[pyvideo] Fix extraction (Closes #10468)

This commit is contained in:
Sergey M․ 2016-08-30 23:51:18 +07:00
parent 3c77a54d5d
commit 245023a861
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D

View File

@ -1,59 +1,73 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import re import re
import os
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import compat_str
from ..utils import int_or_none
class PyvideoIE(InfoExtractor): class PyvideoIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?pyvideo\.org/video/(?P<id>\d+)/(.*)' _VALID_URL = r'https?://(?:www\.)?pyvideo\.org/(?P<category>[^/]+)/(?P<id>[^/?#&.]+)'
_TESTS = [ _TESTS = [{
{ 'url': 'http://pyvideo.org/pycon-us-2013/become-a-logging-expert-in-30-minutes.html',
'url': 'http://pyvideo.org/video/1737/become-a-logging-expert-in-30-minutes',
'md5': '520915673e53a5c5d487c36e0c4d85b5',
'info_dict': { 'info_dict': {
'id': '24_4WWkSmNo', 'id': 'become-a-logging-expert-in-30-minutes',
'ext': 'webm',
'title': 'Become a logging expert in 30 minutes',
'description': 'md5:9665350d466c67fb5b1598de379021f7',
'upload_date': '20130320',
'uploader': 'Next Day Video',
'uploader_id': 'NextDayVideo',
}, },
'add_ie': ['Youtube'], 'playlist_count': 2,
}, }, {
{ 'url': 'http://pyvideo.org/pygotham-2012/gloriajw-spotifywitherikbernhardsson182m4v.html',
'url': 'http://pyvideo.org/video/2542/gloriajw-spotifywitherikbernhardsson182m4v',
'md5': '5fe1c7e0a8aa5570330784c847ff6d12', 'md5': '5fe1c7e0a8aa5570330784c847ff6d12',
'info_dict': { 'info_dict': {
'id': '2542', 'id': '2542',
'ext': 'm4v', 'ext': 'm4v',
'title': 'Gloriajw-SpotifyWithErikBernhardsson182', 'title': 'Gloriajw-SpotifyWithErikBernhardsson182.m4v',
}, },
}, }]
]
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
category = mobj.group('category')
video_id = mobj.group('id') video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id) entries = []
m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', webpage) data = self._download_json(
if m_youtube is not None: 'https://raw.githubusercontent.com/pyvideo/data/master/%s/videos/%s.json'
return self.url_result(m_youtube.group(1), 'Youtube') % (category, video_id), video_id, fatal=False)
title = self._html_search_regex( if data:
r'<div class="section">\s*<h3(?:\s+class="[^"]*"[^>]*)?>([^>]+?)</h3>', print(data)
webpage, 'title', flags=re.DOTALL) for video in data['videos']:
video_url = self._search_regex( video_url = video.get('url')
[r'<source src="(.*?)"', r'<dt>Download</dt>.*?<a href="(.+?)"'], if video_url:
webpage, 'video url', flags=re.DOTALL) if video.get('type') == 'youtube':
entries.append(self.url_result(video_url, 'Youtube'))
return { else:
'id': video_id, entries.append({
'title': os.path.splitext(title)[0], 'id': compat_str(data.get('id') or video_id),
'url': video_url, 'url': video_url,
} 'title': data['title'],
'description': data.get('description') or data.get('summary'),
'thumbnail': data.get('thumbnail_url'),
'duration': int_or_none(data.get('duration')),
})
else:
webpage = self._download_webpage(url, video_id)
title = self._og_search_title(webpage)
media_urls = self._search_regex(
r'(?s)Media URL:(.+?)</li>', webpage, 'media urls')
for m in re.finditer(
r'<a[^>]+href=(["\'])(?P<url>http.+?)\1', media_urls):
media_url = m.group('url')
if re.match(r'https?://www\.youtube\.com/watch\?v=.*', media_url):
entries.append(self.url_result(media_url, 'Youtube'))
else:
entries.append({
'id': video_id,
'url': media_url,
'title': title,
})
return self.playlist_result(entries, video_id)