Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/clipfish.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     determine_ext,
   8     int_or_none,
   9     js_to_json,
  10     parse_iso8601,
  11     remove_end,
  12 )
  13
  14
  15 class ClipfishIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?clipfish\.de/(?:[^/]+/)+video/(?P<id>[0-9]+)'
  17     _TEST = {
  18         'url': 'http://www.clipfish.de/special/game-trailer/video/3966754/fifa-14-e3-2013-trailer/',
  19         'md5': '79bc922f3e8a9097b3d68a93780fd475',
  20         'info_dict': {
  21             'id': '3966754',
  22             'ext': 'mp4',
  23             'title': 'FIFA 14 - E3 2013 Trailer',
  24             'timestamp': 1370938118,
  25             'upload_date': '20130611',
  26             'duration': 82,
  27         }
  28     }
  29
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32
  33         webpage = self._download_webpage(url, video_id)
  34
  35         video_info = self._parse_json(
  36             js_to_json(self._html_search_regex(
  37                 '(?s)videoObject\s*=\s*({.+?});', webpage, 'video object')),
  38             video_id)
  39
  40         formats = []
  41         for video_url in re.findall(r'var\s+videourl\s*=\s*"([^"]+)"', webpage):
  42             ext = determine_ext(video_url)
  43             if ext == 'm3u8':
  44                 formats.append({
  45                     'url': video_url.replace('de.hls.fra.clipfish.de', 'hls.fra.clipfish.de'),
  46                     'ext': 'mp4',
  47                     'format_id': 'hls',
  48                 })
  49             else:
  50                 formats.append({
  51                     'url': video_url,
  52                     'format_id': ext,
  53                 })
  54         self._sort_formats(formats)
  55
  56         title = remove_end(self._og_search_title(webpage), ' - Video')
  57         thumbnail = self._og_search_thumbnail(webpage)
  58         duration = int_or_none(video_info.get('length'))
  59         timestamp = parse_iso8601(self._html_search_meta('uploadDate', webpage, 'upload date'))
  60
  61         return {
  62             'id': video_id,
  63             'title': title,
  64             'formats': formats,
  65             'thumbnail': thumbnail,
  66             'duration': duration,
  67             'timestamp': timestamp,
  68         }