[ign] improve extraction and extract uploader_id
[youtube-dl] / youtube_dl / extractor / howcast.py
1 from __future__ import unicode_literals
2
3 from .common import InfoExtractor
4 from ..utils import parse_iso8601
5
6
7 class HowcastIE(InfoExtractor):
8     _VALID_URL = r'https?://(?:www\.)?howcast\.com/videos/(?P<id>\d+)'
9     _TEST = {
10         'url': 'http://www.howcast.com/videos/390161-How-to-Tie-a-Square-Knot-Properly',
11         'md5': '8b743df908c42f60cf6496586c7f12c3',
12         'info_dict': {
13             'id': '390161',
14             'ext': 'mp4',
15             'title': 'How to Tie a Square Knot Properly',
16             'description': 'md5:dbe792e5f6f1489027027bf2eba188a3',
17             'timestamp': 1276081287,
18             'upload_date': '20100609',
19         },
20         'params': {
21             # m3u8 download
22             'skip_download': True,
23         },
24     }
25
26     def _real_extract(self, url):
27         video_id = self._match_id(url)
28
29         webpage = self._download_webpage(url, video_id)
30
31         embed_code = self._search_regex(
32             r'<iframe[^>]+src="[^"]+\bembed_code=([^\b]+)\b',
33             webpage, 'ooyala embed code')
34
35         return {
36             '_type': 'url_transparent',
37             'ie_key': 'Ooyala',
38             'url': 'ooyala:%s' % embed_code,
39             'id': video_id,
40             'timestamp': parse_iso8601(self._html_search_meta(
41                 'article:published_time', webpage, 'timestamp')),
42         }