InfoExtractor: add some helper methods to extract OpenGraph info

This commit is contained in:
Jaime Marquínez Ferrándiz 2013-07-12 19:00:19 +02:00
parent d8269e1dfb
commit 46720279c2
17 changed files with 54 additions and 96 deletions

View file

@ -257,6 +257,30 @@ class InfoExtractor(object):
return (username, password) return (username, password)
# Helper functions for extracting OpenGraph info
@staticmethod
def _og_regex(property):
return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % property
def _og_search_property(self, property, html, name=None, **kargs):
if name is None:
name = 'OpenGraph %s' % property
return self._html_search_regex(self._og_regex(property), html, name, **kargs)
def _og_search_thumbnail(self, html, **kargs):
return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs)
def _og_search_description(self, html, **kargs):
return self._og_search_property('description', html, fatal=False, **kargs)
def _og_search_title(self, html, **kargs):
return self._og_search_property('title', html, **kargs)
def _og_search_video_url(self, html, name='video url', **kargs):
return self._html_search_regex([self._og_regex('video:secure_url'),
self._og_regex('video')],
html, name, **kargs)
class SearchInfoExtractor(InfoExtractor): class SearchInfoExtractor(InfoExtractor):
""" """
Base class for paged search queries extractors. Base class for paged search queries extractors.

View file

@ -34,8 +34,6 @@ class CSpanIE(InfoExtractor):
description = self._html_search_regex(r'<meta (?:property="og:|name=")description" content="(.*?)"', description = self._html_search_regex(r'<meta (?:property="og:|name=")description" content="(.*?)"',
webpage, 'description', webpage, 'description',
flags=re.MULTILINE|re.DOTALL) flags=re.MULTILINE|re.DOTALL)
thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.*?)"',
webpage, 'thumbnail')
url = self._search_regex(r'<string name="URL">(.*?)</string>', url = self._search_regex(r'<string name="URL">(.*?)</string>',
video_info, 'video url') video_info, 'video url')
@ -49,5 +47,5 @@ class CSpanIE(InfoExtractor):
'url': url, 'url': url,
'play_path': path, 'play_path': path,
'description': description, 'description': description,
'thumbnail': thumbnail, 'thumbnail': self._og_search_thumbnail(webpage),
} }

View file

@ -39,9 +39,6 @@ class DailymotionIE(InfoExtractor):
# Extract URL, uploader and title from webpage # Extract URL, uploader and title from webpage
self.report_extraction(video_id) self.report_extraction(video_id)
video_title = self._html_search_regex(r'<meta property="og:title" content="(.*?)" />',
webpage, 'title')
video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
# Looking for official user # Looking for official user
r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'], r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
@ -76,7 +73,7 @@ class DailymotionIE(InfoExtractor):
'url': video_url, 'url': video_url,
'uploader': video_uploader, 'uploader': video_uploader,
'upload_date': video_upload_date, 'upload_date': video_upload_date,
'title': video_title, 'title': self._og_search_title(webpage),
'ext': video_extension, 'ext': video_extension,
'thumbnail': info['thumbnail_url'] 'thumbnail': info['thumbnail_url']
}] }]

View file

@ -28,14 +28,9 @@ class EHowIE(InfoExtractor):
video_url = self._search_regex(r'(?:file|source)=(http[^\'"&]*)', video_url = self._search_regex(r'(?:file|source)=(http[^\'"&]*)',
webpage, u'video URL') webpage, u'video URL')
final_url = compat_urllib_parse.unquote(video_url) final_url = compat_urllib_parse.unquote(video_url)
thumbnail_url = self._search_regex(r'<meta property="og:image" content="(.+?)" />',
webpage, u'thumbnail URL')
uploader = self._search_regex(r'<meta name="uploader" content="(.+?)" />', uploader = self._search_regex(r'<meta name="uploader" content="(.+?)" />',
webpage, u'uploader') webpage, u'uploader')
title = self._search_regex(r'<meta property="og:title" content="(.+?)" />', title = self._og_search_title(webpage).replace(' | eHow', '')
webpage, u'Video title').replace(' | eHow', '')
description = self._search_regex(r'<meta property="og:description" content="(.+?)" />',
webpage, u'video description')
ext = determine_ext(final_url) ext = determine_ext(final_url)
return { return {
@ -44,8 +39,8 @@ class EHowIE(InfoExtractor):
'url': final_url, 'url': final_url,
'ext': ext, 'ext': ext,
'title': title, 'title': title,
'thumbnail': thumbnail_url, 'thumbnail': self._og_search_thumbnail(webpage),
'description': description, 'description': self._og_search_description(webpage),
'uploader': uploader, 'uploader': uploader,
} }

