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