]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/addanime.py
[ign] Fix extraction of video in articles
[youtube-dl.git] / youtube_dl / extractor / addanime.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7     compat_HTTPError,
8     compat_str,
9     compat_urllib_parse,
10     compat_urllib_parse_urlparse,
11
12     ExtractorError,
13 )
14
15
16 class AddAnimeIE(InfoExtractor):
17
18     _VALID_URL = r'^http://(?:\w+\.)?add-anime\.net/watch_video\.php\?(?:.*?)v=(?P<video_id>[\w_]+)(?:.*)'
19     _TEST = {
20         'url': 'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9',
21         'md5': '72954ea10bc979ab5e2eb288b21425a0',
22         'info_dict': {
23             'id': '24MR3YO5SAS9',
24             'ext': 'mp4',
25             'description': 'One Piece 606',
26             'title': 'One Piece 606',
27         }
28     }
29
30     def _real_extract(self, url):
31         try:
32             mobj = re.match(self._VALID_URL, url)
33             video_id = mobj.group('video_id')
34             webpage = self._download_webpage(url, video_id)
35         except ExtractorError as ee:
36             if not isinstance(ee.cause, compat_HTTPError) or \
37                ee.cause.code != 503:
38                 raise
39
40             redir_webpage = ee.cause.read().decode('utf-8')
41             action = self._search_regex(
42                 r'<form id="challenge-form" action="([^"]+)"',
43                 redir_webpage, 'Redirect form')
44             vc = self._search_regex(
45                 r'<input type="hidden" name="jschl_vc" value="([^"]+)"/>',
46                 redir_webpage, 'redirect vc value')
47             av = re.search(
48                 r'a\.value = ([0-9]+)[+]([0-9]+)[*]([0-9]+);',
49                 redir_webpage)
50             if av is None:
51                 raise ExtractorError(u'Cannot find redirect math task')
52             av_res = int(av.group(1)) + int(av.group(2)) * int(av.group(3))
53
54             parsed_url = compat_urllib_parse_urlparse(url)
55             av_val = av_res + len(parsed_url.netloc)
56             confirm_url = (
57                 parsed_url.scheme + '://' + parsed_url.netloc +
58                 action + '?' +
59                 compat_urllib_parse.urlencode({
60                     'jschl_vc': vc, 'jschl_answer': compat_str(av_val)}))
61             self._download_webpage(
62                 confirm_url, video_id,
63                 note='Confirming after redirect')
64             webpage = self._download_webpage(url, video_id)
65
66         formats = []
67         for format_id in ('normal', 'hq'):
68             rex = r"var %s_video_file = '(.*?)';" % re.escape(format_id)
69             video_url = self._search_regex(rex, webpage, 'video file URLx',
70                                            fatal=False)
71             if not video_url:
72                 continue
73             formats.append({
74                 'format_id': format_id,
75                 'url': video_url,
76             })
77         self._sort_formats(formats)
78         video_title = self._og_search_title(webpage)
79         video_description = self._og_search_description(webpage)
80
81         return {
82             '_type': 'video',
83             'id': video_id,
84             'formats': formats,
85             'title': video_title,
86             'description': video_description
87         }