_ Git - youtube-dl/blob - 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     parse_iso8601,
  18     xpath_text,
  19     determine_ext,
  20 )
  21
  22
  23 class NiconicoIE(InfoExtractor):
  24     IE_NAME = 'niconico'
  25     IE_DESC = 'ニコニコ動画'
  26
  27     _TESTS = [{
  28         'url': 'http://www.nicovideo.jp/watch/sm22312215',
  29         'md5': 'd1a75c0823e2f629128c43e1212760f9',
  30         'info_dict': {
  31             'id': 'sm22312215',
  32             'ext': 'mp4',
  33             'title': 'Big Buck Bunny',
  34             'uploader': 'takuya0301',
  35             'uploader_id': '2698420',
  36             'upload_date': '20131123',
  37             'timestamp': 1385182762,
  38             'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
  39             'duration': 33,
  40         },
  41         'params': {
  42             'username': 'ydl.niconico@gmail.com',
  43             'password': 'youtube-dl',
  44         },
  45     }, {
  46         'url': 'http://www.nicovideo.jp/watch/nm14296458',
  47         'md5': '8db08e0158457cf852a31519fceea5bc',
  48         'info_dict': {
  49             'id': 'nm14296458',
  50             'ext': 'swf',
  51             'title': '【鏡音リン】Dance on media【オリジナル】take2!',
  52             'description': 'md5:689f066d74610b3b22e0f1739add0f58',
  53             'uploader': 'りょうた',
  54             'uploader_id': '18822557',
  55             'upload_date': '20110429',
  56             'timestamp': 1304065916,
  57             'duration': 209,
  58         },
  59         'params': {
  60             'username': 'ydl.niconico@gmail.com',
  61             'password': 'youtube-dl',
  62         },
  63     }, {
  64         # 'video exists but is marked as "deleted"
  65         'url': 'http://www.nicovideo.jp/watch/sm10000',
  66         'md5': '38e53c9aad548f3ecf01ca7680b59b08',
  67         'info_dict': {
  68             'id': 'sm10000',
  69             'ext': 'unknown_video',
  70             'description': 'deleted',
  71             'title': 'ドラえもんエターナル第3話「決戦第3新東京市」＜前編＞',
  72         },
  73         'params': {
  74             'username': 'ydl.niconico@gmail.com',
  75             'password': 'youtube-dl',
  76         }
  77     }]
  78
  79     _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'
  80     _NETRC_MACHINE = 'niconico'
  81     # Determine whether the downloader used authentication to download video
  82     _AUTHENTICATED = False
  83
  84     def _real_initialize(self):
  85         self._login()
  86
  87     def _login(self):
  88         (username, password) = self._get_login_info()
  89         # No authentication to be performed
  90         if not username:
  91             return True
  92
  93         # Log in
  94         login_form_strs = {
  95             'mail': username,
  96             'password': password,
  97         }
  98         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
  99         # chokes on unicode
 100         login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
 101         login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
 102         request = compat_urllib_request.Request(
 103             'https://secure.nicovideo.jp/secure/login', login_data)
 104         login_results = self._download_webpage(
 105             request, None, note='Logging in', errnote='Unable to log in')
 106         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
 107             self._downloader.report_warning('unable to log in: bad username or password')
 108             return False
 109         # Successful login
 110         self._AUTHENTICATED = True
 111         return True
 112
 113     def _real_extract(self, url):
 114         video_id = self._match_id(url)
 115
 116         # Get video webpage. We are not actually interested in it for normal
 117         # cases, but need the cookies in order to be able to download the
 118         # info webpage
 119         webpage = self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
 120
 121         video_info = self._download_xml(
 122             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
 123             note='Downloading video info page')
 124
 125         if self._AUTHENTICATED:
 126             # Get flv info
 127             flv_info_webpage = self._download_webpage(
 128                 'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1',
 129                 video_id, 'Downloading flv info')
 130         else:
 131             # Get external player info
 132             ext_player_info = self._download_webpage(
