X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/509eeaaa67b2f901752c50fc6edf41954dbe3085..a8b640ed81fd63ade1773f4aa2f7808bc80175dd:/youtube_dl/extractor/generic.py diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 3105b47..a27ec34 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -8,24 +8,27 @@ import re from .common import InfoExtractor from .youtube import YoutubeIE from ..utils import ( - compat_urllib_error, compat_urllib_parse, - compat_urllib_request, compat_urlparse, compat_xml_parse_error, + determine_ext, ExtractorError, + float_or_none, HEADRequest, + orderedSet, parse_xml, smuggle_url, unescapeHTML, unified_strdate, + unsmuggle_url, url_basename, ) from .brightcove import BrightcoveIE from .ooyala import OoyalaIE from .rutv import RUTVIE from .smotri import SmotriIE +from .condenast import CondeNastIE class GenericIE(InfoExtractor): @@ -153,7 +156,6 @@ class GenericIE(InfoExtractor): # funnyordie embed { 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns', - 'md5': '7cf780be104d40fea7bae52eed4a470e', 'info_dict': { 'id': '18e820ec3f', 'ext': 'mp4', @@ -178,13 +180,13 @@ class GenericIE(InfoExtractor): # Embedded TED video { 'url': 'http://en.support.wordpress.com/videos/ted-talks/', - 'md5': 'deeeabcc1085eb2ba205474e7235a3d5', + 'md5': '65fdff94098e4a607385a60c5177c638', 'info_dict': { - 'id': '981', + 'id': '1969', 'ext': 'mp4', - 'title': 'My web playroom', - 'uploader': 'Ze Frank', - 'description': 'md5:ddb2a40ecd6b6a147e400e535874947b', + 'title': 'Hidden miracles of the natural world', + 'uploader': 'Louie Schwartzberg', + 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9', } }, # Embeded Ustream video @@ -224,21 +226,6 @@ class GenericIE(InfoExtractor): 'skip_download': 'Requires rtmpdump' } }, - # smotri embed - { - 'url': 'http://rbctv.rbc.ru/archive/news/562949990879132.shtml', - 'md5': 'ec40048448e9284c9a1de77bb188108b', - 'info_dict': { - 'id': 'v27008541fad', - 'ext': 'mp4', - 'title': 'Крым и Севастополь вошли в состав России', - 'description': 'md5:fae01b61f68984c7bd2fa741e11c3175', - 'duration': 900, - 'upload_date': '20140318', - 'uploader': 'rbctv_2012_4', - 'uploader_id': 'rbctv_2012_4', - }, - }, # Condé Nast embed { 'url': 'http://www.wired.com/2014/04/honda-asimo/', @@ -278,70 +265,165 @@ class GenericIE(InfoExtractor): 'skip_download': True, } }, + # MTVSercices embed + { + 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too', + 'md5': '35727f82f58c76d996fc188f9755b0d5', + 'info_dict': { + 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9', + 'ext': 'mp4', + 'title': 'Review', + 'description': 'Mario\'s life in the fast lane has never looked so good.', + }, + }, + # YouTube embed via + { + 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM', + 'info_dict': { + 'id': '4vAffPZIT44', + 'ext': 'mp4', + 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!', + 'uploader': 'Gameloft', + 'uploader_id': 'gameloft', + 'upload_date': '20140828', + 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4', + }, + 'params': { + 'skip_download': True, + } + }, + # Camtasia studio + { + 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/', + 'playlist': [{ + 'md5': '0c5e352edabf715d762b0ad4e6d9ee67', + 'info_dict': { + 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final', + 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1', + 'ext': 'flv', + 'duration': 2235.90, + } + }, { + 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63', + 'info_dict': { + 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP', + 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip', + 'ext': 'flv', + 'duration': 2235.93, + } + }], + 'info_dict': { + 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final', + } + }, + # Flowplayer + { + 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html', + 'md5': '9d65602bf31c6e20014319c7d07fba27', + 'info_dict': { + 'id': '5123ea6d5e5a7', + 'ext': 'mp4', + 'age_limit': 18, + 'uploader': 'www.handjobhub.com', + 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com', + } + }, + # RSS feed + { + 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml', + 'info_dict': { + 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml', + 'title': 'Zero Punctuation', + 'description': 're:' + }, + 'playlist_mincount': 11, + }, + # Multiple brightcove videos + # https://github.com/rg3/youtube-dl/issues/2283 + { + 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html', + 'info_dict': { + 'id': 'always-never', + 'title': 'Always / Never - The New Yorker', + }, + 'playlist_count': 3, + 'params': { + 'extract_flat': False, + 'skip_download': True, + } + }, + # MLB embed + { + 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/', + 'md5': '96f09a37e44da40dd083e12d9a683327', + 'info_dict': { + 'id': '33322633', + 'ext': 'mp4', + 'title': 'Ump changes call to ball', + 'description': 'md5:71c11215384298a172a6dcb4c2e20685', + 'duration': 48, + 'timestamp': 1401537900, + 'upload_date': '20140531', + 'thumbnail': 're:^https?://.