jens/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[redtube] Improve metadata extraction (closes #22492, closes #22615)

Browse source
This commit is contained in:
Sergey M․ 2019-10-05 22:04:49 +07:00
parent 4bf568d36c
commit 560d3b7d7c
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 16 additions and 12 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

28
youtube_dl/extractor/redtube.py
View file

@ -6,6 +6,7 @@ from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
int_or_none, int_or_none,
merge_dicts,
str_to_int, str_to_int,
unified_strdate, unified_strdate,
url_or_none, url_or_none,
@ -45,11 +46,14 @@ class RedTubeIE(InfoExtractor):
if any(s in webpage for s in ['video-deleted-info', '>This video has been removed']): if any(s in webpage for s in ['video-deleted-info', '>This video has been removed']):
raise ExtractorError('Video %s has been removed' % video_id, expected=True) raise ExtractorError('Video %s has been removed' % video_id, expected=True)
title = self._html_search_regex( info = self._search_json_ld(webpage, video_id, default={})
(r'<h(\d)[^>]+class="(?:video_title_text|videoTitle)[^"]*">(?P<title>(?:(?!\1).)+)</h\1>',
r'(?:videoTitle|title)\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',), if not info.get('title'):
webpage, 'title', group='title', info['title'] = self._html_search_regex(
default=None) or self._og_search_title(webpage) (r'<h(\d)[^>]+class="(?:video_title_text|videoTitle)[^"]*">(?P<title>(?:(?!\1).)+)</h\1>',
r'(?:videoTitle|title)\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',),
webpage, 'title', group='title',
default=None) or self._og_search_title(webpage)
formats = [] formats = []
sources = self._parse_json( sources = self._parse_json(
@ -88,28 +92,28 @@ class RedTubeIE(InfoExtractor):
thumbnail = self._og_search_thumbnail(webpage) thumbnail = self._og_search_thumbnail(webpage)
upload_date = unified_strdate(self._search_regex( upload_date = unified_strdate(self._search_regex(
r'<span[^>]+>ADDED ([^<]+)<', r'<span[^>]+>(?:ADDED|Published on) ([^<]+)<',
webpage, 'upload date', fatal=False)) webpage, 'upload date', default=None))
duration = int_or_none(self._og_search_property( duration = int_or_none(self._og_search_property(
'video:duration', webpage, default=None) or self._search_regex( 'video:duration', webpage, default=None) or self._search_regex(
r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None)) r'videoDuration\s*:\s*(\d+)', webpage, 'duration', default=None))
view_count = str_to_int(self._search_regex( view_count = str_to_int(self._search_regex(
(r'<div[^>]*>Views</div>\s*<div[^>]*>\s*([\d,.]+)', (r'<div[^>]*>Views</div>\s*<div[^>]*>\s*([\d,.]+)',
r'<span[^>]*>VIEWS</span>\s*</td>\s*<td>\s*([\d,.]+)'), r'<span[^>]*>VIEWS</span>\s*</td>\s*<td>\s*([\d,.]+)',
webpage, 'view count', fatal=False)) r'<span[^>]+\bclass=["\']video_view_count[^>]*>\s*([\d,.]+)'),
webpage, 'view count', default=None))
# No self-labeling, but they describe themselves as # No self-labeling, but they describe themselves as
# "Home of Videos Porno" # "Home of Videos Porno"
age_limit = 18 age_limit = 18
return { return merge_dicts(info, {
'id': video_id, 'id': video_id,
'ext': 'mp4', 'ext': 'mp4',
'title': title,
'thumbnail': thumbnail, 'thumbnail': thumbnail,
'upload_date': upload_date, 'upload_date': upload_date,
'duration': duration, 'duration': duration,
'view_count': view_count, 'view_count': view_count,
'age_limit': age_limit, 'age_limit': age_limit,
'formats': formats, 'formats': formats,
} })