_ 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
  32     def _real_extract(self, url):
  33         video_id = self._match_id(url)
  34         webpage = self._download_webpage(
  35             'https://i.imgur.com/{id}.gifv'.format(id=video_id), video_id)
  36
  37         width = int_or_none(self._og_search_property(
  38             'video:width', webpage, default=None))
  39         height = int_or_none(self._og_search_property(
  40             'video:height', webpage, default=None))
  41
  42         video_elements = self._search_regex(
  43             r'(?s)<div class="video-elements">(.*?)</div>',
  44             webpage, 'video elements', default=None)
  45         if not video_elements:
  46             raise ExtractorError(
  47                 'No sources found for video %s. Maybe an image?' % video_id,
  48                 expected=True)
  49
  50         formats = []
  51         for m in re.finditer(r'<source\s+src="(?P<src>[^"]+)"\s+type="(?P<type>[^"]+)"', video_elements):
  52             formats.append({
  53                 'format_id': m.group('type').partition('/')[2],
  54                 'url': self._proto_relative_url(m.group('src')),
  55                 'ext': mimetype2ext(m.group('type')),
  56                 'width': width,
  57                 'height': height,
  58                 'http_headers': {
  59                     'User-Agent': 'youtube-dl (like wget)',
  60                 },
  61             })
  62
  63         gif_json = self._search_regex(
  64             r'(?s)var\s+videoItem\s*=\s*(\{.*?\})',
  65             webpage, 'GIF code', fatal=False)
  66         if gif_json:
  67             gifd = self._parse_json(
  68                 gif_json, video_id, transform_source=js_to_json)
  69             formats.append({
  70                 'format_id': 'gif',
  71                 'preference': -10,
  72                 'width': width,
  73                 'height': height,
  74                 'ext': 'gif',
  75                 'acodec': 'none',
  76                 'vcodec': 'gif',
  77                 'container': 'gif',
  78                 'url': self._proto_relative_url(gifd['gifUrl']),
  79                 'filesize': gifd.get('size'),
  80                 'http_headers': {
  81                     'User-Agent': 'youtube-dl (like wget)',
  82                 },
  83             })
  84
  85         self._sort_formats(formats)
  86
  87         return {
  88             'id': video_id,
  89             'formats': formats,
  90             'title': self._og_search_title(webpage),
  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):
 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     }]