_ Git - youtube-dl/blob - youtube_dl/extractor/imgur.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     int_or_none,
   8     js_to_json,
   9     mimetype2ext,
  10     ExtractorError,
  11 )
  12
  13
  14 class ImgurIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?!(?:a|gallery|(?:t(?:opic)?|r)/[^/]+)/)(?P<id>[a-zA-Z0-9]+)'
  16
  17     _TESTS = [{
  18         'url': 'https://i.imgur.com/A61SaA1.gifv',
  19         'info_dict': {
  20             'id': 'A61SaA1',
  21             'ext': 'mp4',
  22             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
  23         },
  24     }, {
  25         'url': 'https://imgur.com/A61SaA1',
  26         'only_matching': True,
  27     }, {
  28         'url': 'https://i.imgur.com/crGpqCV.mp4',
  29         'only_matching': True,
  30     }, {
  31         # no title
  32         'url': 'https://i.imgur.com/jxBXAMC.gifv',
  33         'only_matching': True,
  34     }]
  35
  36     def _real_extract(self, url):
  37         video_id = self._match_id(url)
  38         webpage = self._download_webpage(
  39             'https://i.imgur.com/{id}.gifv'.format(id=video_id), video_id)
  40
  41         width = int_or_none(self._og_search_property(
  42             'video:width', webpage, default=None))
  43         height = int_or_none(self._og_search_property(
  44             'video:height', webpage, default=None))
  45
  46         video_elements = self._search_regex(
  47             r'(?s)<div class="video-elements">(.*?)</div>',
  48             webpage, 'video elements', default=None)
  49         if not video_elements:
  50             raise ExtractorError(
  51                 'No sources found for video %s. Maybe an image?' % video_id,
  52                 expected=True)
  53
  54         formats = []
  55         for m in re.finditer(r'<source\s+src="(?P<src>[^"]+)"\s+type="(?P<type>[^"]+)"', video_elements):
  56             formats.append({
  57                 'format_id': m.group('type').partition('/')[2],
  58                 'url': self._proto_relative_url(m.group('src')),
  59                 'ext': mimetype2ext(m.group('type')),
  60                 'width': width,
  61                 'height': height,
  62                 'http_headers': {
  63                     'User-Agent': 'youtube-dl (like wget)',
  64                 },
  65             })
  66
  67         gif_json = self._search_regex(
  68             r'(?s)var\s+videoItem\s*=\s*(\{.*?\})',
  69             webpage, 'GIF code', fatal=False)
  70         if gif_json:
  71             gifd = self._parse_json(
  72                 gif_json, video_id, transform_source=js_to_json)
  73             formats.append({
  74                 'format_id': 'gif',
  75                 'preference': -10,
  76                 'width': width,
  77                 'height': height,
  78                 'ext': 'gif',
  79                 'acodec': 'none',
  80                 'vcodec': 'gif',
  81                 'container': 'gif',
  82                 'url': self._proto_relative_url(gifd['gifUrl']),
  83                 'filesize': gifd.get('size'),
  84                 'http_headers': {
  85                     'User-Agent': 'youtube-dl (like wget)',
  86                 },
  87             })
  88
  89         self._sort_formats(formats)
  90
  91         return {
  92             'id': video_id,
  93             'formats': formats,
  94             'title': self._og_search_title(webpage, default=video_id),
  95         }
  96
  97
  98 class ImgurGalleryIE(InfoExtractor):
  99     IE_NAME = 'imgur:gallery'
 100     _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:gallery|(?:t(?:opic)?|r)/[^/]+)/(?P<id>[a-zA-Z0-9]+)'
 101
 102     _TESTS = [{
 103         'url': 'http://imgur.com/gallery/Q95ko',
 104         'info_dict': {
 105             'id': 'Q95ko',
 106             'title': 'Adding faces make every GIF better',
 107         },
 108         'playlist_count': 25,
 109     }, {
 110         'url': 'http://imgur.com/topic/Aww/ll5Vk',
 111         'only_matching': True,
 112     }, {
 113         'url': 'https://imgur.com/gallery/YcAQlkx',
 114         'info_dict': {
 115             'id': 'YcAQlkx',
 116             'ext': 'mp4',
 117             'title': 'Classic Steve Carell gif...cracks me up everytime....damn the repost downvotes....',
 118         }
 119     }, {
 120         'url': 'http://imgur.com/topic/Funny/N8rOudd',
 121         'only_matching': True,
 122     }, {
 123         'url': 'http://imgur.com/r/aww/VQcQPhM',
 124         'only_matching': True,
 125     }]
 126
 127     def _real_extract(self, url):
 128         gallery_id = self._match_id(url)
 129
 130         data = self._download_json(
 131             'https://imgur.com/gallery/%s.json' % gallery_id,
 132             gallery_id)['data']['image']
 133
 134         if data.get('is_album'):
 135             entries = [
 136                 self.url_result('http://imgur.com/%s' % image['hash'], ImgurIE.ie_key(), image['hash'])
 137                 for image in data['album_images']['images'] if image.get('hash')]
 138             return self.playlist_result(entries, gallery_id, data.get('title'), data.get('description'))
 139
 140         return self.url_result('http://imgur.com/%s' % gallery_id, ImgurIE.ie_key(), gallery_id)
 141
 142
 143 class ImgurAlbumIE(ImgurGalleryIE):
 144     IE_NAME = 'imgur:album'
 145     _VALID_URL = r'https?://(?:i\.)?imgur\.com/a/(?P<id>[a-zA-Z0-9]+)'
 146
 147     _TESTS = [{
 148         'url': 'http://imgur.com/a/j6Orj',
 149         'info_dict': {
 150             'id': 'j6Orj',
 151             'title': 'A Literary Analysis of "Star Wars: The Force Awakens"',
 152         },
 153         'playlist_count': 12,
 154     }]