]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/aparat.py
Merge tag 'upstream/2014.11.21'
[youtubedl] / youtube_dl / extractor / aparat.py
1 #coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import re
6
7 from .common import InfoExtractor
8 from ..utils import (
9 ExtractorError,
10 HEADRequest,
11 )
12
13
14 class AparatIE(InfoExtractor):
15 _VALID_URL = r'^https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P<id>[a-zA-Z0-9]+)'
16
17 _TEST = {
18 'url': 'http://www.aparat.com/v/wP8On',
19 'md5': '6714e0af7e0d875c5a39c4dc4ab46ad1',
20 'info_dict': {
21 'id': 'wP8On',
22 'ext': 'mp4',
23 'title': 'تیم گلکسی 11 - زومیت',
24 },
25 # 'skip': 'Extremely unreliable',
26 }
27
28 def _real_extract(self, url):
29 m = re.match(self._VALID_URL, url)
30 video_id = m.group('id')
31
32 # Note: There is an easier-to-parse configuration at
33 # http://www.aparat.com/video/video/config/videohash/%video_id
34 # but the URL in there does not work
35 embed_url = ('http://www.aparat.com/video/video/embed/videohash/' +
36 video_id + '/vt/frame')
37 webpage = self._download_webpage(embed_url, video_id)
38
39 video_urls = re.findall(r'fileList\[[0-9]+\]\s*=\s*"([^"]+)"', webpage)
40 for i, video_url in enumerate(video_urls):
41 req = HEADRequest(video_url)
42 res = self._request_webpage(
43 req, video_id, note=u'Testing video URL %d' % i, errnote=False)
44 if res:
45 break
46 else:
47 raise ExtractorError(u'No working video URLs found')
48
49 title = self._search_regex(r'\s+title:\s*"([^"]+)"', webpage, u'title')
50 thumbnail = self._search_regex(
51 r'\s+image:\s*"([^"]+)"', webpage, u'thumbnail', fatal=False)
52
53 return {
54 'id': video_id,
55 'title': title,
56 'url': video_url,
57 'ext': 'mp4',
58 'thumbnail': thumbnail,
59 }