_ 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 class NiconicoIE(InfoExtractor):
  21     IE_NAME = u'niconico'
  22     IE_DESC = u'ニコニコ動画'
  23
  24     _TEST = {
  25         u'url': u'http://www.nicovideo.jp/watch/sm22312215',
  26         u'file': u'sm22312215.mp4',
  27         u'md5': u'd1a75c0823e2f629128c43e1212760f9',
  28         u'info_dict': {
  29             u'title': u'Big Buck Bunny',
  30             u'uploader': u'takuya0301',
  31             u'uploader_id': u'2698420',
  32             u'upload_date': u'20131123',
  33             u'description': u'(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
  34         },
  35         u'params': {
  36             u'username': u'ydl.niconico@gmail.com',
  37             u'password': u'youtube-dl',
  38         },
  39     }
  40
  41     _VALID_URL = r'^(?:https?://)?(?:www\.)?nicovideo\.jp/watch/([a-z][a-z][0-9]+)(?:.*)$'
  42     _LOGIN_URL = 'https://secure.nicovideo.jp/secure/login'
  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('ascii')
  67         request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
  68         try:
  69             self.report_login()
  70             login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
  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         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  75             self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
  76             return False
  77         return True
  78
  79     def _real_extract(self, url):
  80         video_id = self._extract_id(url)
  81
  82         # Get video webpage
  83         self.report_video_webpage_download(video_id)
  84         url = 'http://www.nicovideo.jp/watch/' + video_id
  85         request = compat_urllib_request.Request(url)
  86         try:
  87             video_webpage = compat_urllib_request.urlopen(request).read()
  88         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  89             raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
  90
  91         # Get video info
  92         self.report_video_info_webpage_download(video_id)
  93         url = 'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id
  94         request = compat_urllib_request.Request(url)
  95         try:
  96             video_info_webpage = compat_urllib_request.urlopen(request).read()
  97         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  98             raise ExtractorError(u'Unable to download video info webpage: %s' % compat_str(err))
  99
 100         # Get flv info
 101         self.report_flv_info_webpage_download(video_id)
 102         url = 'http://flapi.nicovideo.jp/api/getflv?v=' + video_id
 103         request = compat_urllib_request.Request(url)
 104         try:
 105             flv_info_webpage = compat_urllib_request.urlopen(request).read()
 106         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 107             raise ExtractorError(u'Unable to download flv info webpage: %s' % compat_str(err))
 108
 109         # Start extracting information
 110         self.report_information_extraction(video_id)
 111         video_info = xml.etree.ElementTree.fromstring(video_info_webpage)
 112
 113         # url
 114         video_real_url = compat_urlparse.parse_qs(flv_info_webpage.decode('utf-8'))['url'][0]
 115
 116         # title
 117         video_title = video_info.find('.//title').text
 118
 119         # ext
 120         video_extension = video_info.find('.//movie_type').text
 121
 122         # format
 123         video_format = video_extension.upper()
 124
 125         # thumbnail
 126         video_thumbnail = video_info.find('.//thumbnail_url').text
 127
 128         # description
 129         video_description = video_info.find('.//description').text
 130
 131         # uploader_id
 132         video_uploader_id = video_info.find('.//user_id').text
 133
 134         # uploader
 135         url = 'http://seiga.nicovideo.jp/api/user/info?id=' + video_uploader_id
 136         request = compat_urllib_request.Request(url)
 137         try:
 138             user_info_webpage = compat_urllib_request.urlopen(request).read()
 139         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
 140             self._downloader.report_warning(u'Unable to download user info webpage: %s' % compat_str(err))
 141
 142         user_info = xml.etree.ElementTree.fromstring(user_info_webpage)
 143         video_uploader = user_info.find('.//nickname').text
 144
 145         # uploder_date
 146         video_upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
 147
 148         # view_count
 149         video_view_count = video_info.find('.//view_counter').text
 150
 151         # webpage_url
 152         video_webpage_url = video_info.find('.//watch_url').text
 153
 154         return {
 155             'id':          video_id,
 156             'url':         video_real_url,
 157             'title':       video_title,
 158             'ext':         video_extension,
 159             'format':      video_format,
 160             'thumbnail':   video_thumbnail,
 161             'description': video_description,
 162             'uploader':    video_uploader,
 163             'upload_date': video_upload_date,
 164             'uploader_id': video_uploader_id,
 165             'view_count':  video_view_count,
 166             'webpage_url': video_webpage_url,
 167         }
 168
 169     def _extract_id(self, url):
 170         mobj = re.match(self._VALID_URL, url)
 171         if mobj is None:
 172             raise ExtractorError(u'Invalid URL: %s' % url)
 173         video_id = mobj.group(1)
 174         return video_id
 175
 176     def report_video_webpage_download(self, video_id):
 177         """Report attempt to download video webpage."""
 178         self.to_screen(u'%s: Downloading video webpage' % video_id)
 179
 180     def report_video_info_webpage_download(self, video_id):
 181         """Report attempt to download video info webpage."""
 182         self.to_screen(u'%s: Downloading video info webpage' % video_id)
 183
 184     def report_flv_info_webpage_download(self, video_id):
 185         """Report attempt to download flv info webpage."""
 186         self.to_screen(u'%s: Downloading flv info webpage' % video_id)
 187
 188     def report_information_extraction(self, video_id):
 189         """Report attempt to extract video information."""
 190         self.to_screen(u'%s: Extracting video information' % video_id)