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
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse,
   9     compat_urllib_request,
  10     compat_urlparse,
  11     compat_str,
  12
  13     ExtractorError,
  14     unified_strdate,
  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         },
  34         'params': {
  35             'username': 'ydl.niconico@gmail.com',
  36             'password': 'youtube-dl',
  37         },
  38     }
  39
  40     _VALID_URL = r'^https?://(?:www\.|secure\.)?nicovideo\.jp/watch/([a-z][a-z][0-9]+)(?:.*)$'
  41     _NETRC_MACHINE = 'niconico'
  42
  43     def _real_initialize(self):
  44         self._login()
  45
  46     def _login(self):
  47         (username, password) = self._get_login_info()
  48         if username is None:
  49             # Login is required
  50             raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  51
  52         # Log in
  53         login_form_strs = {
  54             'mail': username,
  55             'password': password,
  56         }
  57         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
  58         # chokes on unicode
  59         login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
  60         login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
  61         request = compat_urllib_request.Request(
  62             'https://secure.nicovideo.jp/secure/login', login_data)
  63         login_results = self._download_webpage(
  64             request, None, note='Logging in', errnote='Unable to log in')
  65         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
  66             self._downloader.report_warning('unable to log in: bad username or password')
  67             return False
  68         return True
  69
  70     def _real_extract(self, url):
  71         mobj = re.match(self._VALID_URL, url)
  72         video_id = mobj.group(1)
  73
  74         # Get video webpage. We are not actually interested in it, but need
  75         # the cookies in order to be able to download the info webpage
  76         self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
  77
  78         video_info = self._download_xml(
  79             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
  80             note='Downloading video info page')
  81
  82         # Get flv info
  83         flv_info_webpage = self._download_webpage(
  84             'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
  85             video_id, 'Downloading flv info')
  86         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
  87
  88         # Start extracting information
  89         video_title = video_info.find('.//title').text
  90         video_extension = video_info.find('.//movie_type').text
  91         video_format = video_extension.upper()
  92         video_thumbnail = video_info.find('.//thumbnail_url').text
  93         video_description = video_info.find('.//description').text
  94         video_uploader_id = video_info.find('.//user_id').text
  95         video_upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
  96         video_view_count = video_info.find('.//view_counter').text
  97         video_webpage_url = video_info.find('.//watch_url').text
  98
  99         # uploader
 100         video_uploader = video_uploader_id
 101         url = 'http://seiga.nicovideo.jp/api/user/info?id=' + video_uploader_id
 102         try:
 103             user_info = self._download_xml(
 104                 url, video_id, note='Downloading user information')
 105             video_uploader = user_info.find('.//nickname').text
 106         except ExtractorError as err:
 107             self._downloader.report_warning('Unable to download user info webpage: %s' % compat_str(err))
 108
 109         return {
 110             'id': video_id,
 111             'url': video_real_url,
 112             'title': video_title,
 113             'ext': video_extension,
 114             'format': video_format,
 115             'thumbnail': video_thumbnail,
 116             'description': video_description,
 117             'uploader': video_uploader,
 118             'upload_date': video_upload_date,
 119             'uploader_id': video_uploader_id,
 120             'view_count': video_view_count,
 121             'webpage_url': video_webpage_url,
 122         }