*\.jpg$', + }, + }, + # Wistia embed + { + 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson', + 'md5': '8788b683c777a5cf25621eaf286d0c23', + 'info_dict': { + 'id': '1cfaf6b7ea', + 'ext': 'mov', + 'title': 'md5:51364a8d3d009997ba99656004b5e20d', + 'duration': 643.0, + 'filesize': 182808282, + 'uploader': 'education-portal.com', + }, + }, + { + 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz', + 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4', + 'info_dict': { + 'id': 'uxjb0lwrcz', + 'ext': 'mp4', + 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks', + 'duration': 1715.0, + 'uploader': 'thoughtworks.wistia.com', + }, + }, + # Direct download with broken HEAD + { + 'url': 'http://ai-radio.org:8000/radio.opus', + 'info_dict': { + 'id': 'radio', + 'ext': 'opus', + 'title': 'radio', + }, + 'params': { + 'skip_download': True, # infinite live stream + }, + 'expected_warnings': [ + r'501.*Not Implemented' + ], + }, + # Soundcloud embed + { + 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/', + 'info_dict': { + 'id': '174391317', + 'ext': 'mp3', + 'description': 'md5:ff867d6b555488ad3c52572bb33d432c', + 'uploader': 'Sophos Security', + 'title': 'Chet Chat 171 - Oct 29, 2014', + 'upload_date': '20141029', + } + } ] - def report_download_webpage(self, video_id): - """Report webpage download.""" - if not self._downloader.params.get('test', False): - self._downloader.report_warning('Falling back on generic information extractor.') - super(GenericIE, self).report_download_webpage(video_id) - def report_following_redirect(self, new_url): """Report information extraction.""" self._downloader.to_screen('[redirect] Following redirect to %s' % new_url) - def _send_head(self, url): - """Check if it is a redirect, like url shorteners, in case return the new url.""" - - class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): - """ - Subclass the HTTPRedirectHandler to make it use our - HEADRequest also on the redirected URL - """ - def redirect_request(self, req, fp, code, msg, headers, newurl): - if code in (301, 302, 303, 307): - newurl = newurl.replace(' ', '%20') - newheaders = dict((k,v) for k,v in req.headers.items() - if k.lower() not in ("content-length", "content-type")) - try: - # This function was deprecated in python 3.3 and removed in 3.4 - origin_req_host = req.get_origin_req_host() - except AttributeError: - origin_req_host = req.origin_req_host - return HEADRequest(newurl, - headers=newheaders, - origin_req_host=origin_req_host, - unverifiable=True) - else: - raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp) - - class HTTPMethodFallback(compat_urllib_request.BaseHandler): - """ - Fallback to GET if HEAD is not allowed (405 HTTP error) - """ - def http_error_405(self, req, fp, code, msg, headers): - fp.read() - fp.close() - - newheaders = dict((k,v) for k,v in req.headers.items() - if k.lower() not in ("content-length", "content-type")) - return self.parent.open(compat_urllib_request.Request(req.get_full_url(), - headers=newheaders, - origin_req_host=req.get_origin_req_host(), - unverifiable=True)) - - # Build our opener - opener = compat_urllib_request.OpenerDirector() - for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler, - HTTPMethodFallback, HEADRedirectHandler, - compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: - opener.add_handler(handler()) - - response = opener.open(HEADRequest(url)) - if response is None: - raise ExtractorError('Invalid URL protocol') - return response - def _extract_rss(self, url, video_id, doc): playlist_title = doc.find('./channel/title').text playlist_desc_el = doc.find('./channel/description') @@ -361,6 +443,43 @@ class GenericIE(InfoExtractor): 'entries': entries, } + def _extract_camtasia(self, url, video_id, webpage): + """ Returns None if no camtasia video can be found. """ + + camtasia_cfg = self._search_regex( + r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);', + webpage, 'camtasia configuration file', default=None) + if camtasia_cfg is None: + return None + + title = self._html_search_meta('DC.title', webpage, fatal=True) + + camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg) + camtasia_cfg = self._download_xml( + camtasia_url, video_id, + note='Downloading camtasia configuration', + errnote='Failed to download camtasia configuration') + fileset_node = camtasia_cfg.find('./playlist/array/fileset') + + entries = [] + for n in fileset_node.getchildren(): + url_n = n.find('./uri') + if url_n is None: + continue + + entries.append({ + 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0], + 'title': '%s - %s' % (title, n.tag), + 'url': compat_urlparse.urljoin(url, url_n.text), + 