youtube_dlc/extractor/medaltv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     ExtractorError,
  10     float_or_none,
  11     int_or_none,
  12     str_or_none,
  13     try_get,
  14 )
  15
  16
  17 class MedalTVIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?medal\.tv/clips/(?P<id>[0-9]+)'
  19     _TESTS = [{
  20         'url': 'https://medal.tv/clips/34934644/3Is9zyGMoBMr',
  21         'md5': '7b07b064331b1cf9e8e5c52a06ae68fa',
  22         'info_dict': {
  23             'id': '34934644',
  24             'ext': 'mp4',
  25             'title': 'Quad Cold',
  26             'description': 'Medal,https://medal.tv/desktop/',
  27             'uploader': 'MowgliSB',
  28             'timestamp': 1603165266,
  29             'upload_date': '20201020',
  30             'uploader_id': 10619174,
  31         }
  32     }, {
  33         'url': 'https://medal.tv/clips/36787208',
  34         'md5': 'b6dc76b78195fff0b4f8bf4a33ec2148',
  35         'info_dict': {
  36             'id': '36787208',
  37             'ext': 'mp4',
  38             'title': 'u tk me i tk u bigger',
  39             'description': 'Medal,https://medal.tv/desktop/',
  40             'uploader': 'Mimicc',
  41             'timestamp': 1605580939,
  42             'upload_date': '20201117',
  43             'uploader_id': 5156321,
  44         }
  45     }]
  46
  47     def _real_extract(self, url):
  48         video_id = self._match_id(url)
  49         webpage = self._download_webpage(url, video_id)
  50
  51         hydration_data = self._parse_json(self._search_regex(
  52             r'<script[^>]*>\s*(?:var\s*)?hydrationData\s*=\s*({.+?})\s*</script>',
  53             webpage, 'hydration data', default='{}'), video_id)
  54
  55         clip = try_get(
  56             hydration_data, lambda x: x['clips'][video_id], dict) or {}
  57         if not clip:
  58             raise ExtractorError(
  59                 'Could not find video information.', video_id=video_id)
  60
  61         title = clip['contentTitle']
  62
  63         source_width = int_or_none(clip.get('sourceWidth'))
  64         source_height = int_or_none(clip.get('sourceHeight'))
  65
  66         aspect_ratio = source_width / source_height if source_width and source_height else 16 / 9
  67
  68         def add_item(container, item_url, height, id_key='format_id', item_id=None):
  69             item_id = item_id or '%dp' % height
  70             if item_id not in item_url:
  71                 return
  72             width = int(round(aspect_ratio * height))
  73             container.append({
  74                 'url': item_url,
  75                 id_key: item_id,
  76                 'width': width,
  77                 'height': height
  78             })
  79
  80         formats = []
  81         thumbnails = []
  82         for k, v in clip.items():
  83             if not (v and isinstance(v, compat_str)):
  84                 continue
  85             mobj = re.match(r'(contentUrl|thumbnail)(?:(\d+)p)?$', k)
  86             if not mobj:
  87                 continue
  88             prefix = mobj.group(1)
  89             height = int_or_none(mobj.group(2))
  90             if prefix == 'contentUrl':
  91                 add_item(
  92                     formats, v, height or source_height,
  93                     item_id=None if height else 'source')
  94             elif prefix == 'thumbnail':
  95                 add_item(thumbnails, v, height, 'id')
  96
  97         error = clip.get('error')
  98         if not formats and error:
  99             if error == 404:
 100                 raise ExtractorError(
 101                     'That clip does not exist.',
 102                     expected=True, video_id=video_id)
 103             else:
 104                 raise ExtractorError(
 105                     'An unknown error occurred ({0}).'.format(error),
 106                     video_id=video_id)
 107
 108         self._sort_formats(formats)
 109
 110         # Necessary because the id of the author is not known in advance.
 111         # Won't raise an issue if no profile can be found as this is optional.
 112         author = try_get(
 113             hydration_data, lambda x: list(x['profiles'].values())[0], dict) or {}
 114         author_id = str_or_none(author.get('id'))
 115         author_url = 'https://medal.tv/users/{0}'.format(author_id) if author_id else None
 116
 117         return {
 118             'id': video_id,
 119             'title': title,
 120             'formats': formats,
 121             'thumbnails': thumbnails,
 122             'description': clip.get('contentDescription'),
 123             'uploader': author.get('displayName'),
 124             'timestamp': float_or_none(clip.get('created'), 1000),
 125             'uploader_id': author_id,
 126             'uploader_url': author_url,
 127             'duration': int_or_none(clip.get('videoLengthSeconds')),
 128             'view_count': int_or_none(clip.get('views')),
 129             'like_count': int_or_none(clip.get('likes')),
 130             'comment_count': int_or_none(clip.get('comments')),
 131         }