yt_dlp/extractor/box.py

   1 import json
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     parse_iso8601,
   8     update_url_query,
   9     url_or_none,
  10 )
  11 from ..utils.traversal import traverse_obj
  12
  13
  14 class BoxIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<service>app|ent)\.box\.com/s/(?P<shared_name>[^/?#]+)(?:/file/(?P<id>\d+))?'
  16     _TESTS = [{
  17         'url': 'https://mlssoccer.app.box.com/s/0evd2o3e08l60lr4ygukepvnkord1o1x/file/510727257538',
  18         'md5': '1f81b2fd3960f38a40a3b8823e5fcd43',
  19         'info_dict': {
  20             'id': '510727257538',
  21             'ext': 'mp4',
  22             'title': 'Garber   St. Louis will be 28th MLS team  +scarving.mp4',
  23             'uploader': '',
  24             'timestamp': 1566320259,
  25             'upload_date': '20190820',
  26             'uploader_id': '235196876',
  27         },
  28         'params': {'skip_download': 'dash fragment too small'},
  29     }, {
  30         'url': 'https://utexas.app.box.com/s/2x6vanv85fdl8j2eqlcxmv0gp1wvps6e',
  31         'info_dict': {
  32             'id': '787379022466',
  33             'ext': 'mp4',
  34             'title': 'Webinar recording: Take the Leap!.mp4',
  35             'uploader': 'Patricia Mosele',
  36             'timestamp': 1615824864,
  37             'upload_date': '20210315',
  38             'uploader_id': '239068974',
  39         },
  40         'params': {'skip_download': 'dash fragment too small'},
  41     }, {
  42         'url': 'https://thejacksonlaboratory.ent.box.com/s/2x09dm6vcg6y28o0oox1so4l0t8wzt6l/file/1536173056065',
  43         'info_dict': {
  44             'id': '1536173056065',
  45             'ext': 'mp4',
  46             'uploader_id': '18523128264',
  47             'uploader': 'Lexi Hennigan',
  48             'title': 'iPSC Symposium recording part 1.mp4',
  49             'timestamp': 1716228343,
  50             'upload_date': '20240520',
  51         },
  52         'params': {'skip_download': 'dash fragment too small'},
  53     }]
  54
  55     def _real_extract(self, url):
  56         shared_name, file_id, service = self._match_valid_url(url).group('shared_name', 'id', 'service')
  57         webpage = self._download_webpage(url, file_id or shared_name)
  58
  59         if not file_id:
  60             post_stream_data = self._search_json(
  61                 r'Box\.postStreamData\s*=', webpage, 'Box post-stream data', shared_name)
  62             shared_item = traverse_obj(
  63                 post_stream_data, ('/app-api/enduserapp/shared-item', {dict})) or {}
  64             if shared_item.get('itemType') != 'file':
  65                 raise ExtractorError('The requested resource is not a file', expected=True)
  66
  67             file_id = str(shared_item['itemID'])
  68
  69         request_token = self._search_json(
  70             r'Box\.config\s*=', webpage, 'Box config', file_id)['requestToken']
  71         access_token = self._download_json(
  72             f'https://{service}.box.com/app-api/enduserapp/elements/tokens', file_id,
  73             'Downloading token JSON metadata',
  74             data=json.dumps({'fileIDs': [file_id]}).encode(), headers={
  75                 'Content-Type': 'application/json',
  76                 'X-Request-Token': request_token,
  77                 'X-Box-EndUser-API': 'sharedName=' + shared_name,
  78             })[file_id]['read']
  79         shared_link = f'https://{service}.box.com/s/{shared_name}'
  80         f = self._download_json(
  81             'https://api.box.com/2.0/files/' + file_id, file_id,
  82             'Downloading file JSON metadata', headers={
  83                 'Authorization': 'Bearer ' + access_token,
  84                 'BoxApi': 'shared_link=' + shared_link,
  85                 'X-Rep-Hints': '[dash]',  # TODO: extract `hls` formats
  86             }, query={
  87                 'fields': 'authenticated_download_url,created_at,created_by,description,extension,is_download_available,name,representations,size',
  88             })
  89         title = f['name']
  90
  91         query = {
  92             'access_token': access_token,
  93             'shared_link': shared_link,
  94         }
  95
  96         formats = []
  97
  98         for url_tmpl in traverse_obj(f, (
  99             'representations', 'entries', lambda _, v: v['representation'] == 'dash',
 100             'content', 'url_template', {url_or_none},
 101         )):
 102             manifest_url = update_url_query(url_tmpl.replace('{+asset_path}', 'manifest.mpd'), query)
 103             fmts = self._extract_mpd_formats(manifest_url, file_id)
 104             for fmt in fmts:
 105                 fmt['extra_param_to_segment_url'] = urllib.parse.urlparse(manifest_url).query
 106             formats.extend(fmts)
 107
 108         creator = f.get('created_by') or {}
 109
 110         return {
 111             'id': file_id,
 112             'title': title,
 113             'formats': formats,
 114             'description': f.get('description') or None,
 115             'uploader': creator.get('name'),
 116             'timestamp': parse_iso8601(f.get('created_at')),
 117             'uploader_id': creator.get('id'),
 118         }