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/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
  46     @staticmethod
  47     def _extract_url(webpage):
  48         mobj = re.search(
  49             r'<iframe[^>]+src=["\'](?P<url>(?:https?:)?//player\.piksel\.com/v/[a-z0-9]+)',
  50             webpage)
  51         if mobj:
  52             return mobj.group('url')
  53
  54     def _real_extract(self, url):
  55         video_id = self._match_id(url)
  56         webpage = self._download_webpage(url, video_id)
  57         app_token = self._search_regex([
  58             r'clientAPI\s*:\s*"([^"]+)"',
  59             r'data-de-api-key\s*=\s*"([^"]+)"'
  60         ], webpage, 'app token')
  61         response = self._download_json(
  62             'http://player.piksel.com/ws/ws_program/api/%s/mode/json/apiv/5' % app_token,
  63             video_id, query={
  64                 'v': video_id
  65             })['response']
  66         failure = response.get('failure')
  67         if failure:
  68             raise ExtractorError(response['failure']['reason'], expected=True)
  69         video_data = response['WsProgramResponse']['program']['asset']
  70         title = video_data['title']
  71
  72         formats = []
  73
  74         m3u8_url = dict_get(video_data, [
  75             'm3u8iPadURL',
  76             'ipadM3u8Url',
  77             'm3u8AndroidURL',
  78             'm3u8iPhoneURL',
  79             'iphoneM3u8Url'])
  80         if m3u8_url:
  81             formats.extend(self._extract_m3u8_formats(
  82                 m3u8_url, video_id, 'mp4', 'm3u8_native',
  83                 m3u8_id='hls', fatal=False))
  84
  85         asset_type = dict_get(video_data, ['assetType', 'asset_type'])
  86         for asset_file in video_data.get('assetFiles', []):
  87             # TODO: extract rtmp formats
  88             http_url = asset_file.get('http_url')
  89             if not http_url:
  90                 continue
  91             tbr = None
  92             vbr = int_or_none(asset_file.get('videoBitrate'), 1024)
  93             abr = int_or_none(asset_file.get('audioBitrate'), 1024)
  94             if asset_type == 'video':
  95                 tbr = vbr + abr
  96             elif asset_type == 'audio':
  97                 tbr = abr
  98
  99             format_id = ['http']
 100             if tbr:
 101                 format_id.append(compat_str(tbr))
 102
 103             formats.append({
 104                 'format_id': '-'.join(format_id),
 105                 'url': unescapeHTML(http_url),
 106                 'vbr': vbr,
 107                 'abr': abr,
 108                 'width': int_or_none(asset_file.get('videoWidth')),
 109                 'height': int_or_none(asset_file.get('videoHeight')),
 110                 'filesize': int_or_none(asset_file.get('filesize')),
 111                 'tbr': tbr,
 112             })
 113         self._sort_formats(formats)
 114
 115         subtitles = {}
 116         for caption in video_data.get('captions', []):
 117             caption_url = caption.get('url')
 118             if caption_url:
 119                 subtitles.setdefault(caption.get('locale', 'en'), []).append({
 120                     'url': caption_url})
 121
 122         return {
 123             'id': video_id,
 124             'title': title,
 125             'description': video_data.get('description'),
 126             'thumbnail': video_data.get('thumbnailUrl'),
 127             'timestamp': parse_iso8601(video_data.get('dateadd')),
 128             'formats': formats,
 129             'subtitles': subtitles,
 130         }