_ 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
  44     def _real_initialize(self):
  45         self._login()
  46
  47     def _login(self):
  48         (username, password) = self._get_login_info()
  49         if username is None:
  50             # Login is required
  51             raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  52
  53         # Log in
  54         login_form_strs = {
  55             'mail': username,
  56             'password': password,
  57         }
  58         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
  59         # chokes on unicode
  60         login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
  61         login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
  62         request = compat_urllib_request.Request(
  63             'https://secure.nicovideo.jp/secure/login', login_data)
  64         login_results = self._download_webpage(
  65             request, None, note='Logging in', errnote='Unable to log in')
  66         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
  67             self._downloader.report_warning('unable to log in: bad username or password')
  68             return False
  69         return True
  70
  71     def _real_extract(self, url):
  72         mobj = re.match(self._VALID_URL, url)
  73         video_id = mobj.group(1)
  74
  75         # Get video webpage. We are not actually interested in it, but need
  76         # the cookies in order to be able to download the info webpage
  77         self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
  78
  79         video_info = self._download_xml(
  80             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
  81             note='Downloading video info page')
  82
  83         # Get flv info
  84         flv_info_webpage = self._download_webpage(
  85             'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
  86             video_id, 'Downloading flv info')
  87         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
  88
  89         # Start extracting information
  90         title = video_info.find('.//title').text
  91         extension = video_info.find('.//movie_type').text
  92         video_format = extension.upper()
  93         thumbnail = video_info.find('.//thumbnail_url').text
  94         description = video_info.find('.//description').text
  95         upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
  96         view_count = int_or_none(video_info.find('.//view_counter').text)
  97         comment_count = int_or_none(video_info.find('.//comment_num').text)
  98         duration = parse_duration(video_info.find('.//length').text)
  99         webpage_url = video_info.find('.//watch_url').text
 100
 101         if video_info.find('.//ch_id') is not None:
 102             uploader_id = video_info.find('.//ch_id').text
 103             uploader = video_info.find('.//ch_name').text
 104         elif video_info.find('.//user_id') is not None:
 105             uploader_id = video_info.find('.//user_id').text
 106             uploader = video_info.find('.//user_nickname').text
 107         else:
 108             uploader_id = uploader = None
 109
 110         return {
 111             'id': video_id,
 112             'url': video_real_url,
 113             'title': title,
 114             'ext': extension,
 115             'format': video_format,
 116             'thumbnail': thumbnail,
 117             'description': description,
 118             'uploader': uploader,
 119             'upload_date': upload_date,
 120             'uploader_id': uploader_id,
 121             'view_count': view_count,
 122             'comment_count': comment_count,
 123             'duration': duration,
 124             'webpage_url': webpage_url,
 125         }