]>
Commit | Line | Data |
---|---|---|
32dac694 PH |
1 | from __future__ import unicode_literals |
2 | ||
fc287219 | 3 | import re |
fc287219 | 4 | |
0af25f78 | 5 | from .common import InfoExtractor |
1cc79574 | 6 | from ..compat import ( |
e525d9a3 | 7 | compat_str, |
20a6a154 | 8 | compat_xpath, |
1cc79574 PH |
9 | ) |
10 | from ..utils import ( | |
fc287219 | 11 | ExtractorError, |
90834c78 | 12 | find_xpath_attr, |
5aafe895 | 13 | fix_xml_ampersands, |
8765151c | 14 | float_or_none, |
0ef68e04 | 15 | HEADRequest, |
0c75abbb | 16 | RegexNotFoundError, |
5c2266df | 17 | sanitized_Request, |
7cdfc4c9 | 18 | strip_or_none, |
712c7530 | 19 | timeconvert, |
340b0468 | 20 | unescapeHTML, |
0c75abbb | 21 | update_url_query, |
8d9453b9 | 22 | url_basename, |
5767b4ee | 23 | xpath_text, |
fc287219 PH |
24 | ) |
25 | ||
90834c78 | 26 | |
300fcad8 JMF |
27 | def _media_xml_tag(tag): |
28 | return '{http://search.yahoo.com/mrss/}%s' % tag | |
fc287219 | 29 | |
f7e02595 | 30 | |
0af25f78 | 31 | class MTVServicesInfoExtractor(InfoExtractor): |
340b0468 | 32 | _MOBILE_TEMPLATE = None |
a542e372 | 33 | _LANG = None |
8940c1c0 | 34 | |
f7e02595 JMF |
35 | @staticmethod |
36 | def _id_from_uri(uri): | |
37 | return uri.split(':')[-1] | |
38 | ||
0c75abbb YCH |
39 | @staticmethod |
40 | def _remove_template_parameter(url): | |
41 | # Remove the templates, like &device={device} | |
42 | return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url) | |
43 | ||
f7e02595 | 44 | # This was originally implemented for ComedyCentral, but it also works here |
712c7530 YCH |
45 | @classmethod |
46 | def _transform_rtmp_url(cls, rtmp_video_url): | |
f7e02595 JMF |
47 | m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url) |
48 | if not m: | |
712c7530 | 49 | return {'rtmp': rtmp_video_url} |
2774852c | 50 | base = 'http://viacommtvstrmfs.fplive.net/' |
712c7530 | 51 | return {'http': base + m.group('finalid')} |
ab2f744b | 52 | |
8940c1c0 JMF |
53 | def _get_feed_url(self, uri): |
54 | return self._FEED_URL | |
55 | ||
ab2f744b | 56 | def _get_thumbnail_url(self, uri, itemdoc): |
84db8181 JMF |
57 | search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail')) |
58 | thumb_node = itemdoc.find(search_path) | |
59 | if thumb_node is None: | |
60 | return None | |
61 | else: | |
62 | return thumb_node.attrib['url'] | |
f7e02595 | 63 | |
340b0468 JMF |
64 | def _extract_mobile_video_formats(self, mtvn_id): |
65 | webpage_url = self._MOBILE_TEMPLATE % mtvn_id | |
5c2266df | 66 | req = sanitized_Request(webpage_url) |
340b0468 | 67 | # Otherwise we get a webpage that would execute some javascript |
3fcfb8e9 | 68 | req.add_header('User-Agent', 'curl/7') |
340b0468 | 69 | webpage = self._download_webpage(req, mtvn_id, |
9e1a5b84 | 70 | 'Downloading mobile page') |
0ef68e04 JMF |
71 | metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url')) |
72 | req = HEADRequest(metrics_url) | |
73 | response = self._request_webpage(req, mtvn_id, 'Resolving url') | |
74 | url = response.geturl() | |
75 | # Transform the url to get the best quality: | |
76 | url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1) | |
5f6a1245 | 77 | return [{'url': url, 'ext': 'mp4'}] |
340b0468 JMF |
78 | |
79 | def _extract_video_formats(self, mdoc, mtvn_id): | |
