_ 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     unified_strdate,
  18 )
  19
  20
  21 class NiconicoIE(InfoExtractor):
  22     IE_NAME = 'niconico'
  23     IE_DESC = 'ニコニコ動画'
  24
  25     _TESTS = [{
  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         'url': 'http://www.nicovideo.jp/watch/nm14296458',
  44         'md5': '8db08e0158457cf852a31519fceea5bc',
  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:',
  50             'uploader': 'りょうた',
  51             'uploader_id': '18822557',
  52             'upload_date': '20110429',
  53             'duration': 209,
  54         },
  55         'params': {
  56             'username': 'ydl.niconico@gmail.com',
  57             'password': 'youtube-dl',
  58         },
  59     }]
  60
  61     _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'
  62     _NETRC_MACHINE = 'niconico'
  63     # Determine whether the downloader used authentication to download video
  64     _AUTHENTICATED = False
  65
  66     def _real_initialize(self):
  67         self._login()
  68
  69     def _login(self):
  70         (username, password) = self._get_login_info()
  71         # No authentication to be performed
  72         if not username:
  73             return True
  74
  75         # Log in
  76         login_form_strs = {
  77             'mail': username,
  78             'password': password,
  79         }
  80         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
  81         # chokes on unicode
  82         login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
  83         login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
  84         request = compat_urllib_request.Request(
  85             'https://secure.nicovideo.jp/secure/login', login_data)
  86         login_results = self._download_webpage(
  87             request, None, note='Logging in', errnote='Unable to log in')
  88         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
  89             self._downloader.report_warning('unable to log in: bad username or password')
  90             return False
  91         # Successful login
  92         self._AUTHENTICATED = True
  93         return True
  94
  95     def _real_extract(self, url):
  96         video_id = self._match_id(url)
  97
  98         # Get video webpage. We are not actually interested in it, but need
  99         # the cookies in order to be able to download the info webpage
 100         self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
 101
 102         video_info = self._download_xml(
 103             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
 104             note='Downloading video info page')
 105
 106         if self._AUTHENTICATED:
 107             # Get flv info
 108             flv_info_webpage = self._download_webpage(
 109                 'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1',
 110                 video_id, 'Downloading flv info')
 111         else:
 112             # Get external player info
 113             ext_player_info = self._download_webpage(
 114                 'http://ext.nicovideo.jp/thumb_watch/' + video_id, video_id)
 115             thumb_play_key = self._search_regex(
 116                 r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey')
 117
 118             # Get flv info
 119             flv_info_data = compat_urllib_parse.urlencode({
 120                 'k': thumb_play_key,
 121                 'v': video_id
 122             })
 123             flv_info_request = compat_urllib_request.Request(
 124                 'http://ext.nicovideo.jp/thumb_watch', flv_info_data,
 125                 {'Content-Type': 'application/x-www-form-urlencoded'})
 126             flv_info_webpage = self._download_webpage(
 127                 flv_info_request, video_id,
 128                 note='Downloading flv info', errnote='Unable to download flv info')
 129
 130         if 'deleted=' in flv_info_webpage:
 131             raise ExtractorError('The video has been deleted.',
 132                                  expected=True)
 133         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
 134
 135         # Start extracting information
 136         title = video_info.find('.//title').text
 137         extension = video_info.find('.//movie_type').text
 138         video_format = extension.upper()
 139         thumbnail = video_info.find('.//thumbnail_url').text
 140         description = video_info.find('.//description').text
 141         upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
 142         view_count = int_or_none(video_info.find('.//view_counter').text)
 143         comment_count = int_or_none(video_info.find('.//comment_num').text)
 144         duration = parse_duration(video_info.find('.//length').text)
 145         webpage_url = video_info.find('.//watch_url').text
 146
 147         if video_info.find('.//ch_id') is not None:
 148             uploader_id = video_info.find('.//ch_id').text
 149             uploader = video_info.find('.//ch_name').text
 150         elif video_info.find('.//user_id') is not None:
 151             uploader_id = video_info.find('.//user_id').text
 152             uploader = video_info.find('.//user_nickname').text
 153         else:
 154             uploader_id = uploader = None
 155
 156         return {
 157             'id': video_id,
 158             'url': video_real_url,
 159             'title': title,
 160             'ext': extension,
 161             'format': video_format,
 162             'thumbnail': thumbnail,
 163             'description': description,
 164             'uploader': uploader,
 165             'upload_date': upload_date,
 166             'uploader_id': uploader_id,
 167             'view_count': view_count,
 168             'comment_count': comment_count,
 169             'duration': duration,
 170             'webpage_url': webpage_url,
 171         }
 172
 173
 174 class NiconicoPlaylistIE(InfoExtractor):
 175     _VALID_URL = r'https?://www\.nicovideo\.jp/mylist/(?P<id>\d+)'
 176
 177     _TEST = {
 178         'url': 'http://www.nicovideo.jp/mylist/27411728',
 179         'info_dict': {
 180             'id': '27411728',
 181             'title': 'AKB48のオールナイトニッポン',
 182         },
 183         'playlist_mincount': 225,
 184     }
 185
 186     def _real_extract(self, url):
 187         list_id = self._match_id(url)
 188         webpage = self._download_webpage(url, list_id)
 189
 190         entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
 191                                           webpage, 'entries')
 192         entries = json.loads(entries_json)
 193         entries = [{
 194             '_type': 'url',
 195             'ie_key': NiconicoIE.ie_key(),
 196             'url': ('http://www.nicovideo.jp/watch/%s' %
 197                     entry['item_data']['video_id']),
 198         } for entry in entries]
 199
 200         return {
 201             '_type': 'playlist',
 202             'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
 203             'id': list_id,
 204             'entries': entries,
 205         }