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