]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/niconico.py
Implement a different adult sites checking algorithm
[youtube-dl] / youtube_dl / extractor / niconico.py
1 # encoding: utf-8
2
3 import re
4 import socket
5
6 from .common import InfoExtractor
7 from ..utils import (
8     compat_http_client,
9     compat_urllib_error,
10     compat_urllib_parse,
11     compat_urllib_request,
12     compat_urlparse,
13     compat_str,
14
15     ExtractorError,
16     unified_strdate,
17 )
18
19
20 class NiconicoIE(InfoExtractor):
21     IE_NAME = u'niconico'
22     IE_DESC = u'ニコニコ動画'
23
24     _TEST = {
25         u'url': u'http://www.nicovideo.jp/watch/sm22312215',
26         u'file': u'sm22312215.mp4',
27         u'md5': u'd1a75c0823e2f629128c43e1212760f9',
28         u'info_dict': {
29             u'title': u'Big Buck Bunny',
30             u'uploader': u'takuya0301',
31             u'uploader_id': u'2698420',
32             u'upload_date': u'20131123',
33             u'description': u'(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
34         },
35         u'params': {
36             u'username': u'ydl.niconico@gmail.com',
37             u'password': u'youtube-dl',
38         },
39     }
40
41     _VALID_URL = r'^https?://(?:www\.|secure\.)?nicovideo\.jp/watch/([a-z][a-z][0-9]+)(?:.*)$'
42     _NETRC_MACHINE = 'niconico'
43     # If True it will raise an error if no login info is provided
44     _LOGIN_REQUIRED = True
45
46     def _real_initialize(self):
47         self._login()
48
49     def _login(self):
50         (username, password) = self._get_login_info()
51         # No authentication to be performed
52         if username is None:
53             if self._LOGIN_REQUIRED:
54                 raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
55             return False
56
57         # Log in
58         login_form_strs = {
59             u'mail': username,
60             u'password': password,
61         }
62         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
63         # chokes on unicode
64         login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
65         login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
66         request = compat_urllib_request.Request(
67             u'https://secure.nicovideo.jp/secure/login', login_data)
68         login_results = self._download_webpage(
69             request, u'', note=u'Logging in', errnote=u'Unable to log in')
70         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
71             self._downloader.report_warning(u'unable to log in: bad username or password')
72             return False
73         return True
74
75     def _real_extract(self, url):
76         mobj = re.match(self._VALID_URL, url)
77         video_id = mobj.group(1)
78
79         # Get video webpage. We are not actually interested in it, but need
80         # the cookies in order to be able to download the info webpage
81         self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
82
83         video_info = self._download_xml(
84             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
85             note=u'Downloading video info page')
86
87         # Get flv info
88         flv_info_webpage = self._download_webpage(
89             u'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
90             video_id, u'Downloading flv info')
91         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
92
93         # Start extracting information
94         video_title = video_info.find('.//title').text
95         video_extension = video_info.find('.//movie_type').text
96         video_format = video_extension.upper()
97         video_thumbnail = video_info.find('.//thumbnail_url').text
98         video_description = video_info.find('.//description').text
99         video_uploader_id = video_info.find('.//user_id').text
100         video_upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
101         video_view_count = video_info.find('.//view_counter').text
102         video_webpage_url = video_info.find('.//watch_url').text
103
104         # uploader
105         video_uploader = video_uploader_id
106         url = 'http://seiga.nicovideo.jp/api/user/info?id=' + video_uploader_id
107         try:
108             user_info = self._download_xml(
109                 url, video_id, note=u'Downloading user information')
110             video_uploader = user_info.find('.//nickname').text
111         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
112             self._downloader.report_warning(u'Unable to download user info webpage: %s' % compat_str(err))
113
114         return {
115             'id':          video_id,
116             'url':         video_real_url,
117             'title':       video_title,
118             'ext':         video_extension,
119             'format':      video_format,
120             'thumbnail':   video_thumbnail,
121             'description': video_description,
122             'uploader':    video_uploader,
123             'upload_date': video_upload_date,
124             'uploader_id': video_uploader_id,
125             'view_count':  video_view_count,
126             'webpage_url': video_webpage_url,
127         }