Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tumblr.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     urlencode_postdata
  11 )
  12
  13
  14 class TumblrIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?P<blog_name>[^/?#&]+)\.tumblr\.com/(?:post|video)/(?P<id>[0-9]+)(?:$|[/?#])'
  16     _NETRC_MACHINE = 'tumblr'
  17     _LOGIN_URL = 'https://www.tumblr.com/login'
  18     _TESTS = [{
  19         'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
  20         'md5': '479bb068e5b16462f5176a6828829767',
  21         'info_dict': {
  22             'id': '54196191430',
  23             'ext': 'mp4',
  24             'title': 'tatiana maslany news, Orphan Black || DVD extra - behind the scenes ↳...',
  25             'description': 'md5:37db8211e40b50c7c44e95da14f630b7',
  26             'thumbnail': r're:http://.*\.jpg',
  27         }
  28     }, {
  29         'url': 'http://5sostrum.tumblr.com/post/90208453769/yall-forgetting-the-greatest-keek-of-them-all',
  30         'md5': 'bf348ef8c0ef84fbf1cbd6fa6e000359',
  31         'info_dict': {
  32             'id': '90208453769',
  33             'ext': 'mp4',
  34             'title': '5SOS STRUM ;]',
  35             'description': 'md5:dba62ac8639482759c8eb10ce474586a',
  36             'thumbnail': r're:http://.*\.jpg',
  37         }
  38     }, {
  39         'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video',
  40         'md5': '7ae503065ad150122dc3089f8cf1546c',
  41         'info_dict': {
  42             'id': '130323439814',
  43             'ext': 'mp4',
  44             'title': 'HD Video Testing \u2014 Test description for my HD video',
  45             'description': 'md5:97cc3ab5fcd27ee4af6356701541319c',
  46             'thumbnail': r're:http://.*\.jpg',
  47         },
  48         'params': {
  49             'format': 'hd',
  50         },
  51     }, {
  52         'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
  53         'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
  54         'info_dict': {
  55             'id': 'Wmur',
  56             'ext': 'mp4',
  57             'title': 'naked smoking & stretching',
  58             'upload_date': '20150506',
  59             'timestamp': 1430931613,
  60             'age_limit': 18,
  61             'uploader_id': '1638622',
  62             'uploader': 'naked-yogi',
  63         },
  64         'add_ie': ['Vidme'],
  65     }, {
  66         'url': 'http://camdamage.tumblr.com/post/98846056295/',
  67         'md5': 'a9e0c8371ea1ca306d6554e3fecf50b6',
  68         'info_dict': {
  69             'id': '105463834',
  70             'ext': 'mp4',
  71             'title': 'Cam Damage-HD 720p',
  72             'uploader': 'John Moyer',
  73             'uploader_id': 'user32021558',
  74         },
  75         'add_ie': ['Vimeo'],
  76     }, {
  77         'url': 'http://sutiblr.tumblr.com/post/139638707273',
  78         'md5': '2dd184b3669e049ba40563a7d423f95c',
  79         'info_dict': {
  80             'id': 'ir7qBEIKqvq',
  81             'ext': 'mp4',
  82             'title': 'Vine by sutiblr',
  83             'alt_title': 'Vine by sutiblr',
  84             'uploader': 'sutiblr',
  85             'uploader_id': '1198993975374495744',
  86             'upload_date': '20160220',
  87             'like_count': int,
  88             'comment_count': int,
  89             'repost_count': int,
  90         },
  91         'add_ie': ['Vine'],
  92     }, {
  93         'url': 'http://vitasidorkina.tumblr.com/post/134652425014/joskriver-victoriassecret-invisibility-or',
  94         'md5': '01c12ceb82cbf6b2fe0703aa56b3ad72',
  95         'info_dict': {
  96             'id': '-7LnUPGlSo',
  97             'ext': 'mp4',
  98             'title': 'Video by victoriassecret',
  99             'description': 'Invisibility or flight…which superpower would YOU choose? #VSFashionShow #ThisOrThat',
 100             'uploader_id': 'victoriassecret',
 101             'thumbnail': r're:^https?://.*\.jpg'
 102         },
 103         'add_ie': ['Instagram'],
 104     }]
 105
 106     def _real_initialize(self):
 107         self._login()
 108
 109     def _login(self):
 110         username, password = self._get_login_info()
