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     ExtractorError,
  16 )
  17
  18
  19 class NiconicoIE(InfoExtractor):
  20     IE_NAME = 'niconico'
  21     IE_DESC = 'ニコニコ動画'
  22
  23     _TEST = {
  24         'url': 'http://www.nicovideo.jp/watch/sm22312215',
  25         'md5': 'd1a75c0823e2f629128c43e1212760f9',
  26         'info_dict': {
  27             'id': 'sm22312215',
  28             'ext': 'mp4',
  29             'title': 'Big Buck Bunny',
  30             'uploader': 'takuya0301',
  31             'uploader_id': '2698420',
  32             'upload_date': '20131123',
  33             'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
  34             'duration': 33,
  35         },
  36         'params': {
  37             'username': 'ydl.niconico@gmail.com',
  38             'password': 'youtube-dl',
  39         },
  40     }
  41
  42     _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/((?:[a-z]{2})?[0-9]+)'
  43     _NETRC_MACHINE = 'niconico'
  44     # Determine whether the downloader used authentication to download video
  45     _AUTHENTICATED = False
  46
  47     def _real_initialize(self):
  48         self._login()
  49
  50     def _login(self):
  51         (username, password) = self._get_login_info()
  52         # No authentication to be performed
  53         if not username:
  54             return True
  55
  56         # Log in
  57         login_form_strs = {
  58             'mail': username,
  59             'password': password,
  60         }
  61         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
  62         # chokes on unicode
  63         login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
  64         login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
  65         request = compat_urllib_request.Request(
  66             'https://secure.nicovideo.jp/secure/login', login_data)
  67         login_results = self._download_webpage(
  68             request, None, note='Logging in', errnote='Unable to log in')
  69         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
  70             self._downloader.report_warning('unable to log in: bad username or password')
  71             return False
  72         # Successful login
  73         self._AUTHENTICATED = True
  74         return True
  75
  76     def _real_extract(self, url):
  77         mobj = re.match(self._VALID_URL, url)
  78         video_id = mobj.group(1)
  79
  80         # Get video webpage. We are not actually interested in it, but need
  81         # the cookies in order to be able to download the info webpage
  82         self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
  83
  84         video_info = self._download_xml(
  85             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
  86             note='Downloading video info page')
  87
  88         if self._AUTHENTICATED:
  89             # Get flv info
  90             flv_info_webpage = self._download_webpage(
  91                 'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
  92                 video_id, 'Downloading flv info')
  93         else:
  94             # Get external player info
  95             ext_player_info = self._download_webpage(
  96                 'http://ext.nicovideo.jp/thumb_watch/' + video_id, video_id)
  97             thumb_play_key = self._search_regex(
  98                 r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey')
  99
 100             # Get flv info
 101             flv_info_data = compat_urllib_parse.urlencode({
 102                 'k': thumb_play_key,
 103                 'v': video_id
 104             })
 105             flv_info_request = compat_urllib_request.Request(
 106                 'http://ext.nicovideo.jp/thumb_watch', flv_info_data,
 107                 {'Content-Type': 'application/x-www-form-urlencoded'})
 108             flv_info_webpage = self._download_webpage(
 109                 flv_info_request, video_id,
 110                 note='Downloading flv info', errnote='Unable to download flv info')
 111
 112         if 'deleted=' in flv_info_webpage:
 113             raise ExtractorError('The video has been deleted.',
 114                 expected=True)
 115         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
 116
 117         # Start extracting information
 118         title = video_info.find('.//title').text
 119         extension = video_info.find('.//movie_type').text
 120         video_format = extension.upper()
 121         thumbnail = video_info.find('.//thumbnail_url').text
 122         description = video_info.find('.//description').text
 123         upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
 124         view_count = int_or_none(video_info.find('.//view_counter').text)
 125         comment_count = int_or_none(video_info.find('.//comment_num').text)
 126         duration = parse_duration(video_info.find('.//length').text)
 127         webpage_url = video_info.find('.//watch_url').text
 128
 129         if video_info.find('.//ch_id') is not None:
 130             uploader_id = video_info.find('.//ch_id').text
 131             uploader = video_info.find('.//ch_name').text
 132         elif video_info.find('.//user_id') is not None:
 133             uploader_id = video_info.find('.//user_id').text
 134             uploader = video_info.find('.//user_nickname').text
 135         else:
 136             uploader_id = uploader = None
 137
 138         return {
 139             'id': video_id,
 140             'url': video_real_url,
 141             'title': title,
 142             'ext': extension,
 143             'format': video_format,
 144             'thumbnail': thumbnail,
 145             'description': description,
 146             'uploader': uploader,
 147             'upload_date': upload_date,
 148             'uploader_id': uploader_id,
 149             'view_count': view_count,
 150             'comment_count': comment_count,
 151             'duration': duration,
 152             'webpage_url': webpage_url,
 153         }
 154
 155
 156 class NiconicoPlaylistIE(InfoExtractor):
 157     _VALID_URL = r'https?://www\.nicovideo\.jp/mylist/(?P<id>\d+)'
 158
 159     _TEST = {
 160         'url': 'http://www.nicovideo.jp/mylist/27411728',
 161         'info_dict': {
 162             'id': '27411728',
 163             'title': 'AKB48のオールナイトニッポン',
 164         },
 165         'playlist_mincount': 225,
 166     }
 167
 168     def _real_extract(self, url):
 169         list_id = self._match_id(url)
 170         webpage = self._download_webpage(url, list_id)
 171
 172         entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
 173             webpage, 'entries')
 174         entries = json.loads(entries_json)
 175         entries = [{
 176             '_type': 'url',
 177             'ie_key': NiconicoIE.ie_key(),
 178             'url': ('http://www.nicovideo.jp/watch/%s' %
 179                 entry['item_data']['video_id']),
 180         } for entry in entries]
 181
 182         return {
 183             '_type': 'playlist',
 184             'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
 185             'id': list_id,
 186             'entries': entries,
 187         }