yt_dlp/extractor/imgur.py

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