[redtube] Extract all formats, duration, upload date and view count (Closes #9397)

This commit is contained in:
Sergey M․ 2016-05-05 21:02:54 +06:00
parent b1c6a5bac8
commit ac12e888f9
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D

View File

@ -1,7 +1,12 @@
from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import ExtractorError
from ..utils import (
ExtractorError,
int_or_none,
str_to_int,
unified_strdate,
)
class RedTubeIE(InfoExtractor):
@ -13,6 +18,9 @@ class RedTubeIE(InfoExtractor):
'id': '66418',
'ext': 'mp4',
'title': 'Sucked on a toilet',
'upload_date': '20120831',
'duration': 596,
'view_count': int,
'age_limit': 18,
}
}
@ -24,12 +32,40 @@ class RedTubeIE(InfoExtractor):
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)
video_url = self._html_search_regex(
r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
video_title = self._html_search_regex(
r'<h1 class="videoTitle[^"]*">(.+?)</h1>',
webpage, 'title')
video_thumbnail = self._og_search_thumbnail(webpage)
title = self._html_search_regex(
(r'<h1 class="videoTitle[^"]*">(?P<title>.+?)</h1>',
r'videoTitle\s*:\s*(["\'])(?P<title>)\1'),
webpage, 'title', group='title')
formats = []
sources = self._parse_json(
self._search_regex(
r'sources\s*:\s*({.+?})', webpage, 'source', default='{}'),
video_id, fatal=False)
if sources and isinstance(sources, dict):
for format_id, format_url in sources.items():
if format_url:
formats.append({
'url': format_url,
'format_id': format_id,
'height': int_or_none(format_id),
})
else:
video_url = self._html_search_regex(
r'<source src="(.+?)" type="video/mp4">', webpage, 'video URL')
formats.append({'url': video_url})
self._sort_formats(formats)
thumbnail = self._og_search_thumbnail(webpage)
upload_date = unified_strdate(self._search_regex(
r'<span[^>]+class="added-time"[^>]*>ADDED ([^<]+)<',
webpage, 'upload date', fatal=False))
duration = int_or_none(self._search_regex(
r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False))
view_count = str_to_int(self._search_regex(
r'<span[^>]*>VIEWS</span></td>\s*<td>([\d,.]+)',
webpage, 'view count', fatal=False))
# No self-labeling, but they describe themselves as
# "Home of Videos Porno"
@ -37,9 +73,12 @@ class RedTubeIE(InfoExtractor):
return {
'id': video_id,
'url': video_url,
'ext': 'mp4',
'title': video_title,
'thumbnail': video_thumbnail,
'title': title,
'thumbnail': thumbnail,
'upload_date': upload_date,
'duration': duration,
'view_count': view_count,
'age_limit': age_limit,
'formats': formats,
}