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

   1 # encoding: utf-8
   2
   3 import re
   4 import socket
   5 import xml.etree.ElementTree
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     compat_http_client,
  10     compat_urllib_error,
  11     compat_urllib_parse,
  12     compat_urllib_request,
  13     compat_urlparse,
  14     compat_str,
  15
  16     ExtractorError,
  17     unified_strdate,
  18 )
  19
  20
  21 class NiconicoIE(InfoExtractor):
  22     IE_NAME = u'niconico'
  23     IE_DESC = u'ニコニコ動画'
  24
  25     _TEST = {
  26         u'url': u'http://www.nicovideo.jp/watch/sm22312215',
  27         u'file': u'sm22312215.mp4',
  28         u'md5': u'd1a75c0823e2f629128c43e1212760f9',
  29         u'info_dict': {
  30             u'title': u'Big Buck Bunny',
  31             u'uploader': u'takuya0301',
  32             u'uploader_id': u'2698420',
  33             u'upload_date': u'20131123',
  34             u'description': u'(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
  35         },
  36         u'params': {
  37             u'username': u'ydl.niconico@gmail.com',
  38             u'password': u'youtube-dl',
  39         },
  40     }
  41
  42     _VALID_URL = r'^https?://(?:www\.|secure\.)?nicovideo\.jp/watch/([a-z][a-z][0-9]+)(?:.*)$'
  43     _NETRC_MACHINE = 'niconico'
  44     # If True it will raise an error if no login info is provided
  45     _LOGIN_REQUIRED = True
  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 username is None:
  54             if self._LOGIN_REQUIRED:
  55                 raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  56             return False
  57
  58         # Log in
  59         login_form_strs = {
  60             u'mail': username,
  61             u'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             u'https://secure.nicovideo.jp/secure/login', login_data)
  69         login_results = self._download_webpage(
  70             request, u'', note=u'Logging in', errnote=u'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(u'unable to log in: bad username or password')
  73             return False
  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
  81         video_webpage = self._download_webpage(
  82             'http://www.nicovideo.jp/watch/' + video_id, video_id)
  83
  84         video_info_webpage = self._download_webpage(
  85             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
  86             note=u'Downloading video info page')
  87
  88         # Get flv info
  89         flv_info_webpage = self._download_webpage(
  90             u'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
  91             video_id, u'Downloading flv info')
  92         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
  93
  94         # Start extracting information
  95         video_info = xml.etree.ElementTree.fromstring(video_info_webpage)
  96         video_title = video_info.find('.//title').text
  97         video_extension = video_info.find('.//movie_type').text
  98         video_format = video_extension.upper()
  99         video_thumbnail = video_info.find('.//thumbnail_url').text
 100         video_description = video_info.find('.//description').text
 101         video_uploader_id = video_info.find('.//user_id').text
 102         video_upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
 103         video_view_count = video_info.find('.//view_counter').text
 104         video_webpage_url = video_info.find('.//watch_url').text
 105
 106         # uploader
 107         video_uploader = video_uploader_id
 108         url = 'http://seiga.nicovideo.jp/api/user/info?id=' + video_uploader_id
 109         try:
 110             user_info_webpage = self._download_webpage(
 111                 url, video_id, note=u'Downloading user information')
 112         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 113             self._downloader.report_warning(u'Unable to download user info webpage: %s' % compat_str(err))
 114         else:
 115             user_info = xml.etree.ElementTree.fromstring(user_info_webpage)
 116             video_uploader = user_info.find('.//nickname').text
 117
 118         return {
 119             'id':          video_id,
 120             'url':         video_real_url,
 121             'title':       video_title,
 122             'ext':         video_extension,
 123             'format':      video_format,
 124             'thumbnail':   video_thumbnail,
 125             'description': video_description,
 126             'uploader':    video_uploader,
 127             'upload_date': video_upload_date,
 128             'uploader_id': video_uploader_id,
 129             'view_count':  video_view_count,
 130             'webpage_url': video_webpage_url,
 131         }