Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/joj.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     int_or_none,
  10     js_to_json,
  11     try_get,
  12 )
  13
  14
  15 class JojIE(InfoExtractor):
  16     _VALID_URL = r'''(?x)
  17                     (?:
  18                         joj:|
  19                         https?://media\.joj\.sk/embed/
  20                     )
  21                     (?P<id>[^/?#^]+)
  22                 '''
  23     _TESTS = [{
  24         'url': 'https://media.joj.sk/embed/a388ec4c-6019-4a4a-9312-b1bee194e932',
  25         'info_dict': {
  26             'id': 'a388ec4c-6019-4a4a-9312-b1bee194e932',
  27             'ext': 'mp4',
  28             'title': 'NOVÉ BÝVANIE',
  29             'thumbnail': r're:^https?://.*\.jpg$',
  30             'duration': 3118,
  31         }
  32     }, {
  33         'url': 'https://media.joj.sk/embed/9i1cxv',
  34         'only_matching': True,
  35     }, {
  36         'url': 'joj:a388ec4c-6019-4a4a-9312-b1bee194e932',
  37         'only_matching': True,
  38     }, {
  39         'url': 'joj:9i1cxv',
  40         'only_matching': True,
  41     }]
  42
  43     @staticmethod
  44     def _extract_urls(webpage):
  45         return [
  46             mobj.group('url')
  47             for mobj in re.finditer(
  48                 r'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//media\.joj\.sk/embed/(?:(?!\1).)+)\1',
  49                 webpage)]
  50
  51     def _real_extract(self, url):
  52         video_id = self._match_id(url)
  53
  54         webpage = self._download_webpage(
  55             'https://media.joj.sk/embed/%s' % video_id, video_id)
  56
  57         title = self._search_regex(
  58             (r'videoTitle\s*:\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
  59              r'<title>(?P<title>[^<]+)'), webpage, 'title',
  60             default=None, group='title') or self._og_search_title(webpage)
  61
  62         bitrates = self._parse_json(
  63             self._search_regex(
  64                 r'(?s)(?:src|bitrates)\s*=\s*({.+?});', webpage, 'bitrates',
  65                 default='{}'),
  66             video_id, transform_source=js_to_json, fatal=False)
  67
  68         formats = []
  69         for format_url in try_get(bitrates, lambda x: x['mp4'], list) or []:
  70             if isinstance(format_url, compat_str):
  71                 height = self._search_regex(
  72                     r'(\d+)[pP]\.', format_url, 'height', default=None)
  73                 formats.append({
  74                     'url': format_url,
  75                     'format_id': '%sp' % height if height else None,
  76                     'height': int(height),
  77                 })
  78         if not formats:
  79             playlist = self._download_xml(
  80                 'https://media.joj.sk/services/Video.php?clip=%s' % video_id,
  81                 video_id)
  82             for file_el in playlist.findall('./files/file'):
  83                 path = file_el.get('path')
  84                 if not path:
  85                     continue
  86                 format_id = file_el.get('id') or file_el.get('label')
  87                 formats.append({
  88                     'url': 'http://n16.joj.sk/storage/%s' % path.replace(
  89                         'dat/', '', 1),
  90                     'format_id': format_id,
  91                     'height': int_or_none(self._search_regex(
  92                         r'(\d+)[pP]', format_id or path, 'height',
  93                         default=None)),
  94                 })
  95         self._sort_formats(formats)
  96
  97         thumbnail = self._og_search_thumbnail(webpage)
  98
  99         duration = int_or_none(self._search_regex(
 100             r'videoDuration\s*:\s*(\d+)', webpage, 'duration', fatal=False))
 101
 102         return {
 103             'id': video_id,
 104             'title': title,
 105             'thumbnail': thumbnail,
 106             'duration': duration,
 107             'formats': formats,
 108         }