1 from __future__ import unicode_literals
5 from .common import InfoExtractor
6 from ..compat import compat_urlparse
15 class ImgurIE(InfoExtractor):
16 _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:gallery|topic/[^/]+)/)?(?P<id>[a-zA-Z0-9]{6,})(?:[/?#&]+|\.[a-z]+)?$'
19 'url': 'https://i.imgur.com/A61SaA1.gifv',
23 'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
24 'description': 'Imgur: The most awesome images on the Internet.',
27 'url': 'https://imgur.com/A61SaA1',
31 'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
32 'description': 'Imgur: The most awesome images on the Internet.',
35 'url': 'https://imgur.com/gallery/YcAQlkx',
39 'title': 'Classic Steve Carell gif...cracks me up everytime....damn the repost downvotes....',
40 'description': 'Imgur: The most awesome images on the Internet.'
44 'url': 'http://imgur.com/topic/Funny/N8rOudd',
45 'only_matching': True,
48 def _real_extract(self, url):
49 video_id = self._match_id(url)
50 webpage = self._download_webpage(
51 compat_urlparse.urljoin(url, video_id), video_id)
53 width = int_or_none(self._search_regex(
54 r'<param name="width" value="([0-9]+)"',
55 webpage, 'width', fatal=False))
56 height = int_or_none(self._search_regex(
57 r'<param name="height" value="([0-9]+)"',
58 webpage, 'height', fatal=False))
60 video_elements = self._search_regex(
61 r'(?s)<div class="video-elements">(.*?)</div>',
62 webpage, 'video elements', default=None)
63 if not video_elements:
65 'No sources found for video %s. Maybe an image?' % video_id,
69 for m in re.finditer(r'<source\s+src="(?P<src>[^"]+)"\s+type="(?P<type>[^"]+)"', video_elements):
71 'format_id': m.group('type').partition('/')[2],
72 'url': self._proto_relative_url(m.group('src')),
73 'ext': mimetype2ext(m.group('type')),
78 'User-Agent': 'youtube-dl (like wget)',
82 gif_json = self._search_regex(
83 r'(?s)var\s+videoItem\s*=\s*(\{.*?\})',
84 webpage, 'GIF code', fatal=False)
86 gifd = self._parse_json(
87 gif_json, video_id, transform_source=js_to_json)
97 'url': self._proto_relative_url(gifd['gifUrl']),
98 'filesize': gifd.get('size'),
100 'User-Agent': 'youtube-dl (like wget)',
104 self._sort_formats(formats)
109 'description': self._og_search_description(webpage),
110 'title': self._og_search_title(webpage),
114 class ImgurAlbumIE(InfoExtractor):
115 _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:a|gallery|topic/[^/]+)/)?(?P<id>[a-zA-Z0-9]{5})(?:[/?#&]+)?$'
118 'url': 'http://imgur.com/gallery/Q95ko',
122 'playlist_count': 25,
124 'url': 'http://imgur.com/a/j6Orj',
125 'only_matching': True,
127 'url': 'http://imgur.com/topic/Aww/ll5Vk',
128 'only_matching': True,
131 def _real_extract(self, url):
132 album_id = self._match_id(url)
134 album_images = self._download_json(
135 'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id,
136 album_id, fatal=False)
139 data = album_images.get('data')
140 if data and isinstance(data, dict):
141 images = data.get('images')
142 if images and isinstance(images, list):
144 self.url_result('http://imgur.com/%s' % image['hash'])
145 for image in images if image.get('hash')]
146 return self.playlist_result(entries, album_id)
148 # Fallback to single video
149 return self.url_result('http://imgur.com/%s' % album_id, ImgurIE.ie_key())