2013-07-10 17:49:11 +02:00
|
|
|
|
# encoding: utf-8
|
|
|
|
|
|
2014-01-06 01:47:52 +01:00
|
|
|
|
from __future__ import unicode_literals
|
|
|
|
|
|
2013-06-23 20:31:45 +02:00
|
|
|
|
import os
|
|
|
|
|
import re
|
|
|
|
|
|
|
|
|
|
from .common import InfoExtractor
|
2014-01-06 01:42:58 +01:00
|
|
|
|
from .youtube import YoutubeIE
|
2013-06-23 20:31:45 +02:00
|
|
|
|
from ..utils import (
|
|
|
|
|
compat_urllib_error,
|
|
|
|
|
compat_urllib_parse,
|
|
|
|
|
compat_urllib_request,
|
2013-08-28 12:47:27 +02:00
|
|
|
|
compat_urlparse,
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
|
|
|
|
ExtractorError,
|
2013-12-20 17:05:28 +01:00
|
|
|
|
HEADRequest,
|
2013-10-15 12:05:13 +02:00
|
|
|
|
smuggle_url,
|
|
|
|
|
unescapeHTML,
|
2013-12-17 12:33:55 +01:00
|
|
|
|
unified_strdate,
|
|
|
|
|
url_basename,
|
2013-06-23 20:31:45 +02:00
|
|
|
|
)
|
2013-07-10 17:49:11 +02:00
|
|
|
|
from .brightcove import BrightcoveIE
|
2013-12-19 20:28:52 +01:00
|
|
|
|
from .ooyala import OoyalaIE
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
2013-08-24 22:49:52 +02:00
|
|
|
|
|
2013-06-23 20:31:45 +02:00
|
|
|
|
class GenericIE(InfoExtractor):
|
2014-01-06 01:47:52 +01:00
|
|
|
|
IE_DESC = 'Generic downloader that works on some sites'
|
2013-06-23 20:31:45 +02:00
|
|
|
|
_VALID_URL = r'.*'
|
2014-01-06 01:47:52 +01:00
|
|
|
|
IE_NAME = 'generic'
|
2013-07-10 17:49:11 +02:00
|
|
|
|
_TESTS = [
|
|
|
|
|
{
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
|
|
|
|
|
'file': '13601338388002.mp4',
|
|
|
|
|
'md5': '6e15c93721d7ec9e9ca3fdbf07982cfd',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'uploader': 'www.hodiho.fr',
|
|
|
|
|
'title': 'R\u00e9gis plante sa Jeep',
|
2013-07-10 17:49:11 +02:00
|
|
|
|
}
|
|
|
|
|
},
|
2013-10-15 12:05:13 +02:00
|
|
|
|
# embedded vimeo video
|
|
|
|
|
{
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'add_ie': ['Vimeo'],
|
|
|
|
|
'url': 'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references',
|
|
|
|
|
'file': '22444065.mp4',
|
|
|
|
|
'md5': '2903896e23df39722c33f015af0666e2',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'title': 'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011',
|
|
|
|
|
'uploader_id': 'skillsmatter',
|
|
|
|
|
'uploader': 'Skills Matter',
|
2013-10-15 12:05:13 +02:00
|
|
|
|
}
|
2013-10-27 14:40:25 +01:00
|
|
|
|
},
|
|
|
|
|
# bandcamp page with custom domain
|
|
|
|
|
{
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'add_ie': ['Bandcamp'],
|
|
|
|
|
'url': 'http://bronyrock.com/track/the-pony-mash',
|
|
|
|
|
'file': '3235767654.mp3',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'title': 'The Pony Mash',
|
|
|
|
|
'uploader': 'M_Pallante',
|
2013-10-27 14:40:25 +01:00
|
|
|
|
},
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'skip': 'There is a limit of 200 free downloads / month for the test song',
|
2013-10-27 14:40:25 +01:00
|
|
|
|
},
|
2013-11-06 16:40:24 +01:00
|
|
|
|
# embedded brightcove video
|
2013-11-07 21:06:48 +01:00
|
|
|
|
# it also tests brightcove videos that need to set the 'Referer' in the
|
|
|
|
|
# http requests
|
2013-11-06 16:40:24 +01:00
|
|
|
|
{
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'add_ie': ['Brightcove'],
|
|
|
|
|
'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'id': '2765128793001',