054d43bb | 80 | if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc.find('.//src').text) is not None: |
340b0468 | 81 | if mtvn_id is not None and self._MOBILE_TEMPLATE is not None: |
0ef68e04 | 82 | self.to_screen('The normal version is not available from your ' |
9e1a5b84 | 83 | 'country, trying with the mobile version') |
340b0468 | 84 | return self._extract_mobile_video_formats(mtvn_id) |
cc1db7f9 | 85 | raise ExtractorError('This video is not available from your country.', |
9e1a5b84 | 86 | expected=True) |
f7e02595 | 87 | |
f13d0933 PH |
88 | formats = [] |
89 | for rendition in mdoc.findall('.//rendition'): | |
90 | try: | |
91 | _, _, ext = rendition.attrib['type'].partition('/') | |
92 | rtmp_video_url = rendition.find('./src').text | |
ca0f500e PH |
93 | if rtmp_video_url.endswith('siteunavail.png'): |
94 | continue | |
712c7530 YCH |
95 | new_urls = self._transform_rtmp_url(rtmp_video_url) |
96 | formats.extend([{ | |
a72df5f3 YCH |
97 | 'ext': 'flv' if new_url.startswith('rtmp') else ext, |
98 | 'url': new_url, | |
10a1bb3a | 99 | 'format_id': '-'.join(filter(None, [kind, rendition.get('bitrate')])), |
ca0f500e PH |
100 | 'width': int(rendition.get('width')), |
101 | 'height': int(rendition.get('height')), | |
712c7530 | 102 | } for kind, new_url in new_urls.items()]) |
f13d0933 PH |
103 | except (KeyError, TypeError): |
104 | raise ExtractorError('Invalid rendition field.') | |
34d863f3 | 105 | self._sort_formats(formats) |
f13d0933 | 106 | return formats |
f7e02595 | 107 | |
e525d9a3 S |
108 | def _extract_subtitles(self, mdoc, mtvn_id): |
109 | subtitles = {} | |
e525d9a3 S |
110 | for transcript in mdoc.findall('.//transcript'): |
111 | if transcript.get('kind') != 'captions': | |
112 | continue | |
113 | lang = transcript.get('srclang') | |
0af25f78 JMF |
114 | subtitles[lang] = [{ |
115 | 'url': compat_str(typographic.get('src')), | |
116 | 'ext': typographic.get('format') | |
117 | } for typographic in transcript.findall('./typographic')] | |
118 | return subtitles | |
e525d9a3 | 119 | |
f7e02595 JMF |
120 | def _get_video_info(self, itemdoc): |
121 | uri = itemdoc.find('guid').text | |
122 | video_id = self._id_from_uri(uri) | |
123 | self.report_extraction(video_id) | |
04cbc498 | 124 | content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))) |
0c75abbb | 125 | mediagen_url = self._remove_template_parameter(content_el.attrib['url']) |
f7e02595 | 126 | if 'acceptMethods' not in mediagen_url: |
56f447be S |
127 | mediagen_url += '&' if '?' in mediagen_url else '?' |
128 | mediagen_url += 'acceptMethods=fms' | |
6562df76 | 129 | |
e4f320a4 | 130 | mediagen_doc = self._download_xml(mediagen_url, video_id, |
9e1a5b84 | 131 | 'Downloading video urls') |
f7e02595 | 132 | |
0dfe9bc9 S |
133 | item = mediagen_doc.find('./video/item') |
134 | if item is not None and item.get('type') == 'text': | |
135 | message = '%s returned error: ' % self.IE_NAME | |
136 | if item.get('code') is not None: | |
137 | message += '%s - ' % item.get('code') | |
138 | message += item.text | |
139 | raise ExtractorError(message, expected=True) | |
140 | ||
7cdfc4c9 | 141 | description = strip_or_none(xpath_text(itemdoc, 'description')) |
f13d0933 | 142 | |
712c7530 YCH |
143 | timestamp = timeconvert(xpath_text(itemdoc, 'pubDate')) |
144 | ||
90834c78 PH |
145 | title_el = None |
146 | if title_el is None: | |
