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/(?P<id>(?:[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         video_id = self._match_id(url)
  80
  81         # Get video webpage. We are not actually interested in it, but need
  82         # the cookies in order to be able to download the info webpage
  83         self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
  84
  85         video_info = self._download_xml(
  86             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
  87             note='Downloading video info page')
  88
  89         if self._AUTHENTICATED:
  90             # Get flv info
  91             flv_info_webpage = self._download_webpage(
  92                 'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
  93                 video_id, 'Downloading flv info')
  94         else:
  95             # Get external player info
  96             ext_player_info = self._download_webpage(
  97                 'http://ext.nicovideo.jp/thumb_watch/' + video_id, video_id)
  98             thumb_play_key = self._search_regex(
  99                 r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey')
 100
 101             # Get flv info
 102             flv_info_data = compat_urllib_parse.urlencode({
 103                 'k': thumb_play_key,
 104                 'v': video_id
 105             })
 106             flv_info_request = compat_urllib_request.Request(
 107                 'http://ext.nicovideo.jp/thumb_watch', flv_info_data,
 108                 {'Content-Type': 'application/x-www-form-urlencoded'})
 109             flv_info_webpage = self._download_webpage(
 110                 flv_info_request, video_id,
 111                 note='Downloading flv info', errnote='Unable to download flv info')
 112
 113         if 'deleted=' in flv_info_webpage:
 114             raise ExtractorError('The video has been deleted.',
 115                                  expected=True)
 116         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
 117
 118         # Start extracting information
 119         title = video_info.find('.//title').text
 120         extension = video_info.find('.//movie_type').text
 121         video_format = extension.upper()
 122         thumbnail = video_info.find('.//thumbnail_url').text
 123         description = video_info.find('.//description').text
 124         upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
 125         view_count = int_or_none(video_info.find('.//view_counter').text)
 126         comment_count = int_or_none(video_info.find('.//comment_num').text)
 127         duration = parse_duration(video_info.find('.//length').text)
 128         webpage_url = video_info.find('.//watch_url').text
 129
 130         if video_info.find('.//ch_id') is not None:
 131             uploader_id = video_info.find('.//ch_id').text
 132             uploader = video_info.find('.//ch_name').text
 133         elif video_info.find('.//user_id') is not None:
 134             uploader_id = video_info.find('.//user_id').text
 135             uploader = video_info.find('.//user_nickname').text
 136         else:
 137             uploader_id = uploader = None
 138
 139         return {
 140             'id': video_id,
 141             'url': video_real_url,
 142             'title': title,
 143             'ext': extension,
 144             'format': video_format,
 145             'thumbnail': thumbnail,
 146             'description': description,
 147             'uploader': uploader,
 148             'upload_date': upload_date,
 149             'uploader_id': uploader_id,
 150             'view_count': view_count,
 151             'comment_count': comment_count,
 152             'duration': duration,
 153             'webpage_url': webpage_url,
 154         }
 155
 156
 157 class NiconicoPlaylistIE(InfoExtractor):
 158     _VALID_URL = r'https?://www\.nicovideo\.jp/mylist/(?P<id>\d+)'
 159
 160     _TEST = {
 161         'url': 'http://www.nicovideo.jp/mylist/27411728',
 162         'info_dict': {
 163             'id': '27411728',
 164             'title': 'AKB48のオールナイトニッポン',
 165         },
 166         'playlist_mincount': 225,
 167     }
 168
 169     def _real_extract(self, url):
 170         list_id = self._match_id(url)
 171         webpage = self._download_webpage(url, list_id)
 172
 173         entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
 174                                           webpage, 'entries')
 175         entries = json.loads(entries_json)
 176         entries = [{
 177             '_type': 'url',
 178             'ie_key': NiconicoIE.ie_key(),
 179             'url': ('http://www.nicovideo.jp/watch/%s' %
 180                     entry['item_data']['video_id']),
 181         } for entry in entries]
 182
 183         return {
 184             '_type': 'playlist',
 185             'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
 186             'id': list_id,
 187             'entries': entries,
 188         }