v/youtube-dl
1
0
Fork 0
mirror of https://github.com/l1ving/youtube-dl synced 2020-11-18 19:53:54 -08:00
Code Issues Releases Wiki Activity

Update facebook.py

Browse Source
This commit is contained in:
ajj8 2020-03-29 03:06:06 +01:00 committed by GitHub
parent 772f93a75d
commit 09eed10644
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 26 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
youtube_dl/extractor/facebook.py
View File

@ -22,11 +22,12 @@ from ..utils import (
int_or_none, int_or_none,
js_to_json, js_to_json,
limit_length, limit_length,
merge_dicts,
parse_count, parse_count,
sanitized_Request, sanitized_Request,
try_get, try_get,
urlencode_postdata, urlencode_postdata,
merge_dicts, urljoin,
) )
@ -515,8 +516,14 @@ class FacebookPluginsVideoIE(InfoExtractor):
class FacebookUserIE(InfoExtractor): class FacebookUserIE(InfoExtractor):
_VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?facebook\.com/(?:pg/)?(?P<id>[^/?#&]+))/videos(?!/[\w\.])' _VALID_URL = r'https?://(?:[^/]+\.)?facebook\.com/(?:pg/)?(?P<id>[^/?#&]+)/videos'
IE_NAME = "facebook:user" IE_NAME = 'facebook:user'
@classmethod
def suitable(cls, url):
return (False
if FacebookIE.suitable(url)
else super(FacebookUserIE, cls).suitable(url))
_TESTS = [{ _TESTS = [{
# page # page
@ -532,17 +539,17 @@ class FacebookUserIE(InfoExtractor):
}] }]
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) user_id = self._match_id(url)
user_id = mobj.group('id')
page = self._download_webpage( page = self._download_webpage(
url, user_id, 'Downloading user webpage') url, user_id, 'Downloading user webpage')
fb_url = self._html_search_meta( fb_url = self._html_search_meta(
'al:android:url', page, default=None) 'al:android:url', page, default=None)
fb_url_mobj = re.match(r'fb://(?P<type>page|profile)/(?P<id>\d+)', fb_url) fb_url_re = r'fb://(?P<type>page|profile)/(?P<id>\d+)'
if not fb_url_mobj: page_type = self._search_regex(
raise ExtractorError('Could not extract page ID', expected=False) fb_url_re, fb_url, 'page type', group='type')
page_id = fb_url_mobj.group('id') page_id = self._search_regex(
fb_url_re, fb_url, 'page id', group='id')
fb_dtsg_ag = self._search_regex( fb_dtsg_ag = self._search_regex(
r'"async_get_token":"([\w\-:]+)"', r'"async_get_token":"([\w\-:]+)"',
page, 'fb_dtsg_ag', default=None) page, 'fb_dtsg_ag', default=None)
@ -555,15 +562,15 @@ class FacebookUserIE(InfoExtractor):
cursor = None cursor = None
entries = [] entries = []
if fb_url_mobj.group('type') == 'page': if page_type == 'page':
endpoint = 'PagesVideoHubVideoContainerPagelet' endpoint = 'PagesVideoHubVideoContainerPagelet'
data = { data = {
'page': page_id 'page': page_id
} }
page_re = r'<td[^>]+>.+?<a href="(?P<url>[^"]+)".+?</td>' page_re = r'<td[^>]+>.+?<a href="(?P<url>[^"]+)".+?</td>'
elif fb_url_mobj.group('type') == 'profile': elif page_type == 'profile':
if not (fb_dtsg_ag and pagelet_token and collection_token): if not (fb_dtsg_ag and pagelet_token and collection_token):
raise ExtractorError('You must be logged in to extract profile videos', expected=True) raise ExtractorError('You must use cookies to extract profile videos', expected=True)
endpoint = 'VideosByUserAppCollectionPagelet' endpoint = 'VideosByUserAppCollectionPagelet'
data = { data = {
'collection_token': collection_token, 'collection_token': collection_token,
@ -592,17 +599,19 @@ class FacebookUserIE(InfoExtractor):
self._search_regex( self._search_regex(
r'({.+})', js_data_page, r'({.+})', js_data_page,
'js data', default='{}'), 'js data', default='{}'),
user_id, fatal=True) user_id)
for video in re.findall( for video in re.findall(
page_re, js_data['payload']): page_re, js_data['payload']):
video_url = urljoin('https://www.facebook.com', video)
if not FacebookIE.suitable(video_url):
continue
entries.append( entries.append(
self.url_result( self.url_result(
'https://www.facebook.com%s' % video, video_url, FacebookIE.ie_key()))
FacebookIE.ie_key()))
cursor = None cursor = None
if fb_url_mobj.group('type') == 'page': if page_type == 'page':
if 'instances' not in js_data['jsmods']: if 'instances' not in js_data['jsmods']:
break break
for parent in js_data['jsmods']['instances']: for parent in js_data['jsmods']['instances']:
@ -613,7 +622,7 @@ class FacebookUserIE(InfoExtractor):
if type(subchild) is dict and 'cursor' in subchild: if type(subchild) is dict and 'cursor' in subchild:
cursor = subchild['cursor'] cursor = subchild['cursor']
break break
elif fb_url_mobj.group('type') == 'profile': elif page_type == 'profile':
for parent in js_data['jsmods']['require']: for parent in js_data['jsmods']['require']:
if type(parent) is list: if type(parent) is list:
for child in parent: for child in parent: