Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/apa.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     determine_ext,
  10     js_to_json,
  11 )
  12
  13
  14 class APAIE(InfoExtractor):
  15     _VALID_URL = r'https?://[^/]+\.apa\.at/embed/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
  16     _TESTS = [{
  17         'url': 'http://uvp.apa.at/embed/293f6d17-692a-44e3-9fd5-7b178f3a1029',
  18         'md5': '2b12292faeb0a7d930c778c7a5b4759b',
  19         'info_dict': {
  20             'id': 'jjv85FdZ',
  21             'ext': 'mp4',
  22             'title': '"Blau ist mysteriös": Die Blue Man Group im Interview',
  23             'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
  24             'thumbnail': r're:^https?://.*\.jpg$',
  25             'duration': 254,
  26             'timestamp': 1519211149,
  27             'upload_date': '20180221',
  28         },
  29     }, {
  30         'url': 'https://uvp-apapublisher.sf.apa.at/embed/2f94e9e6-d945-4db2-9548-f9a41ebf7b78',
  31         'only_matching': True,
  32     }, {
  33         'url': 'http://uvp-rma.sf.apa.at/embed/70404cca-2f47-4855-bbb8-20b1fae58f76',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://uvp-kleinezeitung.sf.apa.at/embed/f1c44979-dba2-4ebf-b021-e4cf2cac3c81',
  37         'only_matching': True,
  38     }]
  39
  40     @staticmethod
  41     def _extract_urls(webpage):
  42         return [
  43             mobj.group('url')
  44             for mobj in re.finditer(
  45                 r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//[^/]+\.apa\.at/embed/[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}.*?)\1',
  46                 webpage)]
  47
  48     def _real_extract(self, url):
  49         video_id = self._match_id(url)
  50
  51         webpage = self._download_webpage(url, video_id)
  52
  53         jwplatform_id = self._search_regex(
  54             r'media[iI]d\s*:\s*["\'](?P<id>[a-zA-Z0-9]{8})', webpage,
  55             'jwplatform id', default=None)
  56
  57         if jwplatform_id:
  58             return self.url_result(
  59                 'jwplatform:' + jwplatform_id, ie='JWPlatform',
  60                 video_id=video_id)
  61
  62         sources = self._parse_json(
  63             self._search_regex(
  64                 r'sources\s*=\s*(\[.+?\])\s*;', webpage, 'sources'),
  65             video_id, transform_source=js_to_json)
  66
  67         formats = []
  68         for source in sources:
  69             if not isinstance(source, dict):
  70                 continue
  71             source_url = source.get('file')
  72             if not source_url or not isinstance(source_url, compat_str):
  73                 continue
  74             ext = determine_ext(source_url)
  75             if ext == 'm3u8':
  76                 formats.extend(self._extract_m3u8_formats(
  77                     source_url, video_id, 'mp4', entry_protocol='m3u8_native',
  78                     m3u8_id='hls', fatal=False))
  79             else:
  80                 formats.append({
  81                     'url': source_url,
  82                 })
  83         self._sort_formats(formats)
  84
  85         thumbnail = self._search_regex(
  86             r'image\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
  87             'thumbnail', fatal=False, group='url')
  88
  89         return {
  90             'id': video_id,
  91             'title': video_id,
  92             'thumbnail': thumbnail,
  93             'formats': formats,
  94         }