147 | title_el = find_xpath_attr( | |
148 | itemdoc, './/{http://search.yahoo.com/mrss/}category', | |
149 | 'scheme', 'urn:mtvn:video_title') | |
90834c78 | 150 | if title_el is None: |
20a6a154 | 151 | title_el = itemdoc.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title')) |
96cb10a5 | 152 | if title_el is None: |
20a6a154 | 153 | title_el = itemdoc.find(compat_xpath('.//title')) |
1df4229b PH |
154 | if title_el.text is None: |
155 | title_el = None | |
1df4229b | 156 | |
90834c78 PH |
157 | title = title_el.text |
158 | if title is None: | |
159 | raise ExtractorError('Could not find video title') | |
780ee4e5 | 160 | title = title.strip() |
90834c78 | 161 | |
340b0468 JMF |
162 | # This a short id that's used in the webpage urls |
163 | mtvn_id = None | |
164 | mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category', | |
9e1a5b84 | 165 | 'scheme', 'urn:mtvn:id') |
340b0468 JMF |
166 | if mtvn_id_node is not None: |
167 | mtvn_id = mtvn_id_node.text | |
168 | ||
fb7abb31 | 169 | return { |
90834c78 | 170 | 'title': title, |
340b0468 | 171 | 'formats': self._extract_video_formats(mediagen_doc, mtvn_id), |
e525d9a3 | 172 | 'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id), |
f13d0933 PH |
173 | 'id': video_id, |
174 | 'thumbnail': self._get_thumbnail_url(uri, itemdoc), | |
175 | 'description': description, | |
04cbc498 | 176 | 'duration': float_or_none(content_el.attrib.get('duration')), |
712c7530 | 177 | 'timestamp': timestamp, |
f13d0933 PH |
178 | } |
179 | ||
c1e90619 | 180 | def _get_feed_query(self, uri): |
181 | data = {'uri': uri} | |
182 | if self._LANG: | |
183 | data['lang'] = self._LANG | |
0c75abbb | 184 | return data |
c1e90619 | 185 | |
f7e02595 JMF |
186 | def _get_videos_info(self, uri): |
187 | video_id = self._id_from_uri(uri) | |
8940c1c0 | 188 | feed_url = self._get_feed_url(uri) |
0c75abbb | 189 | info_url = update_url_query(feed_url, self._get_feed_query(uri)) |
79fa9db0 S |
190 | return self._get_videos_info_from_url(info_url, video_id) |
191 | ||
192 | def _get_videos_info_from_url(self, url, video_id): | |
e2b38da9 | 193 | idoc = self._download_xml( |
79fa9db0 | 194 | url, video_id, |
32dac694 | 195 | 'Downloading info', transform_source=fix_xml_ampersands) |
712c7530 YCH |
196 | |
197 | title = xpath_text(idoc, './channel/title') | |
198 | description = xpath_text(idoc, './channel/description') | |
199 | ||
5239075b | 200 | return self.playlist_result( |
712c7530 YCH |
201 | [self._get_video_info(item) for item in idoc.findall('.//item')], |
202 | playlist_title=title, playlist_description=description) | |
fc287219 | 203 | |
c1e90619 | 204 | def _extract_mgid(self, webpage): |
8d9453b9 | 205 | try: |
4bbf139a JMF |
206 | # the url can be http://media.mtvnservices.com/fb/{mgid}.swf |
207 | # or http://media.mtvnservices.com/{mgid} | |
208 | og_url = self._og_search_video_url(webpage) | |
209 | mgid = url_basename(og_url) | |
210 | if mgid.endswith('.swf'): | |
211 | mgid = mgid[:-4] | |
8d9453b9 | 212 | except RegexNotFoundError: |
b1298d8e AMW |
213 | mgid = None |
214 | ||
215 | if mgid is None or ':' not in mgid: | |
b9381e43 JMF |
216 | mgid = self._search_regex( |
217 | [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'], | |
fc42bc6e S |
218 | webpage, 'mgid', default=None) |
219 | ||
220 | if not mgid: | |
221 | sm4_embed = self._html_search_meta( | |
222 | 'sm4:video:embed', webpage, 'sm4 embed', default='') | |
223 | mgid = self._search_regex( | |
224 | r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid') | |
c1e90619 | 225 | return mgid |
e525d9a3 | 226 | |
c1e90619 | 227 | def _real_extract(self, url): |
228 | title = url_basename(url) | |
229 | webpage = self._download_webpage(url, title) | |
230 | mgid = self._extract_mgid(webpage) | |
e525d9a3 | 231 | videos_info = self._get_videos_info(mgid) |
e525d9a3 | 232 | return videos_info |
8d9453b9 | 233 | |
84db8181 | 234 | |
8940c1c0 JMF |
235 | class MTVServicesEmbeddedIE(MTVServicesInfoExtractor): |
236 | IE_NAME = 'mtvservices:embedded' | |
237 | _VALID_URL = r'https?://media\.mtvnservices\.com/embed/(?P<mgid>.+?)(\?|/|$)' | |
238 | ||
239 | _TEST = { | |
240 | # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/ | |
241 | 'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906', | |
242 | 'md5': 'cb349b21a7897164cede95bd7bf3fbb9', | |
243 | 'info_dict': { | |
244 | 'id': '1043906', | |
245 | 'ext': 'mp4', | |
246 | 'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds', | |
247 | 'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.', | |
712c7530 YCH |
248 | 'timestamp': 1400126400, |
249 | 'upload_date': '20140515', | |
8940c1c0 JMF |
250 | }, |
251 | } | |
252 | ||
fe1d858e S |
253 | @staticmethod |
254 | def _extract_url(webpage): | |
255 | mobj = re.search( | |
256 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage) | |
257 | if mobj: | |
258 | return mobj.group('url') | |
259 | ||
8940c1c0 JMF |
260 | def _get_feed_url(self, uri): |
261 | video_id = self._id_from_uri(uri) | |
0c75abbb YCH |
262 | config = self._download_json( |
263 | 'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id) | |
264 | return self._remove_template_parameter(config['feedWithQueryParams']) | |
8940c1c0 JMF |
265 | |
266 | def _real_extract(self, url): | |
267 | mobj = re.match(self._VALID_URL, url) | |
268 | mgid = mobj.group('mgid') | |
269 | return self._get_videos_info(mgid) | |
270 | ||
271 | ||
84db8181 | 272 | class MTVIE(MTVServicesInfoExtractor): |
5c541b2c JMF |
273 | _VALID_URL = r'''(?x)^https?:// |
274 | (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$| | |
275 | m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))''' | |
84db8181 JMF |
276 | |
277 | _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/' | |
278 | ||
279 | _TESTS = [ | |
280 | { | |
32dac694 | 281 | 'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml', |
32dac694 PH |
282 | 'md5': '850f3f143316b1e71fa56a4edfd6e0f8', |
283 | 'info_dict': { | |
ca0f500e PH |
284 | 'id': '853555', |
285 | 'ext': 'mp4', | |
32dac694 PH |
286 | 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"', |
287 | 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.', | |
712c7530 YCH |
288 | 'timestamp': 1352610000, |
289 | 'upload_date': '20121111', | |
84db8181 JMF |
290 | }, |
291 | }, | |
84db8181 JMF |
292 | ] |
293 | ||
294 | def _get_thumbnail_url(self, uri, itemdoc): | |
295 | return 'http://mtv.mtvnimages.com/uri/' + uri | |
296 | ||
fc287219 PH |
297 | def _real_extract(self, url): |
298 | mobj = re.match(self._VALID_URL, url) | |
fc287219 | 299 | video_id = mobj.group('videoid') |
c801b205 | 300 | uri = mobj.groupdict().get('mgid') |
