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

   1 # -*- coding: utf-8 -*-
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import parse_filesize, ExtractorError
   8
   9
  10 class TagesschauIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:[^/]+/)*?[^/#?]+?(?P<id>-?[0-9]+)(?:~_[^/#?]+?)?\.html'
  12
  13     _TESTS = [{
  14         'url': 'http://www.tagesschau.de/multimedia/video/video-102143.html',
  15         'md5': '917a228bc7df7850783bc47979673a09',
  16         'info_dict': {
  17             'id': '102143',
  18             'ext': 'mp4',
  19             'title': 'Regierungsumbildung in Athen: Neue Minister in Griechenland vereidigt',
  20             'description': 'md5:171feccd9d9b3dd54d05d501568f6359',
  21             'thumbnail': 're:^http:.*\.jpg$',
  22         },
  23     }, {
  24         'url': 'http://www.tagesschau.de/multimedia/sendung/ts-5727.html',
  25         'md5': '3c54c1f6243d279b706bde660ceec633',
  26         'info_dict': {
  27             'id': '5727',
  28             'ext': 'mp4',
  29             'description': 'md5:695c01bfd98b7e313c501386327aea59',
  30             'title': 'Sendung: tagesschau \t04.12.2014 20:00 Uhr',
  31             'thumbnail': 're:^http:.*\.jpg$',
  32         },
  33     }, {
  34         'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html',
  35         'only_matching': True,
  36     }, {
  37         'url': 'http://www.tagesschau.de/multimedia/sendung/tt-3827.html',
  38         'only_matching': True,
  39     }, {
  40         'url': 'http://www.tagesschau.de/multimedia/sendung/nm-3475.html',
  41         'only_matching': True,
  42     }, {
  43         'url': 'http://www.tagesschau.de/multimedia/sendung/weltspiegel-3167.html',
  44         'only_matching': True,
  45     }, {
  46         'url': 'http://www.tagesschau.de/multimedia/tsvorzwanzig-959.html',
  47         'only_matching': True,
  48     }, {
  49         'url': 'http://www.tagesschau.de/multimedia/sendung/bab/bab-3299~_bab-sendung-209.html',
  50         'only_matching': True,
  51     }, {
  52         'url': 'http://www.tagesschau.de/multimedia/video/video-102303~_bab-sendung-211.html',
  53         'only_matching': True,
  54     }, {
  55         'url': 'http://www.tagesschau.de/multimedia/politikimradio/audio-18407.html',
  56         'only_matching': True,
  57     }]
  58
  59     _FORMATS = {
  60         's': {'width': 256, 'height': 144, 'quality': 1},
  61         'm': {'width': 512, 'height': 288, 'quality': 2},
  62         'l': {'width': 960, 'height': 544, 'quality': 3},
  63     }
  64
  65     def _real_extract(self, url):
  66         video_id = self._match_id(url)
  67         display_id = video_id.lstrip('-')
  68         webpage = self._download_webpage(url, display_id)
  69
  70         player_url = self._html_search_meta(
  71             'twitter:player', webpage, 'player URL', default=None)
  72         if player_url:
  73             playerpage = self._download_webpage(
  74                 player_url, display_id, 'Downloading player page')
  75
  76             medias = re.findall(
  77                 r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"',
  78                 playerpage)
  79             formats = []
  80             for url, ext, res in medias:
  81                 f = {
  82                     'format_id': res + '_' + ext,
  83                     'url': url,
  84                     'ext': ext,
  85                 }
  86                 f.update(self._FORMATS.get(res, {}))
  87                 formats.append(f)
  88             thumbnail_fn = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1]
  89             title = self._og_search_title(webpage).strip()
  90             description = self._og_search_description(webpage).strip()
  91         else:
  92             download_text = self._search_regex(
  93                 r'(?s)<p>Wir bieten dieses Video in folgenden Formaten zum Download an:</p>\s*<div class="controls">(.*?)</div>\s*<p>',
  94                 webpage, 'download links')
  95             links = re.finditer(
  96                 r'<div class="button" title="(?P<title>[^"]*)"><a href="(?P<url>[^"]+)">(?P<name>.+?)</a></div>',
  97                 download_text)
  98             formats = []
  99             for l in links:
 100                 format_id = self._search_regex(
 101                     r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID')
 102                 format = {
 103                     'format_id': format_id,
 104                     'url': l.group('url'),
 105                     'format_name': l.group('name'),
 106                 }
 107                 m = re.match(
 108                     r'''(?x)
 109                         Video:\s*(?P<vcodec>[a-zA-Z0-9/._-]+)\s*&\#10;
 110                         (?P<width>[0-9]+)x(?P<height>[0-9]+)px&\#10;
 111                         (?P<vbr>[0-9]+)kbps&\#10;
 112                         Audio:\s*(?P<abr>[0-9]+)kbps,\s*(?P<audio_desc>[A-Za-z\.0-9]+)&\#10;
 113                         Gr&ouml;&szlig;e:\s*(?P<filesize_approx>[0-9.,]+\s+[a-zA-Z]*B)''',
 114                     l.group('title'))
 115                 if m:
 116                     format.update({
 117                         'format_note': m.group('audio_desc'),
 118                         'vcodec': m.group('vcodec'),
 119                         'width': int(m.group('width')),
 120                         'height': int(m.group('height')),
 121                         'abr': int(m.group('abr')),
 122                         'vbr': int(m.group('vbr')),
 123                         'filesize_approx': parse_filesize(m.group('filesize_approx')),
 124                     })
 125                 formats.append(format)
 126             thumbnail_fn = self._search_regex(
 127                 r'(?s)<img alt="Sendungsbild".*?src="([^"]+)"',
 128                 webpage, 'thumbnail', fatal=False)
 129             description = self._html_search_regex(
 130                 r'(?s)<p class="teasertext">(.*?)</p>',
 131                 webpage, 'description', default=None)
 132             title = self._html_search_regex(
 133                 r'<span class="headline".*?>(.*?)</span>', webpage, 'title')
 134
 135         self._sort_formats(formats)
 136         thumbnail = 'http://www.tagesschau.de' + thumbnail_fn
 137
 138         return {
 139             'id': display_id,
 140             'title': title,
 141             'thumbnail': thumbnail,
 142             'formats': formats,
 143             'description': description,
 144         }