InstagramIE: fix the extraction of the uploader_id and the title

The page title is now 'Instagram', so we build it.
Also extract the description
This commit is contained in:
Jaime Marquínez Ferrándiz 2013-07-18 13:12:27 +02:00
parent f631c3311a
commit 3f40217704

View file

@ -10,7 +10,8 @@ class InstagramIE(InfoExtractor):
u'md5': u'0d2da106a9d2631273e192b372806516', u'md5': u'0d2da106a9d2631273e192b372806516',
u'info_dict': { u'info_dict': {
u"uploader_id": u"naomipq", u"uploader_id": u"naomipq",
u"title": u"Video by naomipq" u"title": u"Video by naomipq",
u'description': u'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
} }
} }
@ -18,20 +19,17 @@ class InstagramIE(InfoExtractor):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group(1) video_id = mobj.group(1)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
html_title = self._html_search_regex( uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
r'<title>(.+?)</title>', webpage, u'uploader id', fatal=False)
webpage, u'title', flags=re.DOTALL) desc = self._search_regex(r'"caption":"(.*?)"', webpage, u'description',
title = re.sub(u'(?: *\(Videos?\))? \u2022 Instagram$', '', html_title).strip() fatal=False)
uploader_id = self._html_search_regex(
r'<div class="media-user" id="media_user">.*?<h2><a href="[^"]*">([^<]*)</a></h2>',
webpage, u'uploader id', fatal=False, flags=re.DOTALL)
ext = 'mp4'
return [{ return [{
'id': video_id, 'id': video_id,
'url': self._og_search_video_url(webpage), 'url': self._og_search_video_url(webpage),
'ext': ext, 'ext': 'mp4',
'title': title, 'title': u'Video by %s' % uploader_id,
'thumbnail': self._og_search_thumbnail(webpage), 'thumbnail': self._og_search_thumbnail(webpage),
'uploader_id' : uploader_id 'uploader_id' : uploader_id,
'description': desc,
}] }]