X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/03342304420e5daeb428ffdcc7bbd2bbfecfa61a..83c8b40add31fb501d587b64fb17a936ffd5dbc5:/youtube_dl/extractor/mtv.py diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index ed06836..0335191 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -4,8 +4,8 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, compat_str, + compat_xpath, ) from ..utils import ( ExtractorError, @@ -13,10 +13,15 @@ from ..utils import ( fix_xml_ampersands, float_or_none, HEADRequest, + NO_DEFAULT, + RegexNotFoundError, sanitized_Request, + strip_or_none, + timeconvert, unescapeHTML, + update_url_query, url_basename, - RegexNotFoundError, + xpath_text, ) @@ -32,14 +37,19 @@ class MTVServicesInfoExtractor(InfoExtractor): def _id_from_uri(uri): return uri.split(':')[-1] - # This was originally implemented for ComedyCentral, but it also works here @staticmethod - def _transform_rtmp_url(rtmp_video_url): + def _remove_template_parameter(url): + # Remove the templates, like &device={device} + return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url) + + # This was originally implemented for ComedyCentral, but it also works here + @classmethod + def _transform_rtmp_url(cls, rtmp_video_url): m = re.match(r'^rtmpe?://.*?/(?Pgsp\..+?/.*)$', rtmp_video_url) if not m: - return rtmp_video_url + return {'rtmp': rtmp_video_url} base = 'http://viacommtvstrmfs.fplive.net/' - return base + m.group('finalid') + return {'http': base + m.group('finalid')} def _get_feed_url(self, uri): return self._FEED_URL @@ -83,13 +93,14 @@ class MTVServicesInfoExtractor(InfoExtractor): rtmp_video_url = rendition.find('./src').text if rtmp_video_url.endswith('siteunavail.png'): continue - formats.append({ - 'ext': ext, - 'url': self._transform_rtmp_url(rtmp_video_url), - 'format_id': rendition.get('bitrate'), + new_urls = self._transform_rtmp_url(rtmp_video_url) + formats.extend([{ + 'ext': 'flv' if new_url.startswith('rtmp') else ext, + 'url': new_url, + 'format_id': '-'.join(filter(None, [kind, rendition.get('bitrate')])), 'width': int(rendition.get('width')), 'height': int(rendition.get('height')), - }) + } for kind, new_url in new_urls.items()]) except (KeyError, TypeError): raise ExtractorError('Invalid rendition field.') self._sort_formats(formats) @@ -112,9 +123,7 @@ class MTVServicesInfoExtractor(InfoExtractor): video_id = self._id_from_uri(uri) self.report_extraction(video_id) content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))) - mediagen_url = content_el.attrib['url'] - # Remove the templates, like &device={device} - mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url) + mediagen_url = self._remove_template_parameter(content_el.attrib['url']) if 'acceptMethods' not in mediagen_url: mediagen_url += '&' if '?' in mediagen_url else '?' mediagen_url += 'acceptMethods=fms' @@ -130,11 +139,9 @@ class MTVServicesInfoExtractor(InfoExtractor): message += item.text raise ExtractorError(message, expected=True) - description_node = itemdoc.find('description') - if description_node is not None: - description = description_node.text.strip() - else: - description = None + description = strip_or_none(xpath_text(itemdoc, 'description')) + + timestamp = timeconvert(xpath_text(itemdoc, 'pubDate')) title_el = None if title_el is None: @@ -142,9 +149,9 @@ class MTVServicesInfoExtractor(InfoExtractor): itemdoc, './/{http://search.yahoo.com/mrss/}category', 'scheme', 'urn:mtvn:video_title') if title_el is None: - title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title') + title_el = itemdoc.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title')) if title_el is None: - title_el = itemdoc.find('.//title') or itemdoc.find('./title') + title_el = itemdoc.find(compat_xpath('.//title')) if title_el.text is None: title_el = None @@ -168,28 +175,34 @@ class MTVServicesInfoExtractor(InfoExtractor): 'thumbnail': self._get_thumbnail_url(uri, itemdoc), 'description': description, 'duration': float_or_none(content_el.attrib.get('duration')), + 'timestamp': timestamp, } def _get_feed_query(self, uri): data = {'uri': uri} if self._LANG: data['lang'] = self._LANG - return compat_urllib_parse.urlencode(data) + return data def _get_videos_info(self, uri): video_id = self._id_from_uri(uri) feed_url = self._get_feed_url(uri) - info_url = feed_url + '?' + self._get_feed_query(uri) + info_url = update_url_query(feed_url, self._get_feed_query(uri)) return self._get_videos_info_from_url(info_url, video_id) def _get_videos_info_from_url(self, url, video_id): idoc = self._download_xml( url, video_id, 'Downloading info', transform_source=fix_xml_ampersands) + + title = xpath_text(idoc, './channel/title') + description = xpath_text(idoc, './channel/description') + return self.playlist_result( - [self._get_video_info(item) for item in idoc.findall('.//item')]) + [self._get_video_info(item) for item in idoc.findall('.