X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/e76f531201cd41dfc0ce00be28bcc5c575c7acc5..e2cfcb2a7ff312228a25442500a885832a1829af:/youtube_dl/extractor/mtv.py diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 302c9bf..2f45568 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -4,18 +4,23 @@ import re from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, - compat_urllib_request, + compat_urllib_parse_urlencode, compat_str, + compat_xpath, ) from ..utils import ( ExtractorError, find_xpath_attr, fix_xml_ampersands, + float_or_none, HEADRequest, + sanitized_Request, + strip_or_none, + timeconvert, unescapeHTML, url_basename, RegexNotFoundError, + xpath_text, ) @@ -32,13 +37,13 @@ class MTVServicesInfoExtractor(InfoExtractor): return uri.split(':')[-1] # This was originally implemented for ComedyCentral, but it also works here - @staticmethod - def _transform_rtmp_url(rtmp_video_url): + @classmethod + def _transform_rtmp_url(cls, rtmp_video_url): m = re.match(r'^rtmpe?://.*?/(?Pgsp\..+?/.*)$', rtmp_video_url) if not m: - return rtmp_video_url + return {'rtmp': rtmp_video_url} base = 'http://viacommtvstrmfs.fplive.net/' - return base + m.group('finalid') + return {'http': base + m.group('finalid')} def _get_feed_url(self, uri): return self._FEED_URL @@ -53,7 +58,7 @@ class MTVServicesInfoExtractor(InfoExtractor): def _extract_mobile_video_formats(self, mtvn_id): webpage_url = self._MOBILE_TEMPLATE % mtvn_id - req = compat_urllib_request.Request(webpage_url) + req = sanitized_Request(webpage_url) # Otherwise we get a webpage that would execute some javascript req.add_header('User-Agent', 'curl/7') webpage = self._download_webpage(req, mtvn_id, @@ -82,13 +87,14 @@ class MTVServicesInfoExtractor(InfoExtractor): rtmp_video_url = rendition.find('./src').text if rtmp_video_url.endswith('siteunavail.png'): continue - formats.append({ - 'ext': ext, - 'url': self._transform_rtmp_url(rtmp_video_url), - 'format_id': rendition.get('bitrate'), + new_urls = self._transform_rtmp_url(rtmp_video_url) + formats.extend([{ + 'ext': 'flv' if new_url.startswith('rtmp') else ext, + 'url': new_url, + 'format_id': '-'.join(filter(None, [kind, rendition.get('bitrate')])), 'width': int(rendition.get('width')), 'height': int(rendition.get('height')), - }) + } for kind, new_url in new_urls.items()]) except (KeyError, TypeError): raise ExtractorError('Invalid rendition field.') self._sort_formats(formats) @@ -110,7 +116,8 @@ class MTVServicesInfoExtractor(InfoExtractor): uri = itemdoc.find('guid').text video_id = self._id_from_uri(uri) self.report_extraction(video_id) - mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url'] + content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))) + mediagen_url = content_el.attrib['url'] # Remove the templates, like &device={device} mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url) if 'acceptMethods' not in mediagen_url: @@ -128,11 +135,9 @@ 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: @@ -140,9 +145,9 @@ 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 @@ -165,28 +170,35 @@ class MTVServicesInfoExtractor(InfoExtractor): 'id': video_id, '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) + def _get_videos_info(self, uri): video_id = self._id_from_uri(uri) feed_url = self._get_feed_url(uri) - data = compat_urllib_parse.urlencode({'uri': uri}) - info_url = feed_url + '?' - if self._LANG: - info_url += 'lang=%s&' % self._LANG - info_url += data + info_url = 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 _real_extract(self, url): - title = url_basename(url) - webpage = self._download_webpage(url, title) + def _extract_mgid(self, webpage): try: # the url can be http://media.mtvnservices.com/fb/{mgid}.swf # or http://media.mtvnservices.com/{mgid} @@ -207,7 +219,12 @@ class MTVServicesInfoExtractor(InfoExtractor): 'sm4:video:embed', webpage, 'sm4 embed', default='') mgid = self._search_regex( r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid') + return mgid + def _real_extract(self, url): + title = url_basename(url) + webpage = self._download_webpage(url, title) + mgid = self._extract_mgid(webpage) videos_info = self._get_videos_info(mgid) return videos_info @@ -225,6 +242,8 @@ 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', }, } @@ -267,6 +286,8 @@ 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', }, }, ] @@ -293,20 +314,6 @@ 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+)-[^/#?]+/*(?:[#?].*)?$' @@ -314,7 +321,7 @@ 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', }, @@ -322,20 +329,21 @@ 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', @@ -347,6 +355,7 @@ class MTVDEIE(MTVServicesInfoExtractor): # rtmp download 'skip_download': True, }, + 'skip': 'Das Video kann zur Zeit nicht abgespielt werden.', }] def _real_extract(self, url): @@ -359,11 +368,14 @@ 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)