yt_dlp/extractor/box.py

   1 import json
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     parse_iso8601,
   7     update_url_query,
   8     url_or_none,
   9 )
  10 from ..utils.traversal import traverse_obj
  11
  12
  13 class BoxIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:[^.]+\.)?app\.box\.com/s/(?P<shared_name>[^/?#]+)/file/(?P<id>\d+)'
  15     _TEST = {
  16         'url': 'https://mlssoccer.app.box.com/s/0evd2o3e08l60lr4ygukepvnkord1o1x/file/510727257538',
  17         'md5': '1f81b2fd3960f38a40a3b8823e5fcd43',
  18         'info_dict': {
  19             'id': '510727257538',
  20             'ext': 'mp4',
  21             'title': 'Garber   St. Louis will be 28th MLS team  +scarving.mp4',
  22             'uploader': '',
  23             'timestamp': 1566320259,
  24             'upload_date': '20190820',
  25             'uploader_id': '235196876',
  26         },
  27         'params': {'skip_download': 'dash fragment too small'},
  28     }
  29
  30     def _real_extract(self, url):
  31         shared_name, file_id = self._match_valid_url(url).groups()
  32         webpage = self._download_webpage(url, file_id)
  33         request_token = self._parse_json(self._search_regex(
  34             r'Box\.config\s*=\s*({.+?});', webpage,
  35             'Box config'), file_id)['requestToken']
  36         access_token = self._download_json(
  37             'https://app.box.com/app-api/enduserapp/elements/tokens', file_id,
  38             'Downloading token JSON metadata',
  39             data=json.dumps({'fileIDs': [file_id]}).encode(), headers={
  40                 'Content-Type': 'application/json',
  41                 'X-Request-Token': request_token,
  42                 'X-Box-EndUser-API': 'sharedName=' + shared_name,
  43             })[file_id]['read']
  44         shared_link = 'https://app.box.com/s/' + shared_name
  45         f = self._download_json(
  46             'https://api.box.com/2.0/files/' + file_id, file_id,
  47             'Downloading file JSON metadata', headers={
  48                 'Authorization': 'Bearer ' + access_token,
  49                 'BoxApi': 'shared_link=' + shared_link,
  50                 'X-Rep-Hints': '[dash]',  # TODO: extract `hls` formats
  51             }, query={
  52                 'fields': 'authenticated_download_url,created_at,created_by,description,extension,is_download_available,name,representations,size'
  53             })
  54         title = f['name']
  55
  56         query = {
  57             'access_token': access_token,
  58             'shared_link': shared_link
  59         }
  60
  61         formats = []
  62
  63         for url_tmpl in traverse_obj(f, (
  64             'representations', 'entries', lambda _, v: v['representation'] == 'dash',
  65             'content', 'url_template', {url_or_none}
  66         )):
  67             manifest_url = update_url_query(url_tmpl.replace('{+asset_path}', 'manifest.mpd'), query)
  68             fmts = self._extract_mpd_formats(manifest_url, file_id)
  69             for fmt in fmts:
  70                 fmt['extra_param_to_segment_url'] = urllib.parse.urlparse(manifest_url).query
  71             formats.extend(fmts)
  72
  73         creator = f.get('created_by') or {}
  74
  75         return {
  76             'id': file_id,
  77             'title': title,
  78             'formats': formats,
  79             'description': f.get('description') or None,
  80             'uploader': creator.get('name'),
  81             'timestamp': parse_iso8601(f.get('created_at')),
  82             'uploader_id': creator.get('id'),
  83         }