//item')], + playlist_title=title, playlist_description=description) - def _extract_mgid(self, webpage): + def _extract_mgid(self, webpage, default=NO_DEFAULT): try: # the url can be http://media.mtvnservices.com/fb/{mgid}.swf # or http://media.mtvnservices.com/{mgid} @@ -209,7 +222,7 @@ class MTVServicesInfoExtractor(InfoExtractor): sm4_embed = self._html_search_meta( 'sm4:video:embed', webpage, 'sm4 embed', default='') mgid = self._search_regex( - r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid') + r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid', default=default) return mgid def _real_extract(self, url): @@ -233,6 +246,8 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor): 'ext': 'mp4', 'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds', 'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.', + 'timestamp': 1400126400, + 'upload_date': '20140515', }, } @@ -245,13 +260,9 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor): def _get_feed_url(self, uri): video_id = self._id_from_uri(uri) - site_id = uri.replace(video_id, '') - config_url = ('http://media.mtvnservices.com/pmt/e1/players/{0}/' - 'context4/context5/config.xml'.format(site_id)) - config_doc = self._download_xml(config_url, video_id) - feed_node = config_doc.find('.//feed') - feed_url = feed_node.text.strip().split('?')[0] - return feed_url + config = self._download_json( + 'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id) + return self._remove_template_parameter(config['feedWithQueryParams']) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -260,6 +271,29 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor): class MTVIE(MTVServicesInfoExtractor): + IE_NAME = 'mtv' + _VALID_URL = r'https?://(?:www\.)?mtv\.com/(?:video-clips|full-episodes)/(?P[^/?#.]+)' + _FEED_URL = 'http://www.mtv.com/feeds/mrss/' + + _TESTS = [{ + 'url': 'http://www.mtv.com/video-clips/vl8qof/unlocking-the-truth-trailer', + 'md5': '1edbcdf1e7628e414a8c5dcebca3d32b', + 'info_dict': { + 'id': '5e14040d-18a4-47c4-a582-43ff602de88e', + 'ext': 'mp4', + 'title': 'Unlocking The Truth|July 18, 2016|1|101|Trailer', + 'description': '"Unlocking the Truth" premieres August 17th at 11/10c.', + 'timestamp': 1468846800, + 'upload_date': '20160718', + }, + }, { + 'url': 'http://www.mtv.com/full-episodes/94tujl/unlocking-the-truth-gates-of-hell-season-1-ep-101', + 'only_matching': True, + }] + + +class MTVVideoIE(MTVServicesInfoExtractor): + IE_NAME = 'mtv:video' _VALID_URL = r'''(?x)^https?:// (?:(?:www\.)?mtv\.com/videos/.+?/(?P[0-9]+)/[^/]+$| m\.mtv\.com/videos/video\.rbml\?.*?id=(?P[^&]+))''' @@ -275,6 +309,8 @@ class MTVIE(MTVServicesInfoExtractor): 'ext': 'mp4', 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"', 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.', + 'timestamp': 1352610000, + 'upload_date': '20121111', }, }, ] @@ -301,20 +337,6 @@ class MTVIE(MTVServicesInfoExtractor): return self._get_videos_info(uri) -class MTVIggyIE(MTVServicesInfoExtractor): - IE_NAME = 'mtviggy.com' - _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+' - _TEST = { - 'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/', - 'info_dict': { - 'id': '984696', - 'ext': 'mp4', - 'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet', - } - } - _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/' - - class MTVDEIE(MTVServicesInfoExtractor): IE_NAME = 'mtv.de' _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P\d+)-[^/#?]+/*(?:[#?].*)?$' @@ -322,7 +344,7 @@ class MTVDEIE(MTVServicesInfoExtractor): 'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum', 'info_dict': { 'id': 'music_video-a50bc5f0b3aa4b3190aa', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'MusicVideo_cro-traum', 'description': 'Cro - Traum', }, @@ -330,20 +352,21 @@ class MTVDEIE(MTVServicesInfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'Blocked at Travis CI', }, { # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97) 'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen', 'info_dict': { 'id': 'local_playlist-f5ae778b9832cc837189', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1', }, 'params': { # rtmp download 'skip_download': True, }, + 'skip': 'Blocked at Travis CI', }, { - # single video in pagePlaylist with different id 'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3', 'info_dict': { 'id': 'local_playlist-4e760566473c4c8c5344', @@ -355,6 +378,7 @@ class MTVDEIE(MTVServicesInfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'Das Video kann zur Zeit nicht abgespielt werden.', }] def _real_extract(self, url): @@ -367,11 +391,14 @@ class MTVDEIE(MTVServicesInfoExtractor): r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'), video_id) + def _mrss_url(item): + return item['mrss'] + item.get('mrssvars', '') + # news pages contain single video in playlist with different id if len(playlist) == 1: - return self._get_videos_info_from_url(playlist[0]['mrss'], video_id) + return self._get_videos_info_from_url(_mrss_url(playlist[0]), video_id) for item in playlist: item_id = item.get('id') if item_id and compat_str(item_id) == video_id: - return self._get_videos_info_from_url(item['mrss'], video_id) + return self._get_videos_info_from_url(_mrss_url(item), video_id)