]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/niconico.py
debian/control: Add recommends on aria2 | wget | curl to use external downloaders.
[youtubedl] / youtube_dl / extractor / niconico.py
1 # encoding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5 import json
6
7 from .common import InfoExtractor
8 from ..compat import (
9 compat_urllib_parse,
10 compat_urllib_request,
11 compat_urlparse,
12 )
13 from ..utils import (
14 ExtractorError,
15 int_or_none,
16 parse_duration,
17 unified_strdate,
18 )
19
20
21 class NiconicoIE(InfoExtractor):
22 IE_NAME = 'niconico'
23 IE_DESC = 'ニコニコ動画'
24
25 _TEST = {
26 'url': 'http://www.nicovideo.jp/watch/sm22312215',
27 'md5': 'd1a75c0823e2f629128c43e1212760f9',
28 'info_dict': {
29 'id': 'sm22312215',
30 'ext': 'mp4',
31 'title': 'Big Buck Bunny',
32 'uploader': 'takuya0301',
33 'uploader_id': '2698420',
34 'upload_date': '20131123',
35 'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
36 'duration': 33,
37 },
38 'params': {
39 'username': 'ydl.niconico@gmail.com',
40 'password': 'youtube-dl',
41 },
42 }
43
44 _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/((?:[a-z]{2})?[0-9]+)'
45 _NETRC_MACHINE = 'niconico'
46 # Determine whether the downloader used authentication to download video
47 _AUTHENTICATED = False
48
49 def _real_initialize(self):
50 self._login()
51
52 def _login(self):
53 (username, password) = self._get_login_info()
54 # No authentication to be performed
55 if not username:
56 return True
57
58 # Log in
59 login_form_strs = {
60 'mail': username,
61 'password': password,
62 }
63 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
64 # chokes on unicode
65 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
66 login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
67 request = compat_urllib_request.Request(
68 'https://secure.nicovideo.jp/secure/login', login_data)
69 login_results = self._download_webpage(
70 request, None, note='Logging in', errnote='Unable to log in')
71 if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
72 self._downloader.report_warning('unable to log in: bad username or password')
73 return False
74 # Successful login
75 self._AUTHENTICATED = True
76 return True
77
78 def _real_extract(self, url):
79 mobj = re.match(self._VALID_URL, url)
80 video_id = mobj.group(1)
81
82 # Get video webpage. We are not actually interested in it, but need
83 # the cookies in order to be able to download the info webpage
84 self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
85
86 video_info = self._download_xml(
87 'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
88 note='Downloading video info page')
89
90 if self._AUTHENTICATED:
91 # Get flv info
92 flv_info_webpage = self._download_webpage(
93 'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
94 video_id, 'Downloading flv info')
95 else:
96 # Get external player info
97 ext_player_info = self._download_webpage(
98 'http://ext.nicovideo.jp/thumb_watch/' + video_id, video_id)
99 thumb_play_key = self._search_regex(
100 r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey')
101
102 # Get flv info
103 flv_info_data = compat_urllib_parse.urlencode({
104 'k': thumb_play_key,
105 'v': video_id
106 })
107 flv_info_request = compat_urllib_request.Request(
108 'http://ext.nicovideo.jp/thumb_watch', flv_info_data,
109 {'Content-Type': 'application/x-www-form-urlencoded'})
110 flv_info_webpage = self._download_webpage(
111 flv_info_request, video_id,
112 note='Downloading flv info', errnote='Unable to download flv info')
113
114 if 'deleted=' in flv_info_webpage:
115 raise ExtractorError('The video has been deleted.',
116 expected=True)
117 video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
118
119 # Start extracting information
120 title = video_info.find('.//title').text
121 extension = video_info.find('.//movie_type').text
122 video_format = extension.upper()
123 thumbnail = video_info.find('.//thumbnail_url').text
124 description = video_info.find('.//description').text
125 upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
126 view_count = int_or_none(video_info.find('.//view_counter').text)
127 comment_count = int_or_none(video_info.find('.//comment_num').text)
128 duration = parse_duration(video_info.find('.//length').text)
129 webpage_url = video_info.find('.//watch_url').text
130
131 if video_info.find('.//ch_id') is not None:
132 uploader_id = video_info.find('.//ch_id').text
133 uploader = video_info.find('.//ch_name').text
134 elif video_info.find('.//user_id') is not None:
135 uploader_id = video_info.find('.//user_id').text
136 uploader = video_info.find('.//user_nickname').text
137 else:
138 uploader_id = uploader = None
139
140 return {
141 'id': video_id,
142 'url': video_real_url,
143 'title': title,
144 'ext': extension,
145 'format': video_format,
146 'thumbnail': thumbnail,
147 'description': description,
148 'uploader': uploader,
149 'upload_date': upload_date,
150 'uploader_id': uploader_id,
151 'view_count': view_count,
152 'comment_count': comment_count,
153 'duration': duration,
154 'webpage_url': webpage_url,
155 }
156
157
158 class NiconicoPlaylistIE(InfoExtractor):
159 _VALID_URL = r'https?://www\.nicovideo\.jp/mylist/(?P<id>\d+)'
160
161 _TEST = {
162 'url': 'http://www.nicovideo.jp/mylist/27411728',
163 'info_dict': {
164 'id': '27411728',
165 'title': 'AKB48のオールナイトニッポン',
166 },
167 'playlist_mincount': 225,
168 }
169
170 def _real_extract(self, url):
171 list_id = self._match_id(url)
172 webpage = self._download_webpage(url, list_id)
173
174 entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
175 webpage, 'entries')
176 entries = json.loads(entries_json)
177 entries = [{
178 '_type': 'url',
179 'ie_key': NiconicoIE.ie_key(),
180 'url': ('http://www.nicovideo.jp/watch/%s' %
181 entry['item_data']['video_id']),
182 } for entry in entries]
183
184 return {
185 '_type': 'playlist',
186 'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
187 'id': list_id,
188 'entries': entries,
189 }