youtube_dl/extractor/box.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     determine_ext,
  10     parse_iso8601,
  11     # try_get,
  12     update_url_query,
  13 )
  14
  15
  16 class BoxIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:[^.]+\.)?app\.box\.com/s/(?P<shared_name>[^/]+)/file/(?P<id>\d+)'
  18     _TEST = {
  19         'url': 'https://mlssoccer.app.box.com/s/0evd2o3e08l60lr4ygukepvnkord1o1x/file/510727257538',
  20         'md5': '1f81b2fd3960f38a40a3b8823e5fcd43',
  21         'info_dict': {
  22             'id': '510727257538',
  23             'ext': 'mp4',
  24             'title': 'Garber   St. Louis will be 28th MLS team  +scarving.mp4',
  25             'uploader': 'MLS Video',
  26             'timestamp': 1566320259,
  27             'upload_date': '20190820',
  28             'uploader_id': '235196876',
  29         }
  30     }
  31
  32     def _real_extract(self, url):
  33         shared_name, file_id = re.match(self._VALID_URL, url).groups()
  34         webpage = self._download_webpage(url, file_id)
  35         request_token = self._parse_json(self._search_regex(
  36             r'Box\.config\s*=\s*({.+?});', webpage,
  37             'Box config'), file_id)['requestToken']
  38         access_token = self._download_json(
  39             'https://app.box.com/app-api/enduserapp/elements/tokens', file_id,
  40             'Downloading token JSON metadata',
  41             data=json.dumps({'fileIDs': [file_id]}).encode(), headers={
  42                 'Content-Type': 'application/json',
  43                 'X-Request-Token': request_token,
  44                 'X-Box-EndUser-API': 'sharedName=' + shared_name,
  45             })[file_id]['read']
  46         shared_link = 'https://app.box.com/s/' + shared_name
  47         f = self._download_json(
  48             'https://api.box.com/2.0/files/' + file_id, file_id,
  49             'Downloading file JSON metadata', headers={
  50                 'Authorization': 'Bearer ' + access_token,
  51                 'BoxApi': 'shared_link=' + shared_link,
  52                 'X-Rep-Hints': '[dash]',  # TODO: extract `hls` formats
  53             }, query={
  54                 'fields': 'authenticated_download_url,created_at,created_by,description,extension,is_download_available,name,representations,size'
  55             })
  56         title = f['name']
  57
  58         query = {
  59             'access_token': access_token,
  60             'shared_link': shared_link
  61         }
  62
  63         formats = []
  64
  65         # for entry in (try_get(f, lambda x: x['representations']['entries'], list) or []):
  66         #     entry_url_template = try_get(
  67         #         entry, lambda x: x['content']['url_template'])
  68         #     if not entry_url_template:
  69         #         continue
  70         #     representation = entry.get('representation')
  71         #     if representation == 'dash':
  72         #         TODO: append query to every fragment URL
  73         #         formats.extend(self._extract_mpd_formats(
  74         #             entry_url_template.replace('{+asset_path}', 'manifest.mpd'),
  75         #             file_id, query=query))
  76
  77         authenticated_download_url = f.get('authenticated_download_url')
  78         if authenticated_download_url and f.get('is_download_available'):
  79             formats.append({
  80                 'ext': f.get('extension') or determine_ext(title),
  81                 'filesize': f.get('size'),
  82                 'format_id': 'download',
  83                 'url': update_url_query(authenticated_download_url, query),
  84             })
  85
  86         self._sort_formats(formats)
  87
  88         creator = f.get('created_by') or {}
  89
  90         return {
  91             'id': file_id,
  92             'title': title,
  93             'formats': formats,
  94             'description': f.get('description') or None,
  95             'uploader': creator.get('name'),
  96             'timestamp': parse_iso8601(f.get('created_at')),
  97             'uploader_id': creator.get('id'),
  98         }