 133                 'http://ext.nicovideo.jp/thumb_watch/' + video_id, video_id)
 134             thumb_play_key = self._search_regex(
 135                 r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey')
 136
 137             # Get flv info
 138             flv_info_data = compat_urllib_parse.urlencode({
 139                 'k': thumb_play_key,
 140                 'v': video_id
 141             })
 142             flv_info_request = compat_urllib_request.Request(
 143                 'http://ext.nicovideo.jp/thumb_watch', flv_info_data,
 144                 {'Content-Type': 'application/x-www-form-urlencoded'})
 145             flv_info_webpage = self._download_webpage(
 146                 flv_info_request, video_id,
 147                 note='Downloading flv info', errnote='Unable to download flv info')
 148
 149         flv_info = compat_urlparse.parse_qs(flv_info_webpage)
 150         if 'url' not in flv_info:
 151             if 'deleted' in flv_info:
 152                 raise ExtractorError('The video has been deleted.',
 153                                      expected=True)
 154             else:
 155                 raise ExtractorError('Unable to find video URL')
 156
 157         video_real_url = flv_info['url'][0]
 158
 159         # Start extracting information
 160         title = xpath_text(video_info, './/title')
 161         if not title:
 162             title = self._html_search_regex(
 163                 r'<span[^>]+class="videoHeaderTitle"[^>]*>([^<]+)</span>',
 164                 webpage, 'video title')
 165
 166         extension = xpath_text(video_info, './/movie_type')
 167         if not extension:
 168             extension = determine_ext(video_real_url)
 169         video_format = extension.upper()
 170         thumbnail = xpath_text(video_info, './/thumbnail_url')
 171         description = xpath_text(video_info, './/description')
 172         timestamp = parse_iso8601(xpath_text(video_info, './/first_retrieve'))
 173         view_count = int_or_none(xpath_text(video_info, './/view_counter'))
 174         comment_count = int_or_none(xpath_text(video_info, './/comment_num'))
 175         duration = parse_duration(xpath_text(video_info, './/length'))
 176         webpage_url = xpath_text(video_info, './/watch_url')
 177
 178         if video_info.find('.//ch_id') is not None:
 179             uploader_id = video_info.find('.//ch_id').text
 180             uploader = video_info.find('.//ch_name').text
 181         elif video_info.find('.//user_id') is not None:
 182             uploader_id = video_info.find('.//user_id').text
 183             uploader = video_info.find('.//user_nickname').text
 184         else:
 185             uploader_id = uploader = None
 186
 187         ret = {
 188             'id': video_id,
 189             'url': video_real_url,
 190             'title': title,
 191             'ext': extension,
 192             'format': video_format,
 193             'thumbnail': thumbnail,
 194             'description': description,
 195             'uploader': uploader,
 196             'timestamp': timestamp,
 197             'uploader_id': uploader_id,
 198             'view_count': view_count,
 199             'comment_count': comment_count,
 200             'duration': duration,
 201             'webpage_url': webpage_url,
 202         }
 203         return dict((k, v) for k, v in ret.items() if v is not None)
 204
 205
 206 class NiconicoPlaylistIE(InfoExtractor):
 207     _VALID_URL = r'https?://www\.nicovideo\.jp/mylist/(?P<id>\d+)'
 208
 209     _TEST = {
 210         'url': 'http://www.nicovideo.jp/mylist/27411728',
 211         'info_dict': {
 212             'id': '27411728',
 213             'title': 'AKB48のオールナイトニッポン',
 214         },
 215         'playlist_mincount': 225,
 216     }
 217
 218     def _real_extract(self, url):
 219         list_id = self._match_id(url)
 220         webpage = self._download_webpage(url, list_id)
 221
 222         entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
 223                                           webpage, 'entries')
 224         entries = json.loads(entries_json)
 225         entries = [{
 226             '_type': 'url',
 227             'ie_key': NiconicoIE.ie_key(),
 228             'url': ('http://www.nicovideo.jp/watch/%s' %
 229                     entry['item_data']['video_id']),
 230         } for entry in entries]
 231
 232         return {
 233             '_type': 'playlist',
 234             'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
 235             'id': list_id,
 236             'entries': entries,
 237         }