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