[facebook] Add support for DASH manifests

This commit is contained in:
Yen Chi Hsuan 2016-01-30 21:31:53 +08:00
parent b323e1707d
commit 0803753fea

View file

@ -6,9 +6,11 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import (
compat_etree_fromstring,
compat_http_client, compat_http_client,
compat_urllib_error, compat_urllib_error,
compat_urllib_parse_unquote, compat_urllib_parse_unquote,
compat_urllib_parse_unquote_plus,
) )
from ..utils import ( from ..utils import (
error_to_compat_str, error_to_compat_str,
@ -44,6 +46,9 @@ class FacebookIE(InfoExtractor):
_CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1' _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
_NETRC_MACHINE = 'facebook' _NETRC_MACHINE = 'facebook'
IE_NAME = 'facebook' IE_NAME = 'facebook'
_CHROME_USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.97 Safari/537.36'
_TESTS = [{ _TESTS = [{
'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf', 'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
'md5': '6a40d33c0eccbb1af76cf0485a052659', 'md5': '6a40d33c0eccbb1af76cf0485a052659',
@ -65,6 +70,15 @@ class FacebookIE(InfoExtractor):
'expected_warnings': [ 'expected_warnings': [
'title' 'title'
] ]
}, {
'note': 'Video with DASH manifest',
'url': 'https://www.facebook.com/video.php?v=957955867617029',
'info_dict': {
'id': '957955867617029',
'ext': 'mp4',
'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...',
'uploader': 'Demy de Zeeuw',
},
}, { }, {
'url': 'https://www.facebook.com/video.php?v=10204634152394104', 'url': 'https://www.facebook.com/video.php?v=10204634152394104',
'only_matching': True, 'only_matching': True,
@ -147,8 +161,9 @@ def _real_initialize(self):
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
url = 'https://www.facebook.com/video/video.php?v=%s' % video_id req = sanitized_Request('https://www.facebook.com/video/video.php?v=%s' % video_id)
webpage = self._download_webpage(url, video_id) req.add_header('User-Agent', self._CHROME_USER_AGENT)
webpage = self._download_webpage(req, video_id)
video_data = None video_data = None
@ -197,9 +212,16 @@ def video_data_list2dict(video_data):
'url': src, 'url': src,
'preference': -10 if format_id == 'progressive' else 0, 'preference': -10 if format_id == 'progressive' else 0,
}) })
dash_manifest = f[0].get('dash_manifest')
if dash_manifest:
formats.extend(self._parse_dash_manifest(
video_id, compat_etree_fromstring(compat_urllib_parse_unquote_plus(dash_manifest)),
default_ns='urn:mpeg:dash:schema:mpd:2011'))
if not formats: if not formats:
raise ExtractorError('Cannot find video formats') raise ExtractorError('Cannot find video formats')
self._sort_formats(formats)
video_title = self._html_search_regex( video_title = self._html_search_regex(
r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title', r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title',
default=None) default=None)