Fix "invalid escape sequences" error on Python 3.6
[youtube-dl] / youtube_dl / extractor / rai.py
1 from __future__ import unicode_literals
2
3 from .common import InfoExtractor
4 from ..compat import compat_urlparse
5 from ..utils import (
6     determine_ext,
7     ExtractorError,
8     find_xpath_attr,
9     fix_xml_ampersands,
10     int_or_none,
11     parse_duration,
12     unified_strdate,
13     update_url_query,
14     xpath_text,
15 )
16
17
18 class RaiBaseIE(InfoExtractor):
19     def _extract_relinker_formats(self, relinker_url, video_id):
20         formats = []
21
22         for platform in ('mon', 'flash', 'native'):
23             relinker = self._download_xml(
24                 relinker_url, video_id,
25                 note='Downloading XML metadata for platform %s' % platform,
26                 transform_source=fix_xml_ampersands,
27                 query={'output': 45, 'pl': platform},
28                 headers=self.geo_verification_headers())
29
30             media_url = find_xpath_attr(relinker, './url', 'type', 'content').text
31             if media_url == 'http://download.rai.it/video_no_available.mp4':
32                 self.raise_geo_restricted()
33
34             ext = determine_ext(media_url)
35             if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
36                 continue
37
38             if ext == 'm3u8':
39                 formats.extend(self._extract_m3u8_formats(
40                     media_url, video_id, 'mp4', 'm3u8_native',
41                     m3u8_id='hls', fatal=False))
42             elif ext == 'f4m':
43                 manifest_url = update_url_query(
44                     media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
45                     {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
46                 formats.extend(self._extract_f4m_formats(
47                     manifest_url, video_id, f4m_id='hds', fatal=False))
48             else:
49                 bitrate = int_or_none(xpath_text(relinker, 'bitrate'))
50                 formats.append({
51                     'url': media_url,
52                     'tbr': bitrate if bitrate > 0 else None,
53                     'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
54                 })
55
56         return formats
57
58     def _extract_from_content_id(self, content_id, base_url):
59         media = self._download_json(
60             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
61             content_id, 'Downloading video JSON')
62
63         thumbnails = []
64         for image_type in ('image', 'image_medium', 'image_300'):
65             thumbnail_url = media.get(image_type)
66             if thumbnail_url:
67                 thumbnails.append({
68                     'url': compat_urlparse.urljoin(base_url, thumbnail_url),
69                 })
70
71         formats = []
72         media_type = media['type']
73         if 'Audio' in media_type:
74             formats.append({
75                 'format_id': media.get('formatoAudio'),
76                 'url': media['audioUrl'],
77                 'ext': media.get('formatoAudio'),
78             })
79         elif 'Video' in media_type:
80             formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id))
81             self._sort_formats(formats)
82         else:
83             raise ExtractorError('not a media file')
84
85         subtitles = {}
86         captions = media.get('subtitlesUrl')
87         if captions:
88             STL_EXT = '.stl'
89             SRT_EXT = '.srt'
90             if captions.endswith(STL_EXT):
91                 captions = captions[:-len(STL_EXT)] + SRT_EXT
92             subtitles['it'] = [{
93                 'ext': 'srt',
94                 'url': captions,
95             }]
96
97         return {
98             'id': content_id,
99             'title': media['name'],
100             'description': media.get('desc'),
101             'thumbnails': thumbnails,
102             'uploader': media.get('author'),
103             'upload_date': unified_strdate(media.get('date')),
104             'duration': parse_duration(media.get('length')),
105             'formats': formats,
106             'subtitles': subtitles,
107         }
108
109
110 class RaiTVIE(RaiBaseIE):
111     _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+(?:media|ondemand)/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
112     _TESTS = [
113         {
114             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
115             'md5': '8970abf8caf8aef4696e7b1f2adfc696',
116             'info_dict': {
117                 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
118                 'ext': 'mp4',
119                 'title': 'Report del 07/04/2014',
120                 'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
121                 'upload_date': '20140407',
122                 'duration': 6160,
123                 'thumbnail': r're:^https?://.*\.jpg$',
124             }
125         },
126         {
127             # no m3u8 stream
128             'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
129             # HDS download, MD5 is unstable
130             'info_dict': {
131                 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
132                 'ext': 'flv',
133                 'title': 'TG PRIMO TEMPO',
134                 'upload_date': '20140612',
135                 'duration': 1758,
136                 'thumbnail': r're:^https?://.*\.jpg$',
137             },
