yt_dlp/extractor/gfycat.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     float_or_none,
  10     qualities,
  11     ExtractorError,
  12 )
  13
  14
  15 class GfycatIE(InfoExtractor):
  16     _VALID_URL = r'(?i)https?://(?:(?:www|giant|thumbs)\.)?gfycat\.com/(?:ru/|ifr/|gifs/detail/)?(?P<id>[^-/?#\."\']+)'
  17     _TESTS = [{
  18         'url': 'http://gfycat.com/DeadlyDecisiveGermanpinscher',
  19         'info_dict': {
  20             'id': 'DeadlyDecisiveGermanpinscher',
  21             'ext': 'mp4',
  22             'title': 'Ghost in the Shell',
  23             'timestamp': 1410656006,
  24             'upload_date': '20140914',
  25             'uploader': 'anonymous',
  26             'duration': 10.4,
  27             'view_count': int,
  28             'like_count': int,
  29             'categories': list,
  30             'age_limit': 0,
  31             'uploader_id': 'anonymous',
  32             'description': '',
  33         }
  34     }, {
  35         'url': 'http://gfycat.com/ifr/JauntyTimelyAmazontreeboa',
  36         'info_dict': {
  37             'id': 'JauntyTimelyAmazontreeboa',
  38             'ext': 'mp4',
  39             'title': 'JauntyTimelyAmazontreeboa',
  40             'timestamp': 1411720126,
  41             'upload_date': '20140926',
  42             'uploader': 'anonymous',
  43             'duration': 3.52,
  44             'view_count': int,
  45             'like_count': int,
  46             'categories': list,
  47             'age_limit': 0,
  48             'uploader_id': 'anonymous',
  49             'description': '',
  50         }
  51     }, {
  52         'url': 'https://gfycat.com/alienatedsolidgreathornedowl',
  53         'info_dict': {
  54             'id': 'alienatedsolidgreathornedowl',
  55             'ext': 'mp4',
  56             'upload_date': '20211226',
  57             'uploader_id': 'reactions',
  58             'timestamp': 1640536930,
  59             'like_count': int,
  60             'description': '',
  61             'title': 'Ingrid Michaelson, Zooey Deschanel - Merry Christmas Happy New Year',
  62             'categories': list,
  63             'age_limit': 0,
  64             'duration': 2.9583333333333335,
  65             'uploader': 'Reaction GIFs',
  66             'view_count': int,
  67         }
  68     }, {
  69         'url': 'https://gfycat.com/ru/RemarkableDrearyAmurstarfish',
  70         'only_matching': True
  71     }, {
  72         'url': 'https://gfycat.com/gifs/detail/UnconsciousLankyIvorygull',
  73         'only_matching': True
  74     }, {
  75         'url': 'https://gfycat.com/acceptablehappygoluckyharborporpoise-baseball',
  76         'only_matching': True
  77     }, {
  78         'url': 'https://thumbs.gfycat.com/acceptablehappygoluckyharborporpoise-size_restricted.gif',
  79         'only_matching': True
  80     }, {
  81         'url': 'https://giant.gfycat.com/acceptablehappygoluckyharborporpoise.mp4',
  82         'only_matching': True
  83     }, {
  84         'url': 'http://gfycat.com/IFR/JauntyTimelyAmazontreeboa',
  85         'only_matching': True
  86     }]
  87
  88     @staticmethod
  89     def _extract_urls(webpage):
  90         return [
  91             mobj.group('url')
  92             for mobj in re.finditer(
  93                 r'<(?:iframe|source)[^>]+\bsrc=["\'](?P<url>%s)' % GfycatIE._VALID_URL,
  94                 webpage)]
  95
  96     def _real_extract(self, url):
  97         video_id = self._match_id(url)
  98
  99         gfy = self._download_json(
 100             'https://api.gfycat.com/v1/gfycats/%s' % video_id,
 101             video_id, 'Downloading video info')
 102         if 'error' in gfy:
 103             raise ExtractorError('Gfycat said: ' + gfy['error'], expected=True)
 104         gfy = gfy['gfyItem']
 105
 106         title = gfy.get('title') or gfy['gfyName']
 107         description = gfy.get('description')
 108         timestamp = int_or_none(gfy.get('createDate'))
 109         uploader = gfy.get('userName') or gfy.get('username')
 110         view_count = int_or_none(gfy.get('views'))
 111         like_count = int_or_none(gfy.get('likes'))
 112         dislike_count = int_or_none(gfy.get('dislikes'))
 113         age_limit = 18 if gfy.get('nsfw') == '1' else 0
 114
 115         width = int_or_none(gfy.get('width'))
 116         height = int_or_none(gfy.get('height'))
 117         fps = int_or_none(gfy.get('frameRate'))
 118         num_frames = int_or_none(gfy.get('numFrames'))
 119
 120         duration = float_or_none(num_frames, fps) if num_frames and fps else None
 121
 122         categories = gfy.get('tags') or gfy.get('extraLemmas') or []
 123
 124         FORMATS = ('gif', 'webm', 'mp4')
 125         quality = qualities(FORMATS)
 126
 127         formats = []
 128         for format_id in FORMATS:
 129             video_url = gfy.get('%sUrl' % format_id)
 130             if not video_url:
 131                 continue
 132             filesize = int_or_none(gfy.get('%sSize' % format_id))
 133             formats.append({
 134                 'url': video_url,
 135                 'format_id': format_id,
 136                 'width': width,
 137                 'height': height,
 138                 'fps': fps,
 139                 'filesize': filesize,
 140                 'quality': quality(format_id),
 141             })
 142         self._sort_formats(formats)
 143
 144         return {
 145             'id': video_id,
 146             'title': title,
 147             'description': description,
 148             'timestamp': timestamp,
 149             'uploader': gfy.get('userDisplayName') or uploader,
 150             'uploader_id': uploader,
 151             'duration': duration,
 152             'view_count': view_count,
 153             'like_count': like_count,
 154             'dislike_count': dislike_count,
 155             'categories': categories,
 156             'age_limit': age_limit,
 157             'formats': formats,
 158         }