diff options
Diffstat (limited to 'youtube_dl/extractor/generic.py')
| -rw-r--r-- | youtube_dl/extractor/generic.py | 84 |
1 files changed, 60 insertions, 24 deletions
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 69e0a7bd2..37671430a 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -25,7 +25,7 @@ class GenericIE(InfoExtractor): { u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', u'file': u'13601338388002.mp4', - u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89', + u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd', u'info_dict': { u"uploader": u"www.hodiho.fr", u"title": u"R\u00e9gis plante sa Jeep" @@ -33,6 +33,7 @@ class GenericIE(InfoExtractor): }, # embedded vimeo video { + u'add_ie': ['Vimeo'], u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references', u'file': u'22444065.mp4', u'md5': u'2903896e23df39722c33f015af0666e2', @@ -41,7 +42,35 @@ class GenericIE(InfoExtractor): u"uploader_id": u"skillsmatter", u"uploader": u"Skills Matter", } - } + }, + # bandcamp page with custom domain + { + u'add_ie': ['Bandcamp'], + u'url': u'http://bronyrock.com/track/the-pony-mash', + u'file': u'3235767654.mp3', + u'info_dict': { + u'title': u'The Pony Mash', + u'uploader': u'M_Pallante', + }, + u'skip': u'There is a limit of 200 free downloads / month for the test song', + }, + # embedded brightcove video + # it also tests brightcove videos that need to set the 'Referer' in the + # http requests + { + u'add_ie': ['Brightcove'], + u'url': u'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', + u'info_dict': { + u'id': u'2765128793001', + u'ext': u'mp4', + u'title': u'Le cours de bourse : l’analyse technique', + u'description': u'md5:7e9ad046e968cb2d1114004aba466fd9', + u'uploader': u'BFM BUSINESS', + }, + u'params': { + u'skip_download': True, + }, + }, ] def report_download_webpage(self, video_id): @@ -133,11 +162,20 @@ class GenericIE(InfoExtractor): raise ExtractorError(u'Failed to download URL: %s' % url) self.report_extraction(video_id) + + # it's tempting to parse this further, but you would + # have to take into account all the variations like + # Video Title - Site Name + # Site Name | Video Title + # Video Title - Tagline | Site Name + # and so on and so forth; it's just not practical + video_title = self._html_search_regex(r'<title>(.*)</title>', + webpage, u'video title', default=u'video', flags=re.DOTALL) + # Look for BrightCove: - m_brightcove = re.search(r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL) - if m_brightcove is not None: + bc_url = BrightcoveIE._extract_brightcove_url(webpage) + if bc_url is not None: self.to_screen(u'Brightcove video detected.') - bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group()) return self.url_result(bc_url, 'Brightcove') # Look for embedded Vimeo player @@ -149,11 +187,20 @@ class GenericIE(InfoExtractor): return self.url_result(surl, 'Vimeo') # Look for embedded YouTube player - mobj = re.search( - r'<iframe[^>]+?src="(https?://(?:www\.)?youtube.com/embed/.+?)"', webpage) - if mobj: - surl = unescapeHTML(mobj.group(1)) - return self.url_result(surl, 'Youtube') + matches = re.findall( + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube.com/embed/.+?)\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) + + # Look for Bandcamp pages with custom domain + mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) + if mobj is not None: + burl = unescapeHTML(mobj.group(1)) + # Don't set the extractor because it can be a track url or an album + return self.url_result(burl) # Start with something easy: JW Player in SWFObject mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage) @@ -162,7 +209,7 @@ class GenericIE(InfoExtractor): mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage) if mobj is None: # Broaden the search a little bit: JWPlayer JS loader - mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"&]*)', webpage) + mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"]*)', webpage) if mobj is None: # Try to find twitter cards info mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage) @@ -189,27 +236,16 @@ class GenericIE(InfoExtractor): video_id = compat_urllib_parse.unquote(os.path.basename(video_url)) # here's a fun little line of code for you: - video_extension = os.path.splitext(video_id)[1][1:] video_id = os.path.splitext(video_id)[0] - # it's tempting to parse this further, but you would - # have to take into account all the variations like - # Video Title - Site Name - # Site Name | Video Title - # Video Title - Tagline | Site Name - # and so on and so forth; it's just not practical - video_title = self._html_search_regex(r'<title>(.*)</title>', - webpage, u'video title', default=u'video', flags=re.DOTALL) - # video uploader is domain name video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*', url, u'video uploader') - return [{ + return { 'id': video_id, 'url': video_url, 'uploader': video_uploader, 'upload_date': None, 'title': video_title, - 'ext': video_extension, - }] + } |