View file

@ -36,11 +36,7 @@ class EscapistIE(InfoExtractor):
videoDesc = self._html_search_regex('<meta name="description" content="([^"]*)"', videoDesc = self._html_search_regex('<meta name="description" content="([^"]*)"',
webpage, u'description', fatal=False) webpage, u'description', fatal=False)
imgUrl = self._html_search_regex('<meta property="og:image" content="([^"]*)"', playerUrl = self._og_search_video_url(webpage, name='player url')
webpage, u'thumbnail', fatal=False)
playerUrl = self._html_search_regex('<meta property="og:video" content="([^"]*)"',
webpage, u'player url')
title = self._html_search_regex('<meta name="title" content="([^"]*)"', title = self._html_search_regex('<meta name="title" content="([^"]*)"',
webpage, u'player url').split(' : ')[-1] webpage, u'player url').split(' : ')[-1]
@ -70,7 +66,7 @@ class EscapistIE(InfoExtractor):
'upload_date': None, 'upload_date': None,
'title': title, 'title': title,
'ext': 'mp4', 'ext': 'mp4',
'thumbnail': imgUrl, 'thumbnail': self._og_search_thumbnail(webpage),
'description': videoDesc, 'description': videoDesc,
'player_url': playerUrl, 'player_url': playerUrl,
} }

View file

@ -47,21 +47,12 @@ class FlickrIE(InfoExtractor):
raise ExtractorError(u'Unable to extract video url') raise ExtractorError(u'Unable to extract video url')
video_url = mobj.group(1) + unescapeHTML(mobj.group(2)) video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
video_title = self._html_search_regex(r'<meta property="og:title" content=(?:"([^"]+)"|\'([^\']+)\')',
webpage, u'video title')
video_description = self._html_search_regex(r'<meta property="og:description" content=(?:"([^"]+)"|\'([^\']+)\')',
webpage, u'description', fatal=False)
thumbnail = self._html_search_regex(r'<meta property="og:image" content=(?:"([^"]+)"|\'([^\']+)\')',
webpage, u'thumbnail', fatal=False)
return [{ return [{
'id': video_id, 'id': video_id,
'url': video_url, 'url': video_url,
'ext': 'mp4', 'ext': 'mp4',
'title': video_title, 'title': self._og_search_title(webpage),
'description': video_description, 'description': self._og_search_description(webpage),
'thumbnail': thumbnail, 'thumbnail': self._og_search_thumbnail(webpage),
'uploader_id': video_uploader_id, 'uploader_id': video_uploader_id,
}] }]

View file

@ -27,14 +27,11 @@ class FunnyOrDieIE(InfoExtractor):
title = self._html_search_regex((r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>", title = self._html_search_regex((r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>",
r'<title>(?P<title>[^<]+?)</title>'), webpage, 'title', flags=re.DOTALL) r'<title>(?P<title>[^<]+?)</title>'), webpage, 'title', flags=re.DOTALL)
video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"',
webpage, u'description', fatal=False, flags=re.DOTALL)
info = { info = {
'id': video_id, 'id': video_id,
'url': video_url, 'url': video_url,
'ext': 'mp4', 'ext': 'mp4',
'title': title, 'title': title,
'description': video_description, 'description': self._og_search_description(webpage, flags=re.DOTALL),
} }
return [info] return [info]

View file

@ -33,16 +33,12 @@ class HotNewHipHopIE(InfoExtractor):
video_title = self._html_search_regex(r"<title>(.*)</title>", video_title = self._html_search_regex(r"<title>(.*)</title>",
webpage_src, u'title') webpage_src, u'title')
# Getting thumbnail and if not thumbnail sets correct title for WSHH candy video.
thumbnail = self._html_search_regex(r'"og:image" content="(.*)"',
webpage_src, u'thumbnail', fatal=False)
results = [{ results = [{
'id': video_id, 'id': video_id,
'url' : video_url, 'url' : video_url,
'title' : video_title, 'title' : video_title,
'thumbnail' : thumbnail, 'thumbnail' : self._og_search_thumbnail(webpage_src),
'ext' : 'mp3', 'ext' : 'mp3',
}] }]
return results return results

