yt_dlp/extractor/tumblr.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     urlencode_postdata
  10 )
  11
  12
  13 class TumblrIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?P<blog_name>[^/?#&]+)\.tumblr\.com/(?:post|video)/(?P<id>[0-9]+)(?:$|[/?#])'
  15     _NETRC_MACHINE = 'tumblr'
  16     _LOGIN_URL = 'https://www.tumblr.com/login'
  17     _TESTS = [{
  18         'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
  19         'md5': '479bb068e5b16462f5176a6828829767',
  20         'info_dict': {
  21             'id': '54196191430',
  22             'ext': 'mp4',
  23             'title': 'tatiana maslany news, Orphan Black || DVD extra - behind the scenes ↳...',
  24             'description': 'md5:37db8211e40b50c7c44e95da14f630b7',
  25             'thumbnail': r're:http://.*\.jpg',
  26         }
  27     }, {
  28         'url': 'http://5sostrum.tumblr.com/post/90208453769/yall-forgetting-the-greatest-keek-of-them-all',
  29         'md5': 'bf348ef8c0ef84fbf1cbd6fa6e000359',
  30         'info_dict': {
  31             'id': '90208453769',
  32             'ext': 'mp4',
  33             'title': '5SOS STRUM ;]',
  34             'description': 'md5:dba62ac8639482759c8eb10ce474586a',
  35             'thumbnail': r're:http://.*\.jpg',
  36         }
  37     }, {
  38         'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video',
  39         'md5': '7ae503065ad150122dc3089f8cf1546c',
  40         'info_dict': {
  41             'id': '130323439814',
  42             'ext': 'mp4',
  43             'title': 'HD Video Testing \u2014 Test description for my HD video',
  44             'description': 'md5:97cc3ab5fcd27ee4af6356701541319c',
  45             'thumbnail': r're:http://.*\.jpg',
  46         },
  47         'params': {
  48             'format': 'hd',
  49         },
  50     }, {
  51         'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
  52         'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
  53         'info_dict': {
  54             'id': 'Wmur',
  55             'ext': 'mp4',
  56             'title': 'naked smoking & stretching',
  57             'upload_date': '20150506',
  58             'timestamp': 1430931613,
  59             'age_limit': 18,
  60             'uploader_id': '1638622',
  61             'uploader': 'naked-yogi',
  62         },
  63         'add_ie': ['Vidme'],
  64     }, {
  65         'url': 'http://camdamage.tumblr.com/post/98846056295/',
  66         'md5': 'a9e0c8371ea1ca306d6554e3fecf50b6',
  67         'info_dict': {
  68             'id': '105463834',
  69             'ext': 'mp4',
  70             'title': 'Cam Damage-HD 720p',
  71             'uploader': 'John Moyer',
  72             'uploader_id': 'user32021558',
  73         },
  74         'add_ie': ['Vimeo'],
  75     }, {
  76         'url': 'http://sutiblr.tumblr.com/post/139638707273',
  77         'md5': '2dd184b3669e049ba40563a7d423f95c',
  78         'info_dict': {
  79             'id': 'ir7qBEIKqvq',
  80             'ext': 'mp4',
  81             'title': 'Vine by sutiblr',
  82             'alt_title': 'Vine by sutiblr',
  83             'uploader': 'sutiblr',
  84             'uploader_id': '1198993975374495744',
  85             'upload_date': '20160220',
  86             'like_count': int,
  87             'comment_count': int,
  88             'repost_count': int,
  89         },
  90         'add_ie': ['Vine'],
  91     }, {
  92         'url': 'http://vitasidorkina.tumblr.com/post/134652425014/joskriver-victoriassecret-invisibility-or',
  93         'md5': '01c12ceb82cbf6b2fe0703aa56b3ad72',
  94         'info_dict': {
  95             'id': '-7LnUPGlSo',
  96             'ext': 'mp4',
  97             'title': 'Video by victoriassecret',
  98             'description': 'Invisibility or flight…which superpower would YOU choose? #VSFashionShow #ThisOrThat',
  99             'uploader_id': 'victoriassecret',
 100             'thumbnail': r're:^https?://.*\.jpg'
 101         },
 102         'add_ie': ['Instagram'],
 103     }]
 104
 105     def _real_initialize(self):
 106         self._login()
 107
 108     def _login(self):
 109         username, password = self._get_login_info()
 110         if username is None:
