[generic] Detect bandcamp pages that use custom domains (closes #1662)

They embed the original url in the 'og:url' property.
This commit is contained in:
Jaime Marquínez Ferrándiz 2013-10-27 14:40:25 +01:00
parent bc63d9d329
commit c19f7764a5

View File

@ -41,7 +41,17 @@ class GenericIE(InfoExtractor):
u"uploader_id": u"skillsmatter",
u"uploader": u"Skills Matter",
}
}
},
# bandcamp page with custom domain
{
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',
},
]
def report_download_webpage(self, video_id):
@ -155,6 +165,12 @@ def _real_extract(self, url):
surl = unescapeHTML(mobj.group(1))
return self.url_result(surl, 'Youtube')
# 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))
return self.url_result(burl, 'Bandcamp')
# Start with something easy: JW Player in SWFObject
mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
if mobj is None: