Akkariin/youtube-dl
1
0
Fork 0
mirror of https://github.com/l1ving/youtube-dl synced 2025-03-10 07:07:19 +08:00
Code Issues Releases Wiki Activity

fix new ui bugs (#333)

Browse Source 
* fix new ui bugs

* fix new ui bugs- pr comments

Co-authored-by: bhodaya <bhodaya@videocites.com>
This commit is contained in:
hodayabu 2020-06-08 10:38:11 +03:00 committed by GitHub
parent 5a4d830b9c
commit 1d4345ec78
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 44 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

61
youtube_dl/extractor/facebook.py
View File

@ -383,7 +383,7 @@ class FacebookIE(InfoExtractor):
self.raise_login_required()
if not video_data:
info_dict = self.get_from_new_ui(webpage, tahoe_data, url)
info_dict = self.get_from_new_ui(webpage, tahoe_data, video_id)
if info_dict:
return webpage, info_dict
@ -448,11 +448,7 @@ class FacebookIE(InfoExtractor):
_lowercase_escape(self._search_regex(r'\"ownerName\":"(.+?)"', tahoe_data.secondary, 'uploader_id', fatal=False)) or \
self._search_regex(r'ownerName"\s*:\s*"([^"]+)"', webpage, 'uploader', default=None) or \
self._og_search_title(webpage, default=None)
timestamp = self._search_regex(
r'datePublished":"(.+?)"', webpage,'timestamp', default=None)\
or self._search_regex(r'datePublished":"(.+?)"', tahoe_data.secondary, 'timestamp', default=None)\
or self._search_regex(r'datePublished":"(.+?)"', tahoe_data.primary, 'timestamp', default=None)
timestamp = self._resolve_timestamp(webpage, tahoe_data)
timestamp = parse_iso8601(timestamp)
if timestamp == None and webpage.find('Paid Partnership') == -1 or\
@ -498,34 +494,32 @@ class FacebookIE(InfoExtractor):
def get_from_new_ui(self, webpage, tahoe_data, url):
def get_from_new_ui(self, webpage, tahoe_data, video_id):
video_title = self._resolve_new_ui_title(webpage, tahoe_data, video_id)
comments_count = self._resolve_new_ui_comments_count(webpage, tahoe_data)
video_title = self._search_regex(r'"headline":"(.+?")', webpage, 'title', fatal=False)
if not video_title:
video_title = self._search_regex(r'"pageTitle">(.+?)<', webpage, 'title', fatal=False)
if not video_title:
video_title = self._extract_video_title(webpage, tahoe_data, video_id)
comments_count = parse_count(self._search_regex(r'"commentCount":(.+?,)', webpage, 'comments_count', fatal=False))
likes = parse_count(self._extract_likes(webpage, tahoe_data))
timestamp = self._search_regex(r'"datePublished":"(.+?)"', webpage, 'timestamp')
timestamp = parse_iso8601(timestamp)
timestamp = self._resolve_new_ui_timestamp(webpage, tahoe_data)
uploader_json = self._search_regex(r'"author":{(.+?)}', webpage, 'uploader')
uploader_handle, uploader = self._extract_uploader_info_new_ui(uploader_json)
ids_json = self._search_regex(r'data-video-channel-id="(.+?)"', webpage, 'ids')
uploader_id = self._resolve_uploader_id(webpage, tahoe_data)
video_id = self._extract_ids_info_new_ui(ids_json)
post_view_counts = parse_count(self._search_regex(r'"postViewCount":(.+?),', tahoe_data.secondary, 'views'))
other_post_view_counts = parse_count(self._search_regex(r'"otherPostsViewCount":(.+?),', tahoe_data.secondary, 'other_views'))
share_counts = parse_count(self._search_regex(r'"sharecount":(.+?),', tahoe_data.secondary, 'other_views'))
thumbnail = self._search_regex(r'"thumbnailUrl":"(.+?)"', webpage, 'thumbnail')
is_live, live_status = self.resolve_new_ui_live_info(webpage, tahoe_data)
formats = self.resolve_new_ui_format(webpage)
info_dict = self.build_info_dict(webpage, tahoe_data, video_id, video_title, formats, uploader, timestamp,
thumbnail, post_view_counts, uploader_id, is_live, live_status, likes,
share_counts, {}, comments_count, other_post_view_counts,
@ -757,6 +751,39 @@ class FacebookIE(InfoExtractor):
fatal=False)
return uploader_id
def _resolve_timestamp(self, webpage, tahoe_data):
timestamp = self._search_regex(
r'datePublished":"(.+?)"', webpage, 'timestamp', default=None) \
or self._search_regex(r'datePublished":"(.+?)"', tahoe_data.secondary, 'timestamp', default=None) \
or self._search_regex(r'datePublished":"(.+?)"', tahoe_data.primary, 'timestamp', default=None)
return timestamp
def _resolve_new_ui_title(self, webpage, tahoe_data, video_id):
video_title = self._search_regex(r'"headline":"(.+?")', webpage, 'title', fatal=False)
if not video_title:
video_title = self._search_regex(r'"pageTitle">(.+?)<', webpage, 'title', fatal=False)
if not video_title:
video_title = self._extract_video_title(webpage, tahoe_data, video_id)
return video_title
def _resolve_new_ui_comments_count(self, webpage, tahoe_data):
comments_count = parse_count(
self._search_regex(r'"commentCount":(.+?,)', webpage, 'comments_count', fatal=False))
if comments_count is None:
comments_count = parse_count(
self._search_regex(r'"commentcount":(.+?,)', tahoe_data.secondary, 'comments_count', fatal=False))
if comments_count is None:
comments_count = parse_count(self._extract_comments_count(webpage, tahoe_data))
return comments_count
def _resolve_new_ui_timestamp(self, webpage, tahoe_data):
timestamp = self._search_regex(r'"datePublished":"(.+?)"', webpage, 'timestamp', fatal=False)
if not timestamp:
timestamp = self._resolve_timestamp(webpage, tahoe_data)
timestamp = parse_iso8601(timestamp)
return timestamp
class FacebookTahoeData:
def __init__(self, extractor, page, video_id):
self._page = page