_ Git - youtube-dl/blob - youtube_dl/extractor/gaskrank.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     float_or_none,
   8     int_or_none,
   9     js_to_json,
  10     unified_strdate,
  11 )
  12
  13
  14 class GaskrankIE(InfoExtractor):
  15     """InfoExtractor for gaskrank.tv"""
  16     _VALID_URL = r'https?://(?:www\.)?gaskrank\.tv/tv/(?P<categories>[^/]+)/(?P<id>[^/]+)\.html?'
  17     _TESTS = [
  18         {
  19             'url': 'http://www.gaskrank.tv/tv/motorrad-fun/strike-einparken-durch-anfaenger-crash-mit-groesserem-flurschaden.htm',
  20             'md5': '1ae88dbac97887d85ebd1157a95fc4f9',
  21             'info_dict': {
  22                 'id': '201601/26955',
  23                 'ext': 'mp4',
  24                 'title': 'Strike! Einparken können nur Männer - Flurschaden hält sich in Grenzen *lol*',
  25                 'thumbnail': r're:^https?://.*\.jpg$',
  26                 'categories': ['motorrad-fun'],
  27                 'display_id': 'strike-einparken-durch-anfaenger-crash-mit-groesserem-flurschaden',
  28                 'uploader_id': 'Bikefun',
  29                 'upload_date': '20170110',
  30                 'uploader_url': None,
  31             }
  32         },
  33         {
  34             'url': 'http://www.gaskrank.tv/tv/racing/isle-of-man-tt-2011-michael-du-15920.htm',
  35             'md5': 'c33ee32c711bc6c8224bfcbe62b23095',
  36             'info_dict': {
  37                 'id': '201106/15920',
  38                 'ext': 'mp4',
  39                 'title': 'Isle of Man - Michael Dunlop vs Guy Martin - schwindelig kucken',
  40                 'thumbnail': r're:^https?://.*\.jpg$',
  41                 'categories': ['racing'],
  42                 'display_id': 'isle-of-man-tt-2011-michael-du-15920',
  43                 'uploader_id': 'IOM',
  44                 'upload_date': '20160506',
  45                 'uploader_url': 'www.iomtt.com',
  46             }
  47         }
  48     ]
  49
  50     def _real_extract(self, url):
  51         """extract information from gaskrank.tv"""
  52         def fix_json(code):
  53             """Removes trailing comma in json: {{},} --> {{}}"""
  54             return re.sub(r',\s*}', r'}', js_to_json(code))
  55
  56         display_id = self._match_id(url)
  57         webpage = self._download_webpage(url, display_id)
  58         categories = [re.match(self._VALID_URL, url).group('categories')]
  59         title = self._search_regex(
  60             r'movieName\s*:\s*\'([^\']*)\'',
  61             webpage, 'title')
  62         thumbnail = self._search_regex(
  63             r'poster\s*:\s*\'([^\']*)\'',
  64             webpage, 'thumbnail', default=None)
  65
  66         mobj = re.search(
  67             r'Video von:\s*(?P<uploader_id>[^|]*?)\s*\|\s*vom:\s*(?P<upload_date>[0-9][0-9]\.[0-9][0-9]\.[0-9][0-9][0-9][0-9])',
  68             webpage)
  69         if mobj is not None:
  70             uploader_id = mobj.groupdict().get('uploader_id')
  71             upload_date = unified_strdate(mobj.groupdict().get('upload_date'))
  72
  73         uploader_url = self._search_regex(
  74             r'Homepage:\s*<[^>]*>(?P<uploader_url>[^<]*)',
  75             webpage, 'uploader_url', default=None)
  76         tags = re.findall(
  77             r'/tv/tags/[^/]+/"\s*>(?P<tag>[^<]*?)<',
  78             webpage)
  79
  80         view_count = self._search_regex(
  81             r'class\s*=\s*"gkRight"(?:[^>]*>\s*<[^>]*)*icon-eye-open(?:[^>]*>\s*<[^>]*)*>\s*(?P<view_count>[0-9\.]*)',
  82             webpage, 'view_count', default=None)
  83         if view_count:
  84             view_count = int_or_none(view_count.replace('.', ''))
  85
  86         average_rating = self._search_regex(
  87             r'itemprop\s*=\s*"ratingValue"[^>]*>\s*(?P<average_rating>[0-9,]+)',
  88             webpage, 'average_rating')
  89         if average_rating:
  90             average_rating = float_or_none(average_rating.replace(',', '.'))
  91
  92         playlist = self._parse_json(
  93             self._search_regex(
  94                 r'playlist\s*:\s*\[([^\]]*)\]',
  95                 webpage, 'playlist', default='{}'),
  96             display_id, transform_source=fix_json, fatal=False)
  97
  98         video_id = self._search_regex(
  99             r'https?://movies\.gaskrank\.tv/([^-]*?)(-[^\.]*)?\.mp4',
 100             playlist.get('0').get('src'), 'video id')
 101
 102         formats = []
 103         for key in playlist:
 104             formats.append({
 105                 'url': playlist[key]['src'],
 106                 'format_id': key,
 107                 'quality': playlist[key].get('quality')})
 108         self._sort_formats(formats, field_preference=['format_id'])
 109
 110         return {
 111             'id': video_id,
 112             'title': title,
 113             'formats': formats,
 114             'thumbnail': thumbnail,
 115             'categories': categories,
 116             'display_id': display_id,
 117             'uploader_id': uploader_id,
 118             'upload_date': upload_date,
 119             'uploader_url': uploader_url,
 120             'tags': tags,
 121             'view_count': view_count,
 122             'average_rating': average_rating,
 123         }