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 ..compat import compat_str
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     urlencode_postdata
  12 )
  13
  14
  15 class TumblrIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?P<blog_name>[^/?#&]+)\.tumblr\.com/(?:post|video)/(?P<id>[0-9]+)(?:$|[/?#])'
  17     _NETRC_MACHINE = 'tumblr'
  18     _LOGIN_URL = 'https://www.tumblr.com/login'
  19     _TESTS = [{
  20         'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
  21         'md5': '479bb068e5b16462f5176a6828829767',
  22         'info_dict': {
  23             'id': '54196191430',
  24             'ext': 'mp4',
  25             'title': 'tatiana maslany news, Orphan Black || DVD extra - behind the scenes ↳...',
  26             'description': 'md5:37db8211e40b50c7c44e95da14f630b7',
  27             'thumbnail': r're:http://.*\.jpg',
  28         }
  29     }, {
  30         'url': 'http://5sostrum.tumblr.com/post/90208453769/yall-forgetting-the-greatest-keek-of-them-all',
  31         'md5': 'bf348ef8c0ef84fbf1cbd6fa6e000359',
  32         'info_dict': {
  33             'id': '90208453769',
  34             'ext': 'mp4',
  35             'title': '5SOS STRUM ;]',
  36             'description': 'md5:dba62ac8639482759c8eb10ce474586a',
  37             'thumbnail': r're:http://.*\.jpg',
  38         }
  39     }, {
  40         'url': 'http://hdvideotest.tumblr.com/post/130323439814/test-description-for-my-hd-video',
  41         'md5': '7ae503065ad150122dc3089f8cf1546c',
  42         'info_dict': {
  43             'id': '130323439814',
  44             'ext': 'mp4',
  45             'title': 'HD Video Testing \u2014 Test description for my HD video',
  46             'description': 'md5:97cc3ab5fcd27ee4af6356701541319c',
  47             'thumbnail': r're:http://.*\.jpg',
  48         },
  49         'params': {
  50             'format': 'hd',
  51         },
  52     }, {
  53         'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
  54         'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
  55         'info_dict': {
  56             'id': 'Wmur',
  57             'ext': 'mp4',
  58             'title': 'naked smoking & stretching',
  59             'upload_date': '20150506',
  60             'timestamp': 1430931613,
  61             'age_limit': 18,
  62             'uploader_id': '1638622',
  63             'uploader': 'naked-yogi',
  64         },
  65         'add_ie': ['Vidme'],
  66     }, {
  67         'url': 'http://camdamage.tumblr.com/post/98846056295/',
  68         'md5': 'a9e0c8371ea1ca306d6554e3fecf50b6',
  69         'info_dict': {
  70             'id': '105463834',
  71             'ext': 'mp4',
  72             'title': 'Cam Damage-HD 720p',
  73             'uploader': 'John Moyer',
  74             'uploader_id': 'user32021558',
  75         },
  76         'add_ie': ['Vimeo'],
  77     }, {
  78         'url': 'http://sutiblr.tumblr.com/post/139638707273',
  79         'md5': '2dd184b3669e049ba40563a7d423f95c',
  80         'info_dict': {
  81             'id': 'ir7qBEIKqvq',
  82             'ext': 'mp4',
  83             'title': 'Vine by sutiblr',
  84             'alt_title': 'Vine by sutiblr',
  85             'uploader': 'sutiblr',
  86             'uploader_id': '1198993975374495744',
  87             'upload_date': '20160220',
  88             'like_count': int,
  89             'comment_count': int,
  90             'repost_count': int,
  91         },
  92         'add_ie': ['Vine'],
  93     }, {
  94         'url': 'http://vitasidorkina.tumblr.com/post/134652425014/joskriver-victoriassecret-invisibility-or',
  95         'md5': '01c12ceb82cbf6b2fe0703aa56b3ad72',
  96         'info_dict': {
  97             'id': '-7LnUPGlSo',
  98             'ext': 'mp4',
  99             'title': 'Video by victoriassecret',
 100             'description': 'Invisibility or flight…which superpower would YOU choose? #VSFashionShow #ThisOrThat',
 101             'uploader_id': 'victoriassecret',
 102             'thumbnail': r're:^https?://.*\.jpg'
 103         },
 104         'add_ie': ['Instagram'],
 105     }]
 106
 107     def _real_initialize(self):
 108         self._login()
 109
 110     def _login(self):
 111         username, password = self._get_login_info()