 111         if username is None:
 112             return
 113
 114         login_page = self._download_webpage(
 115             self._LOGIN_URL, None, 'Downloading login page')
 116
 117         login_form = self._hidden_inputs(login_page)
 118         login_form.update({
 119             'user[email]': username,
 120             'user[password]': password
 121         })
 122
 123         response, urlh = self._download_webpage_handle(
 124             self._LOGIN_URL, None, 'Logging in',
 125             data=urlencode_postdata(login_form), headers={
 126                 'Content-Type': 'application/x-www-form-urlencoded',
 127                 'Referer': self._LOGIN_URL,
 128             })
 129
 130         # Successful login
 131         if '/dashboard' in urlh.geturl():
 132             return
 133
 134         login_errors = self._parse_json(
 135             self._search_regex(
 136                 r'RegistrationForm\.errors\s*=\s*(\[.+?\])\s*;', response,
 137                 'login errors', default='[]'),
 138             None, fatal=False)
 139         if login_errors:
 140             raise ExtractorError(
 141                 'Unable to login: %s' % login_errors[0], expected=True)
 142
 143         self.report_warning('Login has probably failed')
 144
 145     def _real_extract(self, url):
 146         m_url = re.match(self._VALID_URL, url)
 147         video_id = m_url.group('id')
 148         blog = m_url.group('blog_name')
 149
 150         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
 151         webpage, urlh = self._download_webpage_handle(url, video_id)
 152
 153         redirect_url = urlh.geturl()
 154         if 'tumblr.com/safe-mode' in redirect_url or redirect_url.startswith('/safe-mode'):
 155             raise ExtractorError(
 156                 'This Tumblr may contain sensitive media. '
 157                 'Disable safe mode in your account settings '
 158                 'at https://www.tumblr.com/settings/account#safe_mode',
 159                 expected=True)
 160
 161         iframe_url = self._search_regex(
 162             r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
 163             webpage, 'iframe url', default=None)
 164         if iframe_url is None:
 165             return self.url_result(redirect_url, 'Generic')
 166
 167         iframe = self._download_webpage(iframe_url, video_id, 'Downloading iframe page')
 168
 169         duration = None
 170         sources = []
 171
 172         sd_url = self._search_regex(
 173             r'<source[^>]+src=(["\'])(?P<url>.+?)\1', iframe,
 174             'sd video url', default=None, group='url')
 175         if sd_url:
 176             sources.append((sd_url, 'sd'))
 177
 178         options = self._parse_json(
 179             self._search_regex(
 180                 r'data-crt-options=(["\'])(?P<options>.+?)\1', iframe,
 181                 'hd video url', default='', group='options'),
 182             video_id, fatal=False)
 183         if options:
 184             duration = int_or_none(options.get('duration'))
 185             hd_url = options.get('hdUrl')
 186             if hd_url:
 187                 sources.append((hd_url, 'hd'))
 188
 189         formats = [{
 190             'url': video_url,
 191             'ext': 'mp4',
 192             'format_id': format_id,
 193             'height': int_or_none(self._search_regex(
 194                 r'/(\d{3,4})$', video_url, 'height', default=None)),
 195             'quality': quality,
 196         } for quality, (video_url, format_id) in enumerate(sources)]
 197
 198         self._sort_formats(formats)
 199
 200         # The only place where you can get a title, it's not complete,
 201         # but searching in other places doesn't work for all videos
 202         video_title = self._html_search_regex(
 203             r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
 204             webpage, 'title')
 205
 206         return {
 207             'id': video_id,
 208             'title': video_title,
 209             'description': self._og_search_description(webpage, default=None),
 210             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 211             'duration': duration,
 212             'formats': formats,
 213         }