138             'skip': 'Geo-restricted to Italy',
139         },
140         {
141             'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html',
142             'md5': '35cf7c229f22eeef43e48b5cf923bef0',
143             'info_dict': {
144                 'id': '7aafdea9-0e5d-49d5-88a6-7e65da67ae13',
145                 'ext': 'mp4',
146                 'title': 'State of the Net, Antonella La Carpia: regole virali',
147                 'description': 'md5:b0ba04a324126903e3da7763272ae63c',
148                 'upload_date': '20140613',
149             },
150             'skip': 'Error 404',
151         },
152         {
153             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html',
154             'info_dict': {
155                 'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132',
156                 'ext': 'mp4',
157                 'title': 'Alluvione in Sardegna e dissesto idrogeologico',
158                 'description': 'Edizione delle ore 20:30 ',
159             },
160             'skip': 'invalid urls',
161         },
162         {
163             'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
164             'md5': 'e57493e1cb8bc7c564663f363b171847',
165             'info_dict': {
166                 'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
167                 'ext': 'mp4',
168                 'title': 'Il Candidato - Primo episodio: "Le Primarie"',
169                 'description': 'md5:364b604f7db50594678f483353164fb8',
170                 'upload_date': '20140923',
171                 'duration': 386,
172                 'thumbnail': r're:^https?://.*\.jpg$',
173             }
174         },
175     ]
176
177     def _real_extract(self, url):
178         video_id = self._match_id(url)
179
180         return self._extract_from_content_id(video_id, url)
181
182
183 class RaiIE(RaiBaseIE):
184     _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
185     _TESTS = [
186         {
187             'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
188             'md5': '2dd727e61114e1ee9c47f0da6914e178',
189             'info_dict': {
190                 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
191                 'ext': 'mp4',
192                 'title': 'Il pacco',
193                 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
194                 'upload_date': '20141221',
195             },
196         },
197         {
198             # Direct relinker URL
199             'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
200             # HDS live stream, MD5 is unstable
201             'info_dict': {
202                 'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
203                 'ext': 'flv',
204                 'title': 'EuroNews',
205             },
206             'skip': 'Geo-restricted to Italy',
207         },
208         {
209             # Embedded content item ID
210             'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
211             'md5': '84c1135ce960e8822ae63cec34441d63',
212             'info_dict': {
213                 'id': '0960e765-62c8-474a-ac4b-7eb3e2be39c8',
214                 'ext': 'mp4',
215                 'title': 'TG1 ore 20:00 del 02/07/2016',
216                 'upload_date': '20160702',
217             },
218         },
219         {
220             'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
221             # HDS live stream, MD5 is unstable
222             'info_dict': {
223                 'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
224                 'ext': 'flv',
225                 'title': 'La diretta di Rainews24',
226             },
227         },
228     ]
229
230     @classmethod
231     def suitable(cls, url):
232         return False if RaiTVIE.suitable(url) else super(RaiIE, cls).suitable(url)
233
234     def _real_extract(self, url):
235         video_id = self._match_id(url)
236         webpage = self._download_webpage(url, video_id)
237
238         iframe_url = self._search_regex(
239             [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"',
240              r'drawMediaRaiTV\(["\'](.+?)["\']'],
241             webpage, 'iframe', default=None)
242         if iframe_url:
243             if not iframe_url.startswith('http'):
244                 iframe_url = compat_urlparse.urljoin(url, iframe_url)
245             return self.url_result(iframe_url)
246
247         content_item_id = self._search_regex(
248             r'initEdizione\((?P<q1>[\'"])ContentItem-(?P<content_id>[^\'"]+)(?P=q1)',
249             webpage, 'content item ID', group='content_id', default=None)
250         if content_item_id:
251             return self._extract_from_content_id(content_item_id, url)
252
253         relinker_url = compat_urlparse.urljoin(url, self._search_regex(
254             r'(?:var\s+videoURL|mediaInfo\.mediaUri)\s*=\s*(?P<q1>[\'"])(?P<url>(https?:)?//mediapolis\.rai\.it/relinker/relinkerServlet\.htm\?cont=\d+)(?P=q1)',
255             webpage, 'relinker URL', group='url'))
256         formats = self._extract_relinker_formats(relinker_url, video_id)
257         self._sort_formats(formats)
258
259         title = self._search_regex(
260             r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
261             webpage, 'title', group='title', default=None) or self._og_search_title(webpage)
262
263         return {
264             'id': video_id,
265             'title': title,
266             'formats': formats,
267         }