[tube8] Fix metadata extraction (closes #17520)

This commit is contained in:
Sergey M․ 2018-09-11 02:29:45 +07:00
parent 96dbf70de6
commit 79facb2773
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D

View File

@ -45,7 +45,7 @@ def _real_extract(self, url):
r'videoTitle\s*=\s*"([^"]+)', webpage, 'title')
description = self._html_search_regex(
r'>Description:</strong>\s*(.+?)\s*<', webpage, 'description', fatal=False)
r'(?s)Description:</dt>\s*<dd>(.+?)</dd>', webpage, 'description', fatal=False)
uploader = self._html_search_regex(
r'<span class="username">\s*(.+?)\s*<',
webpage, 'uploader', fatal=False)
@ -55,19 +55,19 @@ def _real_extract(self, url):
dislike_count = int_or_none(self._search_regex(
r'rdownVar\s*=\s*"(\d+)"', webpage, 'dislike count', fatal=False))
view_count = str_to_int(self._search_regex(
r'<strong>Views: </strong>([\d,\.]+)\s*</li>',
r'Views:\s*</dt>\s*<dd>([\d,\.]+)',
webpage, 'view count', fatal=False))
comment_count = str_to_int(self._search_regex(
r'<span id="allCommentsCount">(\d+)</span>',
webpage, 'comment count', fatal=False))
category = self._search_regex(
r'Category:\s*</strong>\s*<a[^>]+href=[^>]+>([^<]+)',
r'Category:\s*</dt>\s*<dd>\s*<a[^>]+href=[^>]+>([^<]+)',
webpage, 'category', fatal=False)
categories = [category] if category else None
tags_str = self._search_regex(
r'(?s)Tags:\s*</strong>(.+?)</(?!a)',
r'(?s)Tags:\s*</dt>\s*<dd>(.+?)</(?!a)',
webpage, 'tags', fatal=False)
tags = [t for t in re.findall(
r'<a[^>]+href=[^>]+>([^<]+)', tags_str)] if tags_str else None