'duration': float_or_none(n.find('./duration').text), + }) + + return { + '_type': 'playlist', + 'entries': entries, + 'title': title, + } + def _real_extract(self, url): if url.startswith('//'): return { @@ -372,13 +491,13 @@ class GenericIE(InfoExtractor): if not parsed_url.scheme: default_search = self._downloader.params.get('default_search') if default_search is None: - default_search = 'auto_warning' + default_search = 'fixup_error' - if default_search in ('auto', 'auto_warning'): + if default_search in ('auto', 'auto_warning', 'fixup_error'): if '/' in url: self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') return self.url_result('http://' + url) - else: + elif default_search != 'fixup_error': if default_search == 'auto_warning': if re.match(r'^(?:url|URL)$', url): raise ExtractorError( @@ -386,53 +505,75 @@ class GenericIE(InfoExtractor): expected=True) else: self._downloader.report_warning( - 'Falling back to youtube search for %s . Set --default-search to "auto" to suppress this warning.' % url) + 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url) return self.url_result('ytsearch:' + url) + + if default_search in ('error', 'fixup_error'): + raise ExtractorError( + ('%r is not a valid URL. ' + 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube' + ) % (url, url), expected=True) else: - assert ':' in default_search + if ':' not in default_search: + default_search += ':' return self.url_result(default_search + url) - video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0] + + url, smuggled_data = unsmuggle_url(url) + force_videoid = None + is_intentional = smuggled_data and smuggled_data.get('to_generic') + if smuggled_data and 'force_videoid' in smuggled_data: + force_videoid = smuggled_data['force_videoid'] + video_id = force_videoid + else: + video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0] self.to_screen('%s: Requesting header' % video_id) - try: - response = self._send_head(url) + head_req = HEADRequest(url) + head_response = self._request_webpage( + head_req, video_id, + note=False, errnote='Could not send HEAD request to %s' % url, + fatal=False) + if head_response is not False: # Check for redirect - new_url = response.geturl() + new_url = head_response.geturl() if url != new_url: self.report_following_redirect(new_url) + if force_videoid: + new_url = smuggle_url( + new_url, {'force_videoid': force_videoid}) return self.url_result(new_url) - # Check for direct link to a video - content_type = response.headers.get('Content-Type', '') - m = re.match(r'^(?Paudio|video|application(?=/ogg$))/(?P.+)$', content_type) - if m: - upload_date = response.headers.get('Last-Modified') - if upload_date: - upload_date = unified_strdate(upload_date) - return { - 'id': video_id, - 'title': os.path.splitext(url_basename(url))[0], - 'formats': [{ - 'format_id': m.group('format_id'), - 'url': url, - 'vcodec': 'none' if m.group('type') == 'audio' else None - }], - 'upload_date': upload_date, - } + full_response = None + if head_response is False: + full_response = self._request_webpage(url, video_id) + head_response = full_response + + # Check for direct link to a video + content_type = head_response.headers.get('Content-Type', '') + m = re.match(r'^(?Paudio|video|application(?=/ogg$))/(?P.+)$', content_type) + if m: + upload_date = unified_strdate( + head_response.headers.get('Last-Modified')) + return { + 'id': video_id, + 'title': os.path.splitext(url_basename(url))[0], + 'formats': [{ + 'format_id': m.group('format_id'), + 'url': url, + 'vcodec': 'none' if m.group('type') == 'audio' else None + }], + 'upload_date': upload_date, + } - except compat_urllib_error.HTTPError: - # This may be a stupid server that doesn't like HEAD, our UA, or so - pass + if not self._downloader.params.get('test', False) and not is_intentional: + self._downloader.report_warning('Falling back on generic information extractor.') - try: + if full_response: + webpage = self._webpage_read_content(full_response, url, video_id) + else: webpage = self._download_webpage(url, video_id) - except ValueError: - # since this is the last-resort InfoExtractor, if - # this error is thrown, it'll be thrown here - raise ExtractorError('Failed to download URL: %s' % url) - self.report_extraction(video_id) # Is it an RSS feed? @@ -443,6 +584,11 @@ class GenericIE(InfoExtractor): except compat_xml_parse_error: pass + # Is it a Camtasia project? + camtasia_res = self._extract_camtasia(url, video_id, webpage) + if camtasia_res is not None: + return camtasia_res + # Sometimes embedded video player is hidden behind percent encoding # (e.g. https://github.com/rg3/youtube-dl/issues/2448) # Unescaping the whole page allows to handle those cases in a generic way @@ -458,10 +604,28 @@ class GenericIE(InfoExtractor): r'(?s)(.