|
|
|
|
|
'ext': 'mp4',
|
|
|
|
|
'title': 'Le cours de bourse : l’analyse technique',
|
|
|
|
|
'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
|
|
|
|
|
'uploader': 'BFM BUSINESS',
|
2013-11-06 16:40:24 +01:00
|
|
|
|
},
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'params': {
|
|
|
|
|
'skip_download': True,
|
2013-11-06 16:40:24 +01:00
|
|
|
|
},
|
|
|
|
|
},
|
2013-12-17 12:33:55 +01:00
|
|
|
|
# Direct link to a video
|
|
|
|
|
{
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
|
|
|
|
|
'file': 'trailer.mp4',
|
|
|
|
|
'md5': '67d406c2bcb6af27fa886f31aa934bbe',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'id': 'trailer',
|
|
|
|
|
'title': 'trailer',
|
|
|
|
|
'upload_date': '20100513',
|
2013-12-17 12:33:55 +01:00
|
|
|
|
}
|
2013-12-19 20:28:52 +01:00
|
|
|
|
},
|
|
|
|
|
# ooyala video
|
|
|
|
|
{
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
|
2014-01-21 01:40:34 +01:00
|
|
|
|
'file': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ.mp4',
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
|
|
|
|
|
'ext': 'mp4',
|
2014-01-21 01:40:34 +01:00
|
|
|
|
'title': '2cc213299525360.mov', # that's what we get
|
2013-12-19 20:28:52 +01:00
|
|
|
|
},
|
|
|
|
|
},
|
2013-07-10 17:49:11 +02:00
|
|
|
|
]
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
|
|
|
|
def report_download_webpage(self, video_id):
|
|
|
|
|
"""Report webpage download."""
|
|
|
|
|
if not self._downloader.params.get('test', False):
|
2014-01-06 01:47:52 +01:00
|
|
|
|
self._downloader.report_warning('Falling back on generic information extractor.')
|
2013-06-23 20:31:45 +02:00
|
|
|
|
super(GenericIE, self).report_download_webpage(video_id)
|
|
|
|
|
|
|
|
|
|
def report_following_redirect(self, new_url):
|
|
|
|
|
"""Report information extraction."""
|
2014-01-06 01:47:52 +01:00
|
|
|
|
self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
2013-12-17 12:33:55 +01:00
|
|
|
|
def _send_head(self, url):
|
2013-06-23 20:31:45 +02:00
|
|
|
|
"""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
|
2013-12-20 17:05:28 +01:00
|
|
|
|
HEADRequest also on the redirected URL
|
2013-06-23 20:31:45 +02:00
|
|
|
|
"""
|
|
|
|
|
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"))
|
2013-12-20 17:05:28 +01:00
|
|
|
|
return HEADRequest(newurl,
|
2013-06-23 20:31:45 +02:00
|
|
|
|
headers=newheaders,
|
|
|
|
|
origin_req_host=req.get_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())
|
|
|
|
|
|
2013-12-20 17:05:28 +01:00
|
|
|
|
response = opener.open(HEADRequest(url))
|
2013-06-23 20:31:45 +02:00
|
|
|
|
if response is None:
|
2014-01-06 01:47:52 +01:00
|
|
|
|
raise ExtractorError('Invalid URL protocol')
|
2013-12-17 12:33:55 +01:00
|
|
|
|
return response
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
2013-09-06 18:39:35 +02:00
|
|
|
|
parsed_url = compat_urlparse.urlparse(url)
|
|
|
|
|
if not parsed_url.scheme:
|
|
|
|
|
self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
|
|
|
|
|
return self.url_result('http://' + url)
|
2013-12-17 12:33:55 +01:00
|
|
|
|
video_id = os.path.splitext(url.split('/')[-1])[0]
|
2013-09-06 18:39:35 +02:00
|
|
|
|
|
2014-01-06 01:47:52 +01:00
|
|
|
|
self.to_screen('%s: Requesting header' % video_id)
|
2013-12-27 08:38:42 +01:00
|
|
|
|
|
2013-08-21 04:31:57 +02:00
|
|
|
|
try:
|