View file

@ -18,12 +18,6 @@ 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)
video_url = self._html_search_regex(
r'<meta property="og:video" content="(.+?)"',
webpage, u'video URL')
thumbnail_url = self._html_search_regex(
r'<meta property="og:image" content="(.+?)" />',
webpage, u'thumbnail URL', fatal=False)
html_title = self._html_search_regex( html_title = self._html_search_regex(
r'<title>(.+?)</title>', r'<title>(.+?)</title>',
webpage, u'title', flags=re.DOTALL) webpage, u'title', flags=re.DOTALL)
@ -34,9 +28,9 @@ class InstagramIE(InfoExtractor):
return [{ return [{
'id': video_id, 'id': video_id,
'url': video_url, 'url': self._og_search_video_url(webpage),
'ext': ext, 'ext': ext,
'title': title, 'title': title,
'thumbnail': thumbnail_url, 'thumbnail': self._og_search_thumbnail(webpage),
'uploader_id' : uploader_id 'uploader_id' : uploader_id
}] }]

View file

@ -24,8 +24,7 @@ class KeekIE(InfoExtractor):
thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"', video_title = self._og_search_title(webpage)
webpage, u'title')
uploader = self._html_search_regex(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>', uploader = self._html_search_regex(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>',
webpage, u'uploader', fatal=False) webpage, u'uploader', fatal=False)

View file

@ -33,11 +33,9 @@ class LiveLeakIE(InfoExtractor):
video_url = self._search_regex(r'file: "(.*?)",', video_url = self._search_regex(r'file: "(.*?)",',
webpage, u'video URL') webpage, u'video URL')
video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"', video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
webpage, u'title').replace('LiveLeak.com -', '').strip()
video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"', video_description = self._og_search_description(webpage)
webpage, u'description', fatal=False)
video_uploader = self._html_search_regex(r'By:.*?(\w+)</a>', video_uploader = self._html_search_regex(r'By:.*?(\w+)</a>',
webpage, u'uploader', fatal=False) webpage, u'uploader', fatal=False)

View file

@ -30,8 +30,7 @@ class NBAIE(InfoExtractor):
video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4' video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
shortened_video_id = video_id.rpartition('/')[2] shortened_video_id = video_id.rpartition('/')[2]
title = self._html_search_regex(r'<meta property="og:title" content="(.*?)"', title = self._og_search_title(webpage, default=shortened_video_id).replace('NBA.com: ', '')
webpage, 'title', default=shortened_video_id).replace('NBA.com: ', '')
# It isn't there in the HTML it returns to us # It isn't there in the HTML it returns to us
# uploader_date = self._html_search_regex(r'<b>Date:</b> (.*?)</div>', webpage, 'upload_date', fatal=False) # uploader_date = self._html_search_regex(r'<b>Date:</b> (.*?)</div>', webpage, 'upload_date', fatal=False)

View file

@ -18,12 +18,6 @@ class StatigramIE(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)
video_url = self._html_search_regex(
r'<meta property="og:video:secure_url" content="(.+?)">',
webpage, u'video URL')
thumbnail_url = self._html_search_regex(
r'<meta property="og:image" content="(.+?)" />',
webpage, u'thumbnail URL', fatal=False)
html_title = self._html_search_regex( html_title = self._html_search_regex(
r'<title>(.+?)</title>', r'<title>(.+?)</title>',
webpage, u'title') webpage, u'title')
@ -34,9 +28,9 @@ class StatigramIE(InfoExtractor):
return [{ return [{
'id': video_id, 'id': video_id,
'url': video_url, 'url': self._og_search_video_url(webpage),
'ext': ext, 'ext': ext,
'title': title, 'title': title,
'thumbnail': thumbnail_url, 'thumbnail': self._og_search_thumbnail(webpage),
'uploader_id' : uploader_id 'uploader_id' : uploader_id
}] }]

