format = {
111                 'format_id': format_id,
112                 'preference': -10 if f.get('videoFormat') == 'M3U8' else None,
113                 'language_preference': lang_pref,
114                 'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
115                 'width': int_or_none(f.get('width')),
116                 'height': int_or_none(f.get('height')),
117                 'tbr': int_or_none(f.get('bitrate')),
118                 'quality': qfunc(f.get('quality')),
119             }
120
121             if f.get('mediaType') == 'rtmp':
122                 format['url'] = f['streamer']
123                 format['play_path'] = 'mp4:' + f['url']
124                 format['ext'] = 'flv'
125             else:
126                 format['url'] = f['url']
127
128             formats.append(format)
129
130         self._check_formats(formats, video_id)
131         self._sort_formats(formats)
132
133         info_dict['formats'] = formats
134         return info_dict
135
136
137 class ArteTVPlus7IE(ArteTVBaseIE):
138     IE_NAME = 'arte.tv:+7'
139     _VALID_URL = r'https?://(?:www\.)?arte\.tv/(?P<lang>fr|de|en|es|it|pl)/videos/(?P<id>\d{6}-\d{3}-[AF])'
140
141     _TESTS = [{
142         'url': 'https://www.arte.tv/en/videos/088501-000-A/mexico-stealing-petrol-to-survive/',
143         'info_dict': {
144             'id': '088501-000-A',
145             'ext': 'mp4',
146             'title': 'Mexico: Stealing Petrol to Survive',
147             'upload_date': '20190628',
148         },
149     }]
150
151     def _real_extract(self, url):
152         lang, video_id = re.match(self._VALID_URL, url).groups()
153         return self._extract_from_json_url(
154             'https://api.arte.tv/api/player/v1/config/%s/%s' % (lang, video_id),
155             video_id, lang)
156
157
158 class ArteTVEmbedIE(ArteTVPlus7IE):
159     IE_NAME = 'arte.tv:embed'
160     _VALID_URL = r'''(?x)
161         https://www\.arte\.tv
162         /player/v3/index\.php\?json_url=
163         (?P<json_url>
164             https?://api\.arte\.tv/api/player/v1/config/
165             (?P<lang>[^/]+)/(?P<id>\d{6}-\d{3}-[AF])
166         )
167     '''
168
169     _TESTS = []
170
171     def _real_extract(self, url):
172         json_url, lang, video_id = re.match(self._VALID_URL, url).groups()
173         return self._extract_from_json_url(json_url, video_id, lang)
174
175
176 class ArteTVPlaylistIE(ArteTVBaseIE):
177     IE_NAME = 'arte.tv:playlist'
178     _VALID_URL = r'https?://(?:www\.)?arte\.tv/(?P<lang>fr|de|en|es|it|pl)/videos/(?P<id>RC-\d{6})'
179
180     _TESTS = [{
181         'url': 'https://www.arte.tv/en/videos/RC-016954/earn-a-living/',
182         'info_dict': {
183             'id': 'RC-016954',
184             'title': 'Earn a Living',
185             'description': 'md5:d322c55011514b3a7241f7fb80d494c2',
186         },
187         'playlist_mincount': 6,
188     }]
189
190     def _real_extract(self, url):
191         lang, playlist_id = re.match(self._VALID_URL, url).groups()
192         collection = self._download_json(
193             'https://api.arte.tv/api/player/v1/collectionData/%s/%s?source=videos'
194             % (lang, playlist_id), playlist_id)
195         title = collection.get('title')
196         description = collection.get('shortDescription') or collection.get('teaserText')
197         entries = [
198             self._extract_from_json_url(
199                 video['jsonUrl'], video.get('programId') or playlist_id, lang)
200             for video in collection['videos'] if video.get('jsonUrl')]
201         return self.playlist_result(entries, playlist_id, title, description)