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