yt_dlp/extractor/box.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     parse_iso8601,
  10     # try_get,
  11     update_url_query,
  12 )
  13
  14
  15 class BoxIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:[^.]+\.)?app\.box\.com/s/(?P<shared_name>[^/]+)/file/(?P<id>\d+)'
  17     _TEST = {
  18         'url': 'https://mlssoccer.app.box.com/s/0evd2o3e08l60lr4ygukepvnkord1o1x/file/510727257538',
  19         'md5': '1f81b2fd3960f38a40a3b8823e5fcd43',
  20         'info_dict': {
  21             'id': '510727257538',
  22             'ext': 'mp4',
  23             'title': 'Garber   St. Louis will be 28th MLS team  +scarving.mp4',
  24             'uploader': 'MLS Video',
  25             'timestamp': 1566320259,
  26             'upload_date': '20190820',
  27             'uploader_id': '235196876',
  28         }
  29     }
  30
  31     def _real_extract(self, url):
  32         shared_name, file_id = self._match_valid_url(url).groups()
  33         webpage = self._download_webpage(url, file_id)
  34         request_token = self._parse_json(self._search_regex(
  35             r'Box\.config\s*=\s*({.+?});', webpage,
  36             'Box config'), file_id)['requestToken']
  37         access_token = self._download_json(
  38             'https://app.box.com/app-api/enduserapp/elements/tokens', file_id,
  39             'Downloading token JSON metadata',
  40             data=json.dumps({'fileIDs': [file_id]}).encode(), headers={
  41                 'Content-Type': 'application/json',
  42                 'X-Request-Token': request_token,
  43                 'X-Box-EndUser-API': 'sharedName=' + shared_name,
  44             })[file_id]['read']
  45         shared_link = 'https://app.box.com/s/' + shared_name
  46         f = self._download_json(
  47             'https://api.box.com/2.0/files/' + file_id, file_id,
  48             'Downloading file JSON metadata', headers={
  49                 'Authorization': 'Bearer ' + access_token,
  50                 'BoxApi': 'shared_link=' + shared_link,
  51                 'X-Rep-Hints': '[dash]',  # TODO: extract `hls` formats
  52             }, query={
  53                 'fields': 'authenticated_download_url,created_at,created_by,description,extension,is_download_available,name,representations,size'
  54             })
  55         title = f['name']
  56
  57         query = {
  58             'access_token': access_token,
  59             'shared_link': shared_link
  60         }
  61
  62         formats = []
  63
  64         # for entry in (try_get(f, lambda x: x['representations']['entries'], list) or []):
  65         #     entry_url_template = try_get(
  66         #         entry, lambda x: x['content']['url_template'])
  67         #     if not entry_url_template:
  68         #         continue
  69         #     representation = entry.get('representation')
  70         #     if representation == 'dash':
  71         #         TODO: append query to every fragment URL
  72         #         formats.extend(self._extract_mpd_formats(
  73         #             entry_url_template.replace('{+asset_path}', 'manifest.mpd'),
  74         #             file_id, query=query))
  75
  76         authenticated_download_url = f.get('authenticated_download_url')
  77         if authenticated_download_url and f.get('is_download_available'):
  78             formats.append({
  79                 'ext': f.get('extension') or determine_ext(title),
  80                 'filesize': f.get('size'),
  81                 'format_id': 'download',
  82                 'url': update_url_query(authenticated_download_url, query),
  83             })
  84
  85         self._sort_formats(formats)
  86
  87         creator = f.get('created_by') or {}
  88
  89         return {
  90             'id': file_id,
  91             'title': title,
  92             'formats': formats,
  93             'description': f.get('description') or None,
  94             'uploader': creator.get('name'),
  95             'timestamp': parse_iso8601(f.get('created_at')),
  96             'uploader_id': creator.get('id'),
  97         }