_ 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     }, {
  42         # File downloaded with and without credentials are different, so omit
  43         # the md5 field
  44         'url': 'http://www.nicovideo.jp/watch/nm14296458',
  45         'info_dict': {
  46             'id': 'nm14296458',
  47             'ext': 'swf',
  48             'title': '【鏡音リン】Dance on media【オリジナル】take2!',
  49             'description': 'md5:689f066d74610b3b22e0f1739add0f58',
  50             'uploader': 'りょうた',
  51             'uploader_id': '18822557',
  52             'upload_date': '20110429',
  53             'timestamp': 1304065916,
  54             'duration': 209,
  55         },
  56     }, {
  57         # 'video exists but is marked as "deleted"
  58         'url': 'http://www.nicovideo.jp/watch/sm10000',
  59         'md5': '38e53c9aad548f3ecf01ca7680b59b08',
  60         'info_dict': {
  61             'id': 'sm10000',
  62             'ext': 'unknown_video',
  63             'description': 'deleted',
  64             'title': 'ドラえもんエターナル第3話「決戦第3新東京市」＜前編＞',
  65         },
  66     }]
  67
  68     _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'
  69     _NETRC_MACHINE = 'niconico'
  70     # Determine whether the downloader used authentication to download video
  71     _AUTHENTICATED = False
  72
  73     def _real_initialize(self):
  74         self._login()
  75
  76     def _login(self):
  77         (username, password) = self._get_login_info()
  78         # No authentication to be performed
  79         if not username:
  80             return True
  81
  82         # Log in
  83         login_form_strs = {
  84             'mail': username,
  85             'password': password,
  86         }
  87         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
  88         # chokes on unicode
  89         login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
  90         login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
  91         request = compat_urllib_request.Request(
  92             'https://secure.nicovideo.jp/secure/login', login_data)
  93         login_results = self._download_webpage(
  94             request, None, note='Logging in', errnote='Unable to log in')
  95         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
  96             self._downloader.report_warning('unable to log in: bad username or password')
  97             return False
  98         # Successful login
  99         self._AUTHENTICATED = True
 100         return True
 101
 102     def _real_extract(self, url):
 103         video_id = self._match_id(url)
 104
 105         # Get video webpage. We are not actually interested in it for normal
 106         # cases, but need the cookies in order to be able to download the
 107         # info webpage
 108         webpage = self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
 109
 110         video_info = self._download_xml(
 111             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
 112             note='Downloading video info page')
 113
 114         if self._AUTHENTICATED:
 115             # Get flv info
 116             flv_info_webpage = self._download_webpage(
 117                 'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1',
 118                 video_id, 'Downloading flv info')
 119         else:
 120             # Get external player info
 121             ext_player_info = self._download_webpage(
 122                 'http://ext.nicovideo.jp/thumb_watch/' + video_id, video_id)
 123             thumb_play_key = self._search_regex(
 124                 r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey')
 125
 126             # Get flv info
 127             flv_info_data = compat_urllib_parse.urlencode({
 128                 'k': thumb_play_key,
 129                 'v': video_id
 130             })
 131             flv_info_request = compat_urllib_request.Request(
 132                 'http://ext.nicovideo.jp/thumb_watch', flv_info_data,
 133                 {'Content-Type': 'application/x-www-form-urlencoded'})
 134             flv_info_webpage = self._download_webpage(
 135                 flv_info_request, video_id,
 136                 note='Downloading flv info', errnote='Unable to download flv info')
 137
 138         flv_info = compat_urlparse.parse_qs(flv_info_webpage)
 139         if 'url' not in flv_info:
 140             if 'deleted' in flv_info:
 141                 raise ExtractorError('The video has been deleted.',
 142                                      expected=True)
 143             else:
 144                 raise ExtractorError('Unable to find video URL')
 145
 146         video_real_url = flv_info['url'][0]
 147
 148         # Start extracting information
 149         title = xpath_text(video_info, './/title')
 150         if not title:
 151             title = self._og_search_title(webpage, default=None)
 152         if not title:
 153             title = self._html_search_regex(
 154                 r'<span[^>]+class="videoHeaderTitle"[^>]*>([^<]+)</span>',
 155                 webpage, 'video title')
 156
 157         extension = xpath_text(video_info, './/movie_type')
 158         if not extension:
 159             extension = determine_ext(video_real_url)
 160         video_format = extension.upper()
 161         thumbnail = xpath_text(video_info, './/thumbnail_url')
 162         description = xpath_text(video_info, './/description')
 163         timestamp = parse_iso8601(xpath_text(video_info, './/first_retrieve'))
 164         view_count = int_or_none(xpath_text(video_info, './/view_counter'))
 165         comment_count = int_or_none(xpath_text(video_info, './/comment_num'))
 166         duration = parse_duration(xpath_text(video_info, './/length'))
 167         webpage_url = xpath_text(video_info, './/watch_url')
 168
 169         if video_info.find('.//ch_id') is not None:
 170             uploader_id = video_info.find('.//ch_id').text
 171             uploader = video_info.find('.//ch_name').text
 172         elif video_info.find('.//user_id') is not None:
 173             uploader_id = video_info.find('.//user_id').text
 174             uploader = video_info.find('.//user_nickname').text
 175         else:
 176             uploader_id = uploader = None
 177
 178         ret = {
 179             'id': video_id,
 180             'url': video_real_url,
 181             'title': title,
 182             'ext': extension,
 183             'format': video_format,
 184             'thumbnail': thumbnail,
 185             'description': description,
 186             'uploader': uploader,
 187             'timestamp': timestamp,
 188             'uploader_id': uploader_id,
 189             'view_count': view_count,
 190             'comment_count': comment_count,
 191             'duration': duration,
 192             'webpage_url': webpage_url,
 193         }
 194         return dict((k, v) for k, v in ret.items() if v is not None)
 195
 196
 197 class NiconicoPlaylistIE(InfoExtractor):
 198     _VALID_URL = r'https?://www\.nicovideo\.jp/mylist/(?P<id>\d+)'
 199
 200     _TEST = {
 201         'url': 'http://www.nicovideo.jp/mylist/27411728',
 202         'info_dict': {
 203             'id': '27411728',
 204             'title': 'AKB48のオールナイトニッポン',
 205         },
 206         'playlist_mincount': 225,
 207     }
 208
 209     def _real_extract(self, url):
 210         list_id = self._match_id(url)
 211         webpage = self._download_webpage(url, list_id)
 212
 213         entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
 214                                           webpage, 'entries')
 215         entries = json.loads(entries_json)
 216         entries = [{
 217             '_type': 'url',
 218             'ie_key': NiconicoIE.ie_key(),
 219             'url': ('http://www.nicovideo.jp/watch/%s' %
 220                     entry['item_data']['video_id']),
 221         } for entry in entries]
 222
 223         return {
 224             '_type': 'playlist',
 225             'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
 226             'id': list_id,
 227             'entries': entries,
 228         }