View file

@ -30,15 +30,6 @@ class TeamcocoIE(InfoExtractor):
self.report_extraction(video_id) self.report_extraction(video_id)
video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"',
webpage, u'title')
thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)"',
webpage, u'thumbnail', fatal=False)
video_description = self._html_search_regex(r'<meta property="og:description" content="(.*?)"',
webpage, u'description', fatal=False)
data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
data = self._download_webpage(data_url, video_id, 'Downloading data webpage') data = self._download_webpage(data_url, video_id, 'Downloading data webpage')
@ -49,7 +40,7 @@ class TeamcocoIE(InfoExtractor):
'id': video_id, 'id': video_id,
'url': video_url, 'url': video_url,
'ext': 'mp4', 'ext': 'mp4',
'title': video_title, 'title': self._og_search_title(webpage),
'thumbnail': thumbnail, 'thumbnail': self._og_search_thumbnail(webpage),
'description': video_description, 'description': self._og_search_description(webpage),
}] }]

View file

@ -24,11 +24,8 @@ class TrailerAddictIE(InfoExtractor):
webpage, 'video title').replace(' - Trailer Addict','') webpage, 'video title').replace(' - Trailer Addict','')
view_count = self._search_regex(r'Views: (.+?)<br />', view_count = self._search_regex(r'Views: (.+?)<br />',
webpage, 'Views Count') webpage, 'Views Count')
description = self._search_regex(r'<meta property="og:description" content="(.+?)" />', video_id = self._og_search_property('video', webpage, 'Video id').split('=')[1]
webpage, 'video description')
video_id = self._search_regex(r'<meta property="og:video" content="(.+?)" />',
webpage, 'Video id').split('=')[1]
info_url = "http://www.traileraddict.com/fvar.php?tid=%s" %(str(video_id)) info_url = "http://www.traileraddict.com/fvar.php?tid=%s" %(str(video_id))
info_webpage = self._download_webpage(info_url, video_id , "Downloading the info webpage") info_webpage = self._download_webpage(info_url, video_id , "Downloading the info webpage")
@ -44,6 +41,6 @@ class TrailerAddictIE(InfoExtractor):
'ext' : ext, 'ext' : ext,
'title' : title, 'title' : title,
'thumbnail' : thumbnail_url, 'thumbnail' : thumbnail_url,
'description' : description, 'description' : self._og_search_description(webpage),
'view_count' : view_count, 'view_count' : view_count,
}] }]

View file

@ -22,8 +22,6 @@ class TutvIE(InfoExtractor):
video_id = mobj.group('id') video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
title = self._html_search_regex(
r'<meta property="og:title" content="(.*?)">', webpage, u'title')
internal_id = self._search_regex(r'codVideo=([0-9]+)', webpage, u'internal video ID') internal_id = self._search_regex(r'codVideo=([0-9]+)', webpage, u'internal video ID')
data_url = u'http://tu.tv/flvurl.php?codVideo=' + str(internal_id) data_url = u'http://tu.tv/flvurl.php?codVideo=' + str(internal_id)
@ -36,6 +34,6 @@ class TutvIE(InfoExtractor):
'id': internal_id, 'id': internal_id,
'url': video_url, 'url': video_url,
'ext': ext, 'ext': ext,
'title': title, 'title': self._og_search_title(webpage),
} }
return [info] return [info]

View file

@ -27,12 +27,6 @@ class VineIE(InfoExtractor):
video_url = self._html_search_regex(r'<meta property="twitter:player:stream" content="(.+?)"', video_url = self._html_search_regex(r'<meta property="twitter:player:stream" content="(.+?)"',
webpage, u'video URL') webpage, u'video URL')
video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"',
webpage, u'title')
thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)(\?.*?)?"',
webpage, u'thumbnail', fatal=False)
uploader = self._html_search_regex(r'<div class="user">.*?<h2>(.+?)</h2>', uploader = self._html_search_regex(r'<div class="user">.*?<h2>(.+?)</h2>',
webpage, u'uploader', fatal=False, flags=re.DOTALL) webpage, u'uploader', fatal=False, flags=re.DOTALL)
@ -40,7 +34,7 @@ class VineIE(InfoExtractor):
'id': video_id, 'id': video_id,
'url': video_url, 'url': video_url,
'ext': 'mp4', 'ext': 'mp4',
'title': video_title, 'title': self._og_search_title(webpage),
'thumbnail': thumbnail, 'thumbnail': self._og_search_thumbnail(webpage),
'uploader': uploader, 'uploader': uploader,
}] }]