5c541b2c JMF |
301 | if uri is None: |
302 | webpage = self._download_webpage(url, video_id) | |
5f6a1245 | 303 | |
5c541b2c | 304 | # Some videos come from Vevo.com |
ca0f500e PH |
305 | m_vevo = re.search( |
306 | r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage) | |
5c541b2c | 307 | if m_vevo: |
8bcc8756 | 308 | vevo_id = m_vevo.group(1) |
32dac694 | 309 | self.to_screen('Vevo video detected: %s' % vevo_id) |
5c541b2c | 310 | return self.url_result('vevo:%s' % vevo_id, ie='Vevo') |
5f6a1245 | 311 | |
32dac694 | 312 | uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri') |
f7e02595 | 313 | return self._get_videos_info(uri) |
bc4ba05f JMF |
314 | |
315 | ||
071c1013 PH |
316 | class MTVDEIE(MTVServicesInfoExtractor): |
317 | IE_NAME = 'mtv.de' | |
65488b82 | 318 | _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$' |
79fa9db0 S |
319 | _TESTS = [{ |
320 | 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum', | |
321 | 'info_dict': { | |
322 | 'id': 'music_video-a50bc5f0b3aa4b3190aa', | |
712c7530 | 323 | 'ext': 'flv', |
79fa9db0 S |
324 | 'title': 'MusicVideo_cro-traum', |
325 | 'description': 'Cro - Traum', | |
071c1013 | 326 | }, |
79fa9db0 S |
327 | 'params': { |
328 | # rtmp download | |
329 | 'skip_download': True, | |
330 | }, | |
35f6e0ff | 331 | 'skip': 'Blocked at Travis CI', |
c3c9f879 S |
332 | }, { |
333 | # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97) | |
334 | 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen', | |
335 | 'info_dict': { | |
336 | 'id': 'local_playlist-f5ae778b9832cc837189', | |
712c7530 | 337 | 'ext': 'flv', |
c3c9f879 S |
338 | 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1', |
339 | }, | |
340 | 'params': { | |
341 | # rtmp download | |
342 | 'skip_download': True, | |
343 | }, | |
35f6e0ff | 344 | 'skip': 'Blocked at Travis CI', |
65488b82 | 345 | }, { |
65488b82 S |
346 | 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3', |
347 | 'info_dict': { | |
348 | 'id': 'local_playlist-4e760566473c4c8c5344', | |
349 | 'ext': 'mp4', | |
350 | 'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1', | |
351 | 'description': 'MTV Movies Supercut', | |
352 | }, | |
353 | 'params': { | |
354 | # rtmp download | |
355 | 'skip_download': True, | |
356 | }, | |
712c7530 | 357 | 'skip': 'Das Video kann zur Zeit nicht abgespielt werden.', |
79fa9db0 | 358 | }] |
071c1013 PH |
359 | |
360 | def _real_extract(self, url): | |
79fa9db0 S |
361 | video_id = self._match_id(url) |
362 | ||
363 | webpage = self._download_webpage(url, video_id) | |
364 | ||
365 | playlist = self._parse_json( | |
366 | self._search_regex( | |
367 | r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'), | |
368 | video_id) | |
071c1013 | 369 | |
712c7530 YCH |
370 | def _mrss_url(item): |
371 | return item['mrss'] + item.get('mrssvars', '') | |
372 | ||
65488b82 S |
373 | # news pages contain single video in playlist with different id |
374 | if len(playlist) == 1: | |
712c7530 | 375 | return self._get_videos_info_from_url(_mrss_url(playlist[0]), video_id) |
65488b82 | 376 | |
071c1013 | 377 | for item in playlist: |
79fa9db0 S |
378 | item_id = item.get('id') |
379 | if item_id and compat_str(item_id) == video_id: | |
712c7530 | 380 | return self._get_videos_info_from_url(_mrss_url(item), video_id) |