Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/imgur.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_urlparse
   7 from ..utils import (
   8     int_or_none,
   9     js_to_json,
  10     mimetype2ext,
  11     ExtractorError,
  12 )
  13
  14
  15 class ImgurIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:gallery|topic/[^/]+)/)?(?P<id>[a-zA-Z0-9]{6,})(?:[/?#&]+|\.[a-z]+)?$'
  17
  18     _TESTS = [{
  19         'url': 'https://i.imgur.com/A61SaA1.gifv',
  20         'info_dict': {
  21             'id': 'A61SaA1',
  22             'ext': 'mp4',
  23             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
  24             'description': 'Imgur: The most awesome images on the Internet.',
  25         },
  26     }, {
  27         'url': 'https://imgur.com/A61SaA1',
  28         'info_dict': {
  29             'id': 'A61SaA1',
  30             'ext': 'mp4',
  31             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
  32             'description': 'Imgur: The most awesome images on the Internet.',
  33         },
  34     }, {
  35         'url': 'https://imgur.com/gallery/YcAQlkx',
  36         'info_dict': {
  37             'id': 'YcAQlkx',
  38             'ext': 'mp4',
  39             'title': 'Classic Steve Carell gif...cracks me up everytime....damn the repost downvotes....',
  40             'description': 'Imgur: The most awesome images on the Internet.'
  41
  42         }
  43     }, {
  44         'url': 'http://imgur.com/topic/Funny/N8rOudd',
  45         'only_matching': True,
  46     }]
  47
  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)
  52
  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))
  59
  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:
  64             raise ExtractorError(
  65                 'No sources found for video %s. Maybe an image?' % video_id,
  66                 expected=True)
  67
  68         formats = []
  69         for m in re.finditer(r'<source\s+src="(?P<src>[^"]+)"\s+type="(?P<type>[^"]+)"', video_elements):
  70             formats.append({
  71                 'format_id': m.group('type').partition('/')[2],
  72                 'url': self._proto_relative_url(m.group('src')),
  73                 'ext': mimetype2ext(m.group('type')),
  74                 'acodec': 'none',
  75                 'width': width,
  76                 'height': height,
  77                 'http_headers': {
  78                     'User-Agent': 'youtube-dl (like wget)',
  79                 },
  80             })
  81
  82         gif_json = self._search_regex(
  83             r'(?s)var\s+videoItem\s*=\s*(\{.*?\})',
  84             webpage, 'GIF code', fatal=False)
  85         if gif_json:
  86             gifd = self._parse_json(
  87                 gif_json, video_id, transform_source=js_to_json)
  88             formats.append({
  89                 'format_id': 'gif',
  90                 'preference': -10,
  91                 'width': width,
  92                 'height': height,
  93                 'ext': 'gif',
  94                 'acodec': 'none',
  95                 'vcodec': 'gif',
  96                 'container': 'gif',
  97                 'url': self._proto_relative_url(gifd['gifUrl']),
  98                 'filesize': gifd.get('size'),
  99                 'http_headers': {
 100                     'User-Agent': 'youtube-dl (like wget)',
 101                 },
 102             })
 103
 104         self._sort_formats(formats)
 105
 106         return {
 107             'id': video_id,
 108             'formats': formats,
 109             'description': self._og_search_description(webpage),
 110             'title': self._og_search_title(webpage),
 111         }
 112
 113
 114 class ImgurAlbumIE(InfoExtractor):
 115     _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:a|gallery|topic/[^/]+)/)?(?P<id>[a-zA-Z0-9]{5})(?:[/?#&]+)?$'
 116
 117     _TESTS = [{
 118         'url': 'http://imgur.com/gallery/Q95ko',
 119         'info_dict': {
 120             'id': 'Q95ko',
 121         },
 122         'playlist_count': 25,
 123     }, {
 124         'url': 'http://imgur.com/a/j6Orj',
 125         'only_matching': True,
 126     }, {
 127         'url': 'http://imgur.com/topic/Aww/ll5Vk',
 128         'only_matching': True,
 129     }]
 130
 131     def _real_extract(self, url):
 132         album_id = self._match_id(url)
 133
 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)
 137
 138         if album_images:
 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):
 143                     entries = [
 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)
 147
 148         # Fallback to single video
 149         return self.url_result('http://imgur.com/%s' % album_id, ImgurIE.ie_key())