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/(?P<id>[a-z0-9]+)'
  19     _TESTS = [
  20         {
  21             'url': 'http://player.piksel.com/v/nv60p12f',
  22             'md5': 'd9c17bbe9c3386344f9cfd32fad8d235',
  23             'info_dict': {
  24                 'id': 'nv60p12f',
  25                 'ext': 'mp4',
  26                 'title': 'فن الحياة  - الحلقة 1',
  27                 'description': 'احدث برامج الداعية الاسلامي " مصطفي حسني " فى رمضان 2016علي النهار نور',
  28                 'timestamp': 1465231790,
  29                 'upload_date': '20160606',
  30             }
  31         },
  32         {
  33             # Original source: http://www.uscourts.gov/cameras-courts/state-washington-vs-donald-j-trump-et-al
  34             'url': 'https://player.piksel.com/v/v80kqp41',
  35             'md5': '753ddcd8cc8e4fa2dda4b7be0e77744d',
  36             'info_dict': {
  37                 'id': 'v80kqp41',
  38                 'ext': 'mp4',
  39                 'title': 'WAW- State of Washington vs. Donald J. Trump, et al',
  40                 '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.',
  41                 'timestamp': 1486171129,
  42                 'upload_date': '20170204',
  43             }
  44         }
  45     ]
  46
  47     @staticmethod
  48     def _extract_url(webpage):
  49         mobj = re.search(
  50             r'<iframe[^>]+src=["\'](?P<url>(?:https?:)?//player\.piksel\.com/v/[a-z0-9]+)',
  51             webpage)
  52         if mobj:
  53             return mobj.group('url')
  54
  55     def _real_extract(self, url):
  56         video_id = self._match_id(url)
  57         webpage = self._download_webpage(url, video_id)
  58         app_token = self._search_regex([
  59             r'clientAPI\s*:\s*"([^"]+)"',
  60             r'data-de-api-key\s*=\s*"([^"]+)"'
  61         ], webpage, 'app token')
  62         response = self._download_json(
  63             'http://player.piksel.com/ws/ws_program/api/%s/mode/json/apiv/5' % app_token,
  64             video_id, query={
  65                 'v': video_id
  66             })['response']
  67         failure = response.get('failure')
  68         if failure:
  69             raise ExtractorError(response['failure']['reason'], expected=True)
  70         video_data = response['WsProgramResponse']['program']['asset']
  71         title = video_data['title']
  72
  73         formats = []
  74
  75         m3u8_url = dict_get(video_data, [
  76             'm3u8iPadURL',
  77             'ipadM3u8Url',
  78             'm3u8AndroidURL',
  79             'm3u8iPhoneURL',
  80             'iphoneM3u8Url'])
  81         if m3u8_url:
  82             formats.extend(self._extract_m3u8_formats(
  83                 m3u8_url, video_id, 'mp4', 'm3u8_native',
  84                 m3u8_id='hls', fatal=False))
  85
  86         asset_type = dict_get(video_data, ['assetType', 'asset_type'])
  87         for asset_file in video_data.get('assetFiles', []):
  88             # TODO: extract rtmp formats
  89             http_url = asset_file.get('http_url')
  90             if not http_url:
  91                 continue
  92             tbr = None
  93             vbr = int_or_none(asset_file.get('videoBitrate'), 1024)
  94             abr = int_or_none(asset_file.get('audioBitrate'), 1024)
  95             if asset_type == 'video':
  96                 tbr = vbr + abr
  97             elif asset_type == 'audio':
  98                 tbr = abr
  99
 100             format_id = ['http']
 101             if tbr:
 102                 format_id.append(compat_str(tbr))
 103
 104             formats.append({
 105                 'format_id': '-'.join(format_id),
 106                 'url': unescapeHTML(http_url),
 107                 'vbr': vbr,
 108                 'abr': abr,
 109                 'width': int_or_none(asset_file.get('videoWidth')),
 110                 'height': int_or_none(asset_file.get('videoHeight')),
 111                 'filesize': int_or_none(asset_file.get('filesize')),
 112                 'tbr': tbr,
 113             })
 114         self._sort_formats(formats)
 115
 116         return {
 117             'id': video_id,
 118             'title': title,
 119             'description': video_data.get('description'),
 120             'thumbnail': video_data.get('thumbnailUrl'),
 121             'timestamp': parse_iso8601(video_data.get('dateadd')),
 122             'formats': formats,
 123         }