2013-12-17 12:33:55 +01:00
|
|
|
|
response = self._send_head(url)
|
|
|
|
|
|
|
|
|
|
# Check for redirect
|
|
|
|
|
new_url = response.geturl()
|
|
|
|
|
if url != new_url:
|
|
|
|
|
self.report_following_redirect(new_url)
|
2013-12-17 12:04:33 +01:00
|
|
|
|
return self.url_result(new_url)
|
2013-12-17 12:33:55 +01:00
|
|
|
|
|
|
|
|
|
# Check for direct link to a video
|
|
|
|
|
content_type = response.headers.get('Content-Type', '')
|
2013-12-17 16:26:32 +01:00
|
|
|
|
m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
|
2013-12-17 12:33:55 +01:00
|
|
|
|
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,
|
2014-01-06 01:47:52 +01:00
|
|
|
|
'vcodec': 'none' if m.group('type') == 'audio' else None
|
2013-12-17 12:33:55 +01:00
|
|
|
|
}],
|
|
|
|
|
'upload_date': upload_date,
|
|
|
|
|
}
|
|
|
|
|
|
2013-08-21 04:31:57 +02:00
|
|
|
|
except compat_urllib_error.HTTPError:
|
|
|
|
|
# This may be a stupid server that doesn't like HEAD, our UA, or so
|
|
|
|
|
pass
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
|
|
|
|
try:
|
|
|
|
|
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
|
2014-01-06 01:47:52 +01:00
|
|
|
|
raise ExtractorError('Failed to download URL: %s' % url)
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
|
|
|
|
self.report_extraction(video_id)
|
2013-11-18 13:28:26 +01:00
|
|
|
|
|
|
|
|
|
# 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
|
2013-12-06 09:15:04 +01:00
|
|
|
|
video_title = self._html_search_regex(
|
2014-01-06 01:47:52 +01:00
|
|
|
|
r'(?s)<title>(.*?)</title>', webpage, 'video title',
|
|
|
|
|
default='video')
|
2013-12-06 09:15:04 +01:00
|
|
|
|
|
|
|
|
|
# video uploader is domain name
|
|
|
|
|
video_uploader = self._search_regex(
|
2014-01-06 01:47:52 +01:00
|
|
|
|
r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
|
2013-11-18 13:28:26 +01:00
|
|
|
|
|
2013-08-26 21:29:31 +02:00
|
|
|
|
# Look for BrightCove:
|
2013-11-06 16:40:24 +01:00
|
|
|
|
bc_url = BrightcoveIE._extract_brightcove_url(webpage)
|
|
|
|
|
if bc_url is not None:
|
2014-01-06 01:47:52 +01:00
|
|
|
|
self.to_screen('Brightcove video detected.')
|
2014-01-07 05:34:14 +01:00
|
|
|
|
surl = smuggle_url(bc_url, {'Referer': url})
|
|
|
|
|
return self.url_result(surl, 'Brightcove')
|
2013-07-10 17:49:11 +02:00
|
|
|
|
|
2013-12-22 03:34:13 +01:00
|
|
|
|
# Look for embedded (iframe) Vimeo player
|
2013-10-15 12:05:13 +02:00
|
|
|
|
mobj = re.search(
|
2014-01-08 22:42:52 +01:00
|
|
|
|
r'<iframe[^>]+?src="((?:https?:)?//player.vimeo.com/video/.+?)"', webpage)
|
2013-10-15 12:05:13 +02:00
|
|
|
|
if mobj:
|
|
|
|
|
player_url = unescapeHTML(mobj.group(1))
|
|
|
|
|
surl = smuggle_url(player_url, {'Referer': url})
|
|
|
|
|
return self.url_result(surl, 'Vimeo')
|
|
|
|
|
|
2013-12-22 03:34:13 +01:00
|
|
|
|
# Look for embedded (swf embed) Vimeo player
|
|
|
|
|
mobj = re.search(
|
|
|
|
|
r'<embed[^>]+?src="(https?://(?:www\.)?vimeo.com/moogaloop.swf.+?)"', webpage)
|
|
|
|
|
if mobj:
|
|
|
|
|
return self.url_result(mobj.group(1), 'Vimeo')
|
|
|
|
|
|
2013-10-18 11:44:57 +02:00
|
|
|
|
# Look for embedded YouTube player
|
2013-12-19 20:44:30 +01:00
|
|
|
|
matches = re.findall(r'''(?x)
|
|
|
|
|
(?:<iframe[^>]+?src=|embedSWF\(\s*)
|
|
|
|
|
(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
|
|
|
|
|
(?:embed|v)/.+?)
