[nuvid] Fix extraction

Closes #7620
This commit is contained in:
Yen Chi Hsuan 2016-04-28 17:51:20 +08:00
parent 7f776fa4b5
commit 0cbcbdd89d
No known key found for this signature in database
GPG Key ID: 3FDDD575826C5C30

View File

@ -5,8 +5,6 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
parse_duration, parse_duration,
sanitized_Request,
unified_strdate,
) )
@ -20,7 +18,6 @@ class NuvidIE(InfoExtractor):
'ext': 'mp4', 'ext': 'mp4',
'title': 'Horny babes show their awesome bodeis and', 'title': 'Horny babes show their awesome bodeis and',
'duration': 129, 'duration': 129,
'upload_date': '20140508',
'age_limit': 18, 'age_limit': 18,
} }
} }
@ -28,28 +25,31 @@ class NuvidIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
formats = [] page_url = 'http://m.nuvid.com/video/%s' % video_id
webpage = self._download_webpage(
page_url, video_id, 'Downloading video page')
# When dwnld_speed exists and has a value larger than the MP4 file's
# bitrate, Nuvid returns the MP4 URL
# It's unit is 100bytes/millisecond, see mobile-nuvid-min.js for the algorithm
self._set_cookie('nuvid.com', 'dwnld_speed', '10.0')
mp4_webpage = self._download_webpage(
page_url, video_id, 'Downloading video page for MP4 format')
for dwnld_speed, format_id in [(0, '3gp'), (5, 'mp4')]: html5_video_re = r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']',
request = sanitized_Request( video_url = self._html_search_regex(html5_video_re, webpage, video_id)
'http://m.nuvid.com/play/%s' % video_id) mp4_video_url = self._html_search_regex(html5_video_re, mp4_webpage, video_id)
request.add_header('Cookie', 'skip_download_page=1; dwnld_speed=%d; adv_show=1' % dwnld_speed) formats = [{
webpage = self._download_webpage( 'url': video_url,
request, video_id, 'Downloading %s page' % format_id) }]
video_url = self._html_search_regex( if mp4_video_url != video_url:
r'<a\s+href="([^"]+)"\s+class="b_link">', webpage, '%s video URL' % format_id, fatal=False)
if not video_url:
continue
formats.append({ formats.append({
'url': video_url, 'url': mp4_video_url,
'format_id': format_id,
}) })
webpage = self._download_webpage(
'http://m.nuvid.com/video/%s' % video_id, video_id, 'Downloading video page')
title = self._html_search_regex( title = self._html_search_regex(
[r'<span title="([^"]+)">', [r'<span title="([^"]+)">',
r'<div class="thumb-holder video">\s*<h5[^>]*>([^<]+)</h5>'], webpage, 'title').strip() r'<div class="thumb-holder video">\s*<h5[^>]*>([^<]+)</h5>',
r'<span[^>]+class="title_thumb">([^<]+)</span>'], webpage, 'title').strip()
thumbnails = [ thumbnails = [
{ {
'url': thumb_url, 'url': thumb_url,
@ -57,9 +57,8 @@ def _real_extract(self, url):
] ]
thumbnail = thumbnails[0]['url'] if thumbnails else None thumbnail = thumbnails[0]['url'] if thumbnails else None
duration = parse_duration(self._html_search_regex( duration = parse_duration(self._html_search_regex(
r'<i class="fa fa-clock-o"></i>\s*(\d{2}:\d{2})', webpage, 'duration', fatal=False)) [r'<i class="fa fa-clock-o"></i>\s*(\d{2}:\d{2})',
upload_date = unified_strdate(self._html_search_regex( r'<span[^>]+class="view_time">([^<]+)</span>'], webpage, 'duration', fatal=False))
r'<i class="fa fa-user"></i>\s*(\d{4}-\d{2}-\d{2})', webpage, 'upload date', fatal=False))
return { return {
'id': video_id, 'id': video_id,
@ -67,7 +66,6 @@ def _real_extract(self, url):
'thumbnails': thumbnails, 'thumbnails': thumbnails,
'thumbnail': thumbnail, 'thumbnail': thumbnail,
'duration': duration, 'duration': duration,
'upload_date': upload_date,
'age_limit': 18, 'age_limit': 18,
'formats': formats, 'formats': formats,
} }