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|r)/[^/]+)/)?(?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         'url': 'http://imgur.com/r/aww/VQcQPhM',
  48         'only_matching': True,
  49     }]
  50
  51     def _real_extract(self, url):
  52         video_id = self._match_id(url)
  53         webpage = self._download_webpage(
  54             compat_urlparse.urljoin(url, video_id), video_id)
  55
  56         width = int_or_none(self._og_search_property(
  57             'video:width', webpage, default=None))
  58         height = int_or_none(self._og_search_property(
  59             'video:height', webpage, default=None))
  60
  61         video_elements = self._search_regex(
  62             r'(?s)<div class="video-elements">(.*?)</div>',
  63             webpage, 'video elements', default=None)
  64         if not video_elements:
  65             raise ExtractorError(
  66                 'No sources found for video %s. Maybe an image?' % video_id,
  67                 expected=True)
  68
  69         formats = []
  70         for m in re.finditer(r'<source\s+src="(?P<src>[^"]+)"\s+type="(?P<type>[^"]+)"', video_elements):
  71             formats.append({
  72                 'format_id': m.group('type').partition('/')[2],
  73                 'url': self._proto_relative_url(m.group('src')),
  74                 'ext': mimetype2ext(m.group('type')),
  75                 'acodec': 'none',
  76                 'width': width,
  77                 'height': height,
  78                 'http_headers': {
  79                     'User-Agent': 'youtube-dl (like wget)',
  80                 },
  81             })
  82
  83         gif_json = self._search_regex(
  84             r'(?s)var\s+videoItem\s*=\s*(\{.*?\})',
  85             webpage, 'GIF code', fatal=False)
  86         if gif_json:
  87             gifd = self._parse_json(
  88                 gif_json, video_id, transform_source=js_to_json)
  89             formats.append({
  90                 'format_id': 'gif',
  91                 'preference': -10,
  92                 'width': width,
  93                 'height': height,
  94                 'ext': 'gif',
  95                 'acodec': 'none',
  96                 'vcodec': 'gif',
  97                 'container': 'gif',
  98                 'url': self._proto_relative_url(gifd['gifUrl']),
  99                 'filesize': gifd.get('size'),
 100                 'http_headers': {
 101                     'User-Agent': 'youtube-dl (like wget)',
 102                 },
 103             })
 104
 105         self._sort_formats(formats)
 106
 107         return {
 108             'id': video_id,
 109             'formats': formats,
 110             'description': self._og_search_description(webpage),
 111             'title': self._og_search_title(webpage),
 112         }
 113
 114
 115 class ImgurAlbumIE(InfoExtractor):
 116     _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:a|gallery|topic/[^/]+)/)?(?P<id>[a-zA-Z0-9]{5})(?:[/?#&]+)?$'
 117
 118     _TESTS = [{
 119         'url': 'http://imgur.com/gallery/Q95ko',
 120         'info_dict': {
 121             'id': 'Q95ko',
 122         },
 123         'playlist_count': 25,
 124     }, {
 125         'url': 'http://imgur.com/a/j6Orj',
 126         'only_matching': True,
 127     }, {
 128         'url': 'http://imgur.com/topic/Aww/ll5Vk',
 129         'only_matching': True,
 130     }]
 131
 132     def _real_extract(self, url):
 133         album_id = self._match_id(url)
 134
 135         album_images = self._download_json(
 136             'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id,
 137             album_id, fatal=False)
 138
 139         if album_images:
 140             data = album_images.get('data')
 141             if data and isinstance(data, dict):
 142                 images = data.get('images')
 143                 if images and isinstance(images, list):
 144                     entries = [
 145                         self.url_result('http://imgur.com/%s' % image['hash'])
 146                         for image in images if image.get('hash')]
 147                     return self.playlist_result(entries, album_id)
 148
 149         # Fallback to single video
 150         return self.url_result('http://imgur.com/%s' % album_id, ImgurIE.ie_key())