]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/addanime.py
[youtube] extract more format metadata
[youtube-dl] / youtube_dl / extractor / addanime.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..compat import (
7     compat_HTTPError,
8     compat_str,
9     compat_urllib_parse_urlencode,
10     compat_urllib_parse_urlparse,
11 )
12 from ..utils import (
13     ExtractorError,
14     qualities,
15 )
16
17
18 class AddAnimeIE(InfoExtractor):
19     _VALID_URL = r'https?://(?:\w+\.)?add-anime\.net/(?:watch_video\.php\?(?:.*?)v=|video/)(?P<id>[\w_]+)'
20     _TESTS = [{
21         'url': 'http://www.add-anime.net/watch_video.php?v=24MR3YO5SAS9',
22         'md5': '72954ea10bc979ab5e2eb288b21425a0',
23         'info_dict': {
24             'id': '24MR3YO5SAS9',
25             'ext': 'mp4',
26             'description': 'One Piece 606',
27             'title': 'One Piece 606',
28         },
29         'skip': 'Video is gone',
30     }, {
31         'url': 'http://add-anime.net/video/MDUGWYKNGBD8/One-Piece-687',
32         'only_matching': True,
33     }]
34
35     def _real_extract(self, url):
36         video_id = self._match_id(url)
37
38         try:
39             webpage = self._download_webpage(url, video_id)
40         except ExtractorError as ee:
41             if not isinstance(ee.cause, compat_HTTPError) or \
42                ee.cause.code != 503:
43                 raise
44
45             redir_webpage = ee.cause.read().decode('utf-8')
46             action = self._search_regex(
47                 r'<form id="challenge-form" action="([^"]+)"',
48                 redir_webpage, 'Redirect form')
49             vc = self._search_regex(
50                 r'<input type="hidden" name="jschl_vc" value="([^"]+)"/>',
51                 redir_webpage, 'redirect vc value')
52             av = re.search(
53                 r'a\.value = ([0-9]+)[+]([0-9]+)[*]([0-9]+);',
54                 redir_webpage)
55             if av is None:
56                 raise ExtractorError('Cannot find redirect math task')
57             av_res = int(av.group(1)) + int(av.group(2)) * int(av.group(3))
58
59             parsed_url = compat_urllib_parse_urlparse(url)
60             av_val = av_res + len(parsed_url.netloc)
61             confirm_url = (
62                 parsed_url.scheme + '://' + parsed_url.netloc +
63                 action + '?' +
64                 compat_urllib_parse_urlencode({
65                     'jschl_vc': vc, 'jschl_answer': compat_str(av_val)}))
66             self._download_webpage(
67                 confirm_url, video_id,
68                 note='Confirming after redirect')
69             webpage = self._download_webpage(url, video_id)
70
71         FORMATS = ('normal', 'hq')
72         quality = qualities(FORMATS)
73         formats = []
74         for format_id in FORMATS:
75             rex = r"var %s_video_file = '(.*?)';" % re.escape(format_id)
76             video_url = self._search_regex(rex, webpage, 'video file URLx',
77                                            fatal=False)
78             if not video_url:
79                 continue
80             formats.append({
81                 'format_id': format_id,
82                 'url': video_url,
83                 'quality': quality(format_id),
84             })
85         self._sort_formats(formats)
86         video_title = self._og_search_title(webpage)
87         video_description = self._og_search_description(webpage)
88
89         return {
90             '_type': 'video',
91             'id': video_id,
92             'formats': formats,
93             'title': video_title,
94             'description': video_description
95         }