|
|
|
|
|
\1''', webpage)
|
2013-11-18 13:28:26 +01:00
|
|
|
|
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)
|
2013-10-18 11:44:57 +02:00
|
|
|
|
|
2013-12-01 01:21:33 +01:00
|
|
|
|
# Look for embedded Dailymotion player
|
|
|
|
|
matches = re.findall(
|
2013-12-06 09:15:04 +01:00
|
|
|
|
r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
|
2013-12-01 01:21:33 +01:00
|
|
|
|
if matches:
|
|
|
|
|
urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion')
|
|
|
|
|
for tuppl in matches]
|
|
|
|
|
return self.playlist_result(
|
|
|
|
|
urlrs, playlist_id=video_id, playlist_title=video_title)
|
|
|
|
|
|
2013-12-06 09:15:04 +01:00
|
|
|
|
# Look for embedded Wistia player
|
|
|
|
|
match = re.search(
|
|
|
|
|
r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
|
|
|
|
|
if match:
|
|
|
|
|
return {
|
|
|
|
|
'_type': 'url_transparent',
|
|
|
|
|
'url': unescapeHTML(match.group('url')),
|
|
|
|
|
'ie_key': 'Wistia',
|
|
|
|
|
'uploader': video_uploader,
|
|
|
|
|
'title': video_title,
|
|
|
|
|
'id': video_id,
|
|
|
|
|
}
|
|
|
|
|
|
2013-12-16 20:08:23 +01:00
|
|
|
|
# Look for embedded blip.tv player
|
2013-12-30 06:15:02 +01:00
|
|
|
|
mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
|
2013-12-16 20:08:23 +01:00
|
|
|
|
if mobj:
|
2013-12-30 06:15:02 +01:00
|
|
|
|
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)
|
2013-12-16 20:08:23 +01:00
|
|
|
|
if mobj:
|
2013-12-30 06:15:02 +01:00
|
|
|
|
return self.url_result(mobj.group(1), 'BlipTV')
|
2013-12-16 20:08:23 +01:00
|
|
|
|
|
2013-10-27 14:40:25 +01:00
|
|
|
|
# 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))
|
2013-11-22 16:05:14 +01:00
|
|
|
|
# Don't set the extractor because it can be a track url or an album
|
|
|
|
|
return self.url_result(burl)
|
2013-10-27 14:40:25 +01:00
|
|
|
|
|
2013-12-16 21:45:21 +01:00
|
|
|
|
# Look for embedded Vevo player
|
|
|
|
|
mobj = re.search(
|
|
|
|
|
r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
|
|
|
|
|
if mobj is not None:
|
|
|
|
|
return self.url_result(mobj.group('url'))
|
|
|
|
|
|
2013-12-19 20:28:52 +01:00
|
|
|
|
# Look for Ooyala videos
|
|
|
|
|
mobj = re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=([^"&]+)', webpage)
|
|
|
|
|
if mobj is not None:
|
|
|
|
|
return OoyalaIE._build_url_result(mobj.group(1))
|
|
|
|
|
|
2013-12-20 17:05:28 +01:00
|
|
|
|
# Look for Aparat videos
|
2014-01-07 11:04:27 +01:00
|
|
|
|
mobj = re.search(r'<iframe src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
|
2013-12-20 17:05:28 +01:00
|
|
|
|
if mobj is not None:
|
|
|
|
|
return self.url_result(mobj.group(1), 'Aparat')
|
|
|
|
|
|
2014-01-07 08:07:46 +01:00
|
|
|
|
# Look for MPORA videos
|
2014-01-07 11:04:27 +01:00
|
|
|
|
mobj = re.search(r'<iframe .*?src="(http://mpora\.com/videos/[^"]+)"', webpage)
|
2014-01-07 08:07:46 +01:00
|
|
|
|
if mobj is not None:
|
|
|
|
|
return self.url_result(mobj.group(1), 'Mpora')
|
2014-01-08 02:11:46 +01:00
|
|
|
|
|
2014-01-08 02:07:11 +01:00
|
|
|
|
# Look for embedded Novamov player
|
|
|
|
|
mobj = re.search(
|
|
|
|
|
r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:(?:embed|www)\.)?novamov\.com/embed\.php.+?)\1', webpage)
