Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/piksel.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     dict_get,
  11     int_or_none,
  12     unescapeHTML,
  13     parse_iso8601,
  14 )
  15
  16
  17 class PikselIE(InfoExtractor):
  18     _VALID_URL = r'https?://player\.piksel\.com/v/(?:refid/[^/]+/prefid/)?(?P<id>[a-z0-9_]+)'
  19     _TESTS = [
  20         {
  21             'url': 'http://player.piksel.com/v/ums2867l',
  22             'md5': '34e34c8d89dc2559976a6079db531e85',
  23             'info_dict': {
  24                 'id': 'ums2867l',
  25                 'ext': 'mp4',
  26                 'title': 'GX-005 with Caption',
  27                 'timestamp': 1481335659,
  28                 'upload_date': '20161210'
  29             }
  30         },
  31         {
  32             # Original source: http://www.uscourts.gov/cameras-courts/state-washington-vs-donald-j-trump-et-al
  33             'url': 'https://player.piksel.com/v/v80kqp41',
  34             'md5': '753ddcd8cc8e4fa2dda4b7be0e77744d',
  35             'info_dict': {
  36                 'id': 'v80kqp41',
  37                 'ext': 'mp4',
  38                 'title': 'WAW- State of Washington vs. Donald J. Trump, et al',
  39                 'description': 'State of Washington vs. Donald J. Trump, et al, Case Number 17-CV-00141-JLR, TRO Hearing, Civil Rights Case, 02/3/2017, 1:00 PM (PST), Seattle Federal Courthouse, Seattle, WA, Judge James L. Robart presiding.',
  40                 'timestamp': 1486171129,
  41                 'upload_date': '20170204'
  42             }
  43         },
  44         {
  45             # https://www3.nhk.or.jp/nhkworld/en/ondemand/video/2019240/
  46             'url': 'http://player.piksel.com/v/refid/nhkworld/prefid/nw_vod_v_en_2019_240_20190823233000_02_1566873477',
  47             'only_matching': True,
  48         }
  49     ]
  50
  51     @staticmethod
  52     def _extract_url(webpage):
  53         mobj = re.search(
  54             r'<iframe[^>]+src=["\'](?P<url>(?:https?:)?//player\.piksel\.com/v/[a-z0-9]+)',
  55             webpage)
  56         if mobj:
  57             return mobj.group('url')
  58
  59     def _real_extract(self, url):
  60         display_id = self._match_id(url)
  61         webpage = self._download_webpage(url, display_id)
  62         video_id = self._search_regex(
  63             r'data-de-program-uuid=[\'"]([a-z0-9]+)',
  64             webpage, 'program uuid', default=display_id)
  65         app_token = self._search_regex([
  66             r'clientAPI\s*:\s*"([^"]+)"',
  67             r'data-de-api-key\s*=\s*"([^"]+)"'
  68         ], webpage, 'app token')
  69         response = self._download_json(
  70             'http://player.piksel.com/ws/ws_program/api/%s/mode/json/apiv/5' % app_token,
  71             video_id, query={
  72                 'v': video_id
  73             })['response']
  74         failure = response.get('failure')
  75         if failure:
  76             raise ExtractorError(response['failure']['reason'], expected=True)
  77         video_data = response['WsProgramResponse']['program']['asset']
  78         title = video_data['title']
  79
  80         formats = []
  81
  82         m3u8_url = dict_get(video_data, [
  83             'm3u8iPadURL',
  84             'ipadM3u8Url',
  85             'm3u8AndroidURL',
  86             'm3u8iPhoneURL',
  87             'iphoneM3u8Url'])
  88         if m3u8_url:
  89             formats.extend(self._extract_m3u8_formats(
  90                 m3u8_url, video_id, 'mp4', 'm3u8_native',
  91                 m3u8_id='hls', fatal=False))
  92
  93         asset_type = dict_get(video_data, ['assetType', 'asset_type'])
  94         for asset_file in video_data.get('assetFiles', []):
  95             # TODO: extract rtmp formats
  96             http_url = asset_file.get('http_url')
  97             if not http_url:
  98                 continue
  99             tbr = None
 100             vbr = int_or_none(asset_file.get('videoBitrate'), 1024)
 101             abr = int_or_none(asset_file.get('audioBitrate'), 1024)
 102             if asset_type == 'video':
 103                 tbr = vbr + abr
 104             elif asset_type == 'audio':
 105                 tbr = abr
 106
 107             format_id = ['http']
 108             if tbr:
 109                 format_id.append(compat_str(tbr))
 110
 111             formats.append({
 112                 'format_id': '-'.join(format_id),
 113                 'url': unescapeHTML(http_url),
 114                 'vbr': vbr,
 115                 'abr': abr,
 116                 'width': int_or_none(asset_file.get('videoWidth')),
 117                 'height': int_or_none(asset_file.get('videoHeight')),
 118                 'filesize': int_or_none(asset_file.get('filesize')),
 119                 'tbr': tbr,
 120             })
 121         self._sort_formats(formats)
 122
 123         subtitles = {}
 124         for caption in video_data.get('captions', []):
 125             caption_url = caption.get('url')
 126             if caption_url:
 127                 subtitles.setdefault(caption.get('locale', 'en'), []).append({
 128                     'url': caption_url})
 129
 130         return {
 131             'id': video_id,
 132             'title': title,
 133             'description': video_data.get('description'),
 134             'thumbnail': video_data.get('thumbnailUrl'),
 135             'timestamp': parse_iso8601(video_data.get('dateadd')),
 136             'formats': formats,
 137             'subtitles': subtitles,
 138         }