*?)', webpage, 'video title', default='video') + # Try to detect age limit automatically + age_limit = self._rta_search(webpage) + # And then there are the jokers who advertise that they use RTA, + # but actually don't. + AGE_LIMIT_MARKERS = [ + r'Proudly Labeled RTA', + ] + if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS): + age_limit = 18 + # video uploader is domain name video_uploader = self._search_regex( r'^(?:https?://)?([^/]*)/.*', url, 'video uploader') + # Helper method + def _playlist_from_matches(matches, getter, ie=None): + urlrs = orderedSet( + self.url_result(self._proto_relative_url(getter(m)), ie) + for m in matches) + return self.playlist_result( + urlrs, playlist_id=video_id, playlist_title=video_title) + # Look for BrightCove: bc_urls = BrightcoveIE._extract_brightcove_urls(webpage) if bc_urls: @@ -485,58 +649,79 @@ class GenericIE(InfoExtractor): if mobj: player_url = unescapeHTML(mobj.group('url')) surl = smuggle_url(player_url, {'Referer': url}) - return self.url_result(surl, 'Vimeo') + return self.url_result(surl) # Look for embedded (swf embed) Vimeo player mobj = re.search( - r']+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) + r']+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) if mobj: - return self.url_result(mobj.group(1), 'Vimeo') + return self.url_result(mobj.group(1)) # Look for embedded YouTube player matches = re.findall(r'''(?x) (?: ]+?src=| + data-video-url=| ]+?src=| - embedSWF\(?:\s* + embedSWF\(?:\s*| + new\s+SWFObject\( ) (["\']) - (?P(?:https?:)?//(?:www\.)?youtube\.com/ - (?:embed|v)/.+?) + (?P(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/ + (?:embed|v|p)/.+?) \1''', webpage) if matches: - urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') - for tuppl in matches] - return self.playlist_result( - urlrs, playlist_id=video_id, playlist_title=video_title) + return _playlist_from_matches( + matches, lambda m: unescapeHTML(m[1])) # Look for embedded Dailymotion player matches = re.findall( r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage) if matches: - urlrs = [self.url_result(unescapeHTML(tuppl[1])) - for tuppl in matches] - return self.playlist_result( - urlrs, playlist_id=video_id, playlist_title=video_title) + return _playlist_from_matches( + matches, lambda m: unescapeHTML(m[1])) + + # Look for embedded Dailymotion playlist player (#3822) + m = re.search( + r']+?src=(["\'])(?P(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage) + if m: + playlists = re.findall( + r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url'))) + if playlists: + return _playlist_from_matches( + playlists, lambda p: '//dailymotion.com/playlist/%s' % p) # Look for embedded Wistia player match = re.search( - r']+?src=(["\'])(?P(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) + r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) if match: + embed_url = self._proto_relative_url( + unescapeHTML(match.group('url'))) return { '_type': 'url_transparent', - 'url': unescapeHTML(match.group('url')), + 'url': embed_url, 'ie_key': 'Wistia', 'uploader': video_uploader, 'title': video_title, 'id': video_id, } + + match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P[^"\']+)', webpage) + if match: + return { + '_type': 'url_transparent', + 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')), + 'ie_key': 'Wistia', + 'uploader': video_uploader, + 'title': video_title, + 'id': match.group('id') + } # Look for embedded blip.tv player mobj = re.search(r']*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage) if mobj: return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV') - mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage) + mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage) if mobj: return self.url_result(mobj.group(1), 'BlipTV') @@ -609,6 +794,11 @@ class GenericIE(InfoExtractor): if mobj is not None: return self.url_result(mobj.group('url'), 'VK') + # Look for embedded ivi player + mobj = re.search(r']+?src=(["\'])(?Phttps?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage) + if mobj is not None: + return self.url_result(mobj.group('url'), 'Ivi') + # Look for embedded Huffington Post player mobj = re.search( r']+?src=(["\'])(?Phttps?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage) @@ -626,10 +816,8 @@ class GenericIE(InfoExtractor): # Look for funnyordie embed matches = re.findall(r']+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage) if matches: - urlrs = [self.url_result(unescapeHTML(eurl), 'FunnyOrDie') - for eurl in matches] - return self.playlist_result( - urlrs, playlist_id=video_id, playlist_title=video_title) + return _playlist_from_matches( + matches, getter=unescapeHTML, ie='FunnyOrDie') # Look for embedded RUTV player rutv_url = RUTVIE._extract_url(webpage) @@ -662,7 +850,7 @@ class GenericIE(InfoExtractor): # Look for embeded soundcloud player mobj = re.search( - r'