|
|
|
|
|
if mobj is not None:
|
|
|
|
|
return self.url_result(mobj.group('url'), 'Novamov')
|
2014-01-07 08:07:46 +01:00
|
|
|
|
|
2014-01-21 18:10:14 +01:00
|
|
|
|
# Look for embedded Facebook player
|
|
|
|
|
mobj = re.search(
|
|
|
|
|
r'<iframe[^>]+?src=(["\'])(?P<url>https://www.facebook.com/video/embed.+?)\1', webpage)
|
|
|
|
|
if mobj is not None:
|
|
|
|
|
return self.url_result(mobj.group('url'), 'Facebook')
|
|
|
|
|
|
2013-06-23 20:31:45 +02:00
|
|
|
|
# Start with something easy: JW Player in SWFObject
|
|
|
|
|
mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
|
2014-01-05 05:34:06 +01:00
|
|
|
|
if mobj is None:
|
|
|
|
|
# Look for gorilla-vid style embedding
|
|
|
|
|
mobj = re.search(r'(?s)jw_plugins.*?file:\s*["\'](.*?)["\']', webpage)
|
2013-06-23 20:31:45 +02:00
|
|
|
|
if mobj is None:
|
|
|
|
|
# Broaden the search a little bit
|
|
|
|
|
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
|
2014-01-17 02:13:00 +01:00
|
|
|
|
mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)
|
2013-06-23 20:31:45 +02:00
|
|
|
|
if mobj is None:
|
|
|
|
|
# Try to find twitter cards info
|
|
|
|
|
mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
|
|
|
|
|
if mobj is None:
|
|
|
|
|
# We look for Open Graph info:
|
|
|
|
|
# We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
|
|
|
|
|
m_video_type = re.search(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
|
|
|
|
|
# We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
|
|
|
|
|
if m_video_type is not None:
|
|
|
|
|
mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
|
2013-08-21 04:32:22 +02:00
|
|
|
|
if mobj is None:
|
|
|
|
|
# HTML5 video
|
2013-09-05 18:02:17 +02:00
|
|
|
|
mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
|
2013-06-23 20:31:45 +02:00
|
|
|
|
if mobj is None:
|
2014-01-06 01:47:52 +01:00
|
|
|
|
raise ExtractorError('Unsupported URL: %s' % url)
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
|
|
|
|
# It's possible that one of the regexes
|
|
|
|
|
# matched, but returned an empty group:
|
|
|
|
|
if mobj.group(1) is None:
|
2014-01-06 01:47:52 +01:00
|
|
|
|
raise ExtractorError('Did not find a valid video URL at %s' % url)
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
2013-09-05 18:02:17 +02:00
|
|
|
|
video_url = mobj.group(1)
|
2013-08-28 12:47:27 +02:00
|
|
|
|
video_url = compat_urlparse.urljoin(url, video_url)
|
2013-09-05 18:02:17 +02:00
|
|
|
|
video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
|
2013-06-23 20:31:45 +02:00
|
|
|
|
|
2014-01-06 01:42:58 +01:00
|
|
|
|
# Sometimes, jwplayer extraction will result in a YouTube URL
|
|
|
|
|
if YoutubeIE.suitable(video_url):
|
|
|
|
|
return self.url_result(video_url, 'Youtube')
|
|
|
|
|
|
2013-06-23 20:31:45 +02:00
|
|
|
|
# here's a fun little line of code for you:
|
|
|
|
|
video_id = os.path.splitext(video_id)[0]
|
|
|
|
|
|
2013-11-25 03:35:52 +01:00
|
|
|
|
return {
|
2014-01-06 01:42:58 +01:00
|
|
|
|
'id': video_id,
|
|
|
|
|
'url': video_url,
|
2013-06-23 20:31:45 +02:00
|
|
|
|
'uploader': video_uploader,
|
2014-01-06 01:42:58 +01:00
|
|
|
|
'title': video_title,
|
2013-11-25 03:35:52 +01:00
|
|
|
|
}
|