yt_dlp/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>[^/?#&]+)'
  19     _TESTS = [{
  20         'url': 'https://medal.tv/clips/2mA60jWAGQCBH',
  21         'md5': '7b07b064331b1cf9e8e5c52a06ae68fa',
  22         'info_dict': {
  23             'id': '2mA60jWAGQCBH',
  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/2um24TWdty0NA',
  34         'md5': 'b6dc76b78195fff0b4f8bf4a33ec2148',
  35         'info_dict': {
  36             'id': '2um24TWdty0NA',
  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         'url': 'https://medal.tv/clips/37rMeFpryCC-9',
  47         'only_matching': True,
  48     }, {
  49         'url': 'https://medal.tv/clips/2WRj40tpY_EU9',
  50         'only_matching': True,
  51     }]
  52
  53     def _real_extract(self, url):
  54         video_id = self._match_id(url)
  55         webpage = self._download_webpage(url, video_id)
  56
  57         hydration_data = self._parse_json(self._search_regex(
  58             r'<script[^>]*>\s*(?:var\s*)?hydrationData\s*=\s*({.+?})\s*</script>',
  59             webpage, 'hydration data', default='{}'), video_id)
  60
  61         clip = try_get(
  62             hydration_data, lambda x: x['clips'][video_id], dict) or {}
  63         if not clip:
  64             raise ExtractorError(
  65                 'Could not find video information.', video_id=video_id)
  66
  67         title = clip['contentTitle']
  68
  69         source_width = int_or_none(clip.get('sourceWidth'))
  70         source_height = int_or_none(clip.get('sourceHeight'))
  71
  72         aspect_ratio = source_width / source_height if source_width and source_height else 16 / 9
  73
  74         def add_item(container, item_url, height, id_key='format_id', item_id=None):
  75             item_id = item_id or '%dp' % height
  76             if item_id not in item_url:
  77                 return
  78             width = int(round(aspect_ratio * height))
  79             container.append({
  80                 'url': item_url,
  81                 id_key: item_id,
  82                 'width': width,
  83                 'height': height
  84             })
  85
  86         formats = []
  87         thumbnails = []
  88         for k, v in clip.items():
  89             if not (v and isinstance(v, compat_str)):
  90                 continue
  91             mobj = re.match(r'(contentUrl|thumbnail)(?:(\d+)p)?$', k)
  92             if not mobj:
  93                 continue
  94             prefix = mobj.group(1)
  95             height = int_or_none(mobj.group(2))
  96             if prefix == 'contentUrl':
  97                 add_item(
  98                     formats, v, height or source_height,
  99                     item_id=None if height else 'source')
 100             elif prefix == 'thumbnail':
 101                 add_item(thumbnails, v, height, 'id')
 102
 103         error = clip.get('error')
 104         if not formats and error:
 105             if error == 404:
 106                 self.raise_no_formats(
 107                     'That clip does not exist.',
 108                     expected=True, video_id=video_id)
 109             else:
 110                 self.raise_no_formats(
 111                     'An unknown error occurred ({0}).'.format(error),
 112                     video_id=video_id)
 113
 114         self._sort_formats(formats)
 115
 116         # Necessary because the id of the author is not known in advance.
 117         # Won't raise an issue if no profile can be found as this is optional.
 118         author = try_get(
 119             hydration_data, lambda x: list(x['profiles'].values())[0], dict) or {}
 120         author_id = str_or_none(author.get('id'))
 121         author_url = 'https://medal.tv/users/{0}'.format(author_id) if author_id else None
 122
 123         return {
 124             'id': video_id,
 125             'title': title,
 126             'formats': formats,
 127             'thumbnails': thumbnails,
 128             'description': clip.get('contentDescription'),
 129             'uploader': author.get('displayName'),
 130             'timestamp': float_or_none(clip.get('created'), 1000),
 131             'uploader_id': author_id,
 132             'uploader_url': author_url,
 133             'duration': int_or_none(clip.get('videoLengthSeconds')),
 134             'view_count': int_or_none(clip.get('views')),
 135             'like_count': int_or_none(clip.get('likes')),
 136             'comment_count': int_or_none(clip.get('comments')),
 137         }