 112         if username is None:
 113             return
 114
 115         login_page = self._download_webpage(
 116             self._LOGIN_URL, None, 'Downloading login page')
 117
 118         login_form = self._hidden_inputs(login_page)
 119         login_form.update({
 120             'user[email]': username,
 121             'user[password]': password
 122         })
 123
 124         response, urlh = self._download_webpage_handle(
 125             self._LOGIN_URL, None, 'Logging in',
 126             data=urlencode_postdata(login_form), headers={
 127                 'Content-Type': 'application/x-www-form-urlencoded',
 128                 'Referer': self._LOGIN_URL,
 129             })
 130
 131         # Successful login
 132         if '/dashboard' in urlh.geturl():
 133             return
 134
 135         login_errors = self._parse_json(
 136             self._search_regex(
 137                 r'RegistrationForm\.errors\s*=\s*(\[.+?\])\s*;', response,
 138                 'login errors', default='[]'),
 139             None, fatal=False)
 140         if login_errors:
 141             raise ExtractorError(
 142                 'Unable to login: %s' % login_errors[0], expected=True)
 143
 144         self.report_warning('Login has probably failed')
 145
 146     def _real_extract(self, url):
 147         m_url = re.match(self._VALID_URL, url)
 148         video_id = m_url.group('id')
 149         blog = m_url.group('blog_name')
 150
 151         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
 152         webpage, urlh = self._download_webpage_handle(url, video_id)
 153
 154         redirect_url = compat_str(urlh.geturl())
 155         if 'tumblr.com/safe-mode' in redirect_url or redirect_url.startswith('/safe-mode'):
 156             raise ExtractorError(
 157                 'This Tumblr may contain sensitive media. '
 158                 'Disable safe mode in your account settings '
 159                 'at https://www.tumblr.com/settings/account#safe_mode',
 160                 expected=True)
 161
 162         iframe_url = self._search_regex(
 163             r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
 164             webpage, 'iframe url', default=None)
 165         if iframe_url is None:
 166             return self.url_result(redirect_url, 'Generic')
 167
 168         iframe = self._download_webpage(iframe_url, video_id, 'Downloading iframe page')
 169
 170         duration = None
 171         sources = []
 172
 173         sd_url = self._search_regex(
 174             r'<source[^>]+src=(["\'])(?P<url>.+?)\1', iframe,
 175             'sd video url', default=None, group='url')
 176         if sd_url:
 177             sources.append((sd_url, 'sd'))
 178
 179         options = self._parse_json(
 180             self._search_regex(
 181                 r'data-crt-options=(["\'])(?P<options>.+?)\1', iframe,
 182                 'hd video url', default='', group='options'),
 183             video_id, fatal=False)
 184         if options:
 185             duration = int_or_none(options.get('duration'))
 186             hd_url = options.get('hdUrl')
 187             if hd_url:
 188                 sources.append((hd_url, 'hd'))
 189
 190         formats = [{
 191             'url': video_url,
 192             'ext': 'mp4',
 193             'format_id': format_id,
 194             'height': int_or_none(self._search_regex(
 195                 r'/(\d{3,4})$', video_url, 'height', default=None)),
 196             'quality': quality,
 197         } for quality, (video_url, format_id) in enumerate(sources)]
 198
 199         self._sort_formats(formats)
 200
 201         # The only place where you can get a title, it's not complete,
 202         # but searching in other places doesn't work for all videos
 203         video_title = self._html_search_regex(
 204             r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
 205             webpage, 'title')
 206
 207         return {
 208             'id': video_id,
 209             'title': video_title,
 210             'description': self._og_search_description(webpage, default=None),
 211             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 212             'duration': duration,
 213             'formats': formats,
 214         }