 111             return
 112
 113         login_page = self._download_webpage(
 114             self._LOGIN_URL, None, 'Downloading login page')
 115
 116         login_form = self._hidden_inputs(login_page)
 117         login_form.update({
 118             'user[email]': username,
 119             'user[password]': password
 120         })
 121
 122         response, urlh = self._download_webpage_handle(
 123             self._LOGIN_URL, None, 'Logging in',
 124             data=urlencode_postdata(login_form), headers={
 125                 'Content-Type': 'application/x-www-form-urlencoded',
 126                 'Referer': self._LOGIN_URL,
 127             })
 128
 129         # Successful login
 130         if '/dashboard' in urlh.geturl():
 131             return
 132
 133         login_errors = self._parse_json(
 134             self._search_regex(
 135                 r'RegistrationForm\.errors\s*=\s*(\[.+?\])\s*;', response,
 136                 'login errors', default='[]'),
 137             None, fatal=False)
 138         if login_errors:
 139             raise ExtractorError(
 140                 'Unable to login: %s' % login_errors[0], expected=True)
 141
 142         self.report_warning('Login has probably failed')
 143
 144     def _real_extract(self, url):
 145         m_url = self._match_valid_url(url)
 146         video_id = m_url.group('id')
 147         blog = m_url.group('blog_name')
 148
 149         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
 150         webpage, urlh = self._download_webpage_handle(url, video_id)
 151
 152         redirect_url = urlh.geturl()
 153         if 'tumblr.com/safe-mode' in redirect_url or redirect_url.startswith('/safe-mode'):
 154             raise ExtractorError(
 155                 'This Tumblr may contain sensitive media. '
 156                 'Disable safe mode in your account settings '
 157                 'at https://www.tumblr.com/settings/account#safe_mode',
 158                 expected=True)
 159
 160         iframe_url = self._search_regex(
 161             r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
 162             webpage, 'iframe url', default=None)
 163         if iframe_url is None:
 164             return self.url_result(redirect_url, 'Generic')
 165
 166         iframe = self._download_webpage(iframe_url, video_id, 'Downloading iframe page')
 167
 168         duration = None
 169         sources = []
 170
 171         sd_url = self._search_regex(
 172             r'<source[^>]+src=(["\'])(?P<url>.+?)\1', iframe,
 173             'sd video url', default=None, group='url')
 174         if sd_url:
 175             sources.append((sd_url, 'sd'))
 176
 177         options = self._parse_json(
 178             self._search_regex(
 179                 r'data-crt-options=(["\'])(?P<options>.+?)\1', iframe,
 180                 'hd video url', default='', group='options'),
 181             video_id, fatal=False)
 182         if options:
 183             duration = int_or_none(options.get('duration'))
 184             hd_url = options.get('hdUrl')
 185             if hd_url:
 186                 sources.append((hd_url, 'hd'))
 187
 188         formats = [{
 189             'url': video_url,
 190             'ext': 'mp4',
 191             'format_id': format_id,
 192             'height': int_or_none(self._search_regex(
 193                 r'/(\d{3,4})$', video_url, 'height', default=None)),
 194             'quality': quality,
 195         } for quality, (video_url, format_id) in enumerate(sources)]
 196
 197         self._sort_formats(formats)
 198
 199         # The only place where you can get a title, it's not complete,
 200         # but searching in other places doesn't work for all videos
 201         video_title = self._html_search_regex(
 202             r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
 203             webpage, 'title')
 204
 205         return {
 206             'id': video_id,
 207             'title': video_title,
 208             'description': self._og_search_description(webpage, default=None),
 209             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 210             'duration': duration,
 211             'formats': formats,
 212         }