_ Git - youtube-dl/blob - youtube_dl/extractor/niconico.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse,
   9     compat_urllib_request,
  10     compat_urlparse,
  11     ExtractorError,
  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 uses authentication to download video
  44     _AUTHENTICATE = False
  45
  46     def _real_initialize(self):
  47         if self._downloader.params.get('username', None) is not None:
  48             self._AUTHENTICATE = True
  49
  50         if self._AUTHENTICATE:
  51             self._login()
  52
  53     def _login(self):
  54         (username, password) = self._get_login_info()
  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         return True
  73
  74     def _real_extract(self, url):
  75         mobj = re.match(self._VALID_URL, url)
  76         video_id = mobj.group(1)
  77
  78         # Get video webpage. We are not actually interested in it, but need
  79         # the cookies in order to be able to download the info webpage
  80         self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
  81
  82         video_info = self._download_xml(
  83             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
  84             note='Downloading video info page')
  85
  86         if self._AUTHENTICATE:
  87             # Get flv info
  88             flv_info_webpage = self._download_webpage(
  89                 'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
  90                 video_id, 'Downloading flv info')
  91         else:
  92             # Get external player info
  93             ext_player_info = self._download_webpage(
  94                 'http://ext.nicovideo.jp/thumb_watch/' + video_id, video_id)
  95             thumb_play_key = self._search_regex(
  96                 r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey')
  97
  98             # Get flv info
  99             flv_info_data = compat_urllib_parse.urlencode({
 100                 'k': thumb_play_key,
 101                 'v': video_id
 102             })
 103             flv_info_request = compat_urllib_request.Request(
 104                 'http://ext.nicovideo.jp/thumb_watch', flv_info_data,
 105                 {'Content-Type': 'application/x-www-form-urlencoded'})
 106             flv_info_webpage = self._download_webpage(
 107                 flv_info_request, video_id,
 108                 note='Downloading flv info', errnote='Unable to download flv info')
 109
 110         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
 111
 112         # Start extracting information
 113         title = video_info.find('.//title').text
 114         extension = video_info.find('.//movie_type').text
 115         video_format = extension.upper()
 116         thumbnail = video_info.find('.//thumbnail_url').text
 117         description = video_info.find('.//description').text
 118         upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
 119         view_count = int_or_none(video_info.find('.//view_counter').text)
 120         comment_count = int_or_none(video_info.find('.//comment_num').text)
 121         duration = parse_duration(video_info.find('.//length').text)
 122         webpage_url = video_info.find('.//watch_url').text
 123
 124         if video_info.find('.//ch_id') is not None:
 125             uploader_id = video_info.find('.//ch_id').text
 126             uploader = video_info.find('.//ch_name').text
 127         elif video_info.find('.//user_id') is not None:
 128             uploader_id = video_info.find('.//user_id').text
 129             uploader = video_info.find('.//user_nickname').text
 130         else:
 131             uploader_id = uploader = None
 132
 133         return {
 134             'id': video_id,
 135             'url': video_real_url,
 136             'title': title,
 137             'ext': extension,
 138             'format': video_format,
 139             'thumbnail': thumbnail,
 140             'description': description,
 141             'uploader': uploader,
 142             'upload_date': upload_date,
 143             'uploader_id': uploader_id,
 144             'view_count': view_count,
 145             'comment_count': comment_count,
 146             'duration': duration,
 147             'webpage_url': webpage_url,
 148         }