_ 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
   8
   9
  10 class TagesschauIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/[^/]+/(?:[^/]+/)*?[^/#?]+?(?P<id>-?[0-9]+)(?:~_?[^/#?]+?)?\.html'
  12
  13     _TESTS = [{
  14         'url': 'http://www.tagesschau.de/multimedia/video/video-102143.html',
  15         'md5': 'f7c27a0eff3bfe8c7727e65f8fe1b1e6',
  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:^https?:.*\.jpg$',
  22         },
  23     }, {
  24         'url': 'http://www.tagesschau.de/multimedia/sendung/ts-5727.html',
  25         'md5': '3c54c1f6243d279b706bde660ceec633',
  26         'info_dict': {
  27             'id': '5727',
  28             'ext': 'mp4',
  29             'description': 'md5:695c01bfd98b7e313c501386327aea59',
  30             'title': 'Sendung: tagesschau \t04.12.2014 20:00 Uhr',
  31             'thumbnail': 're:^https?:.*\.jpg$',
  32         },
  33     }, {
  34         'url': 'http://www.tagesschau.de/multimedia/politikimradio/audio-18407.html',
  35         'md5': 'aef45de271c4bf0a5db834aa40bf774c',
  36         'info_dict': {
  37             'id': '18407',
  38             'ext': 'mp3',
  39             'title': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich',
  40             'description': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich',
  41             'thumbnail': 're:^https?:.*\.jpg$',
  42         },
  43     }, {
  44         'url': 'http://www.tagesschau.de/inland/afd-parteitag-135.html',
  45         'info_dict': {
  46             'id': '135',
  47             'title': 'Möchtegern-Underdog mit Machtanspruch',
  48         },
  49         'playlist_count': 2,
  50     }, {
  51         'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html',
  52         'only_matching': True,
  53     }, {
  54         'url': 'http://www.tagesschau.de/multimedia/sendung/tt-3827.html',
  55         'only_matching': True,
  56     }, {
  57         'url': 'http://www.tagesschau.de/multimedia/sendung/nm-3475.html',
  58         'only_matching': True,
  59     }, {
  60         'url': 'http://www.tagesschau.de/multimedia/sendung/weltspiegel-3167.html',
  61         'only_matching': True,
  62     }, {
  63         'url': 'http://www.tagesschau.de/multimedia/tsvorzwanzig-959.html',
  64         'only_matching': True,
  65     }, {
  66         'url': 'http://www.tagesschau.de/multimedia/sendung/bab/bab-3299~_bab-sendung-209.html',
  67         'only_matching': True,
  68     }, {
  69         'url': 'http://www.tagesschau.de/multimedia/video/video-102303~_bab-sendung-211.html',
  70         'only_matching': True,
  71     }, {
  72         'url': 'http://www.tagesschau.de/multimedia/video/video-179517~player.html',
  73         'only_matching': True,
  74     }]
  75
  76     _FORMATS = {
  77         'xs': {'quality': 0},
  78         's': {'width': 320, 'height': 180, 'quality': 1},
  79         'm': {'width': 512, 'height': 288, 'quality': 2},
  80         'l': {'width': 960, 'height': 540, 'quality': 3},
  81         'xl': {'width': 1280, 'height': 720, 'quality': 4},
  82         'xxl': {'quality': 5},
  83     }
  84
  85     def _extract_formats(self, download_text):
  86         links = re.finditer(
  87             r'<div class="button" title="(?P<title>[^"]*)"><a href="(?P<url>[^"]+)">(?P<name>.+?)</a></div>',
  88             download_text)
  89         formats = []
  90         for l in links:
  91             format_id = self._search_regex(
  92                 r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID')
  93             format = {
  94                 'format_id': format_id,
  95                 'url': l.group('url'),
  96                 'format_name': l.group('name'),
  97             }
  98             m = re.match(
  99                 r'''(?x)
 100                     Video:\s*(?P<vcodec>[a-zA-Z0-9/._-]+)\s*&\#10;
 101                     (?P<width>[0-9]+)x(?P<height>[0-9]+)px&\#10;
 102                     (?P<vbr>[0-9]+)kbps&\#10;
 103                     Audio:\s*(?P<abr>[0-9]+)kbps,\s*(?P<audio_desc>[A-Za-z\.0-9]+)&\#10;
 104                     Gr&ouml;&szlig;e:\s*(?P<filesize_approx>[0-9.,]+\s+[a-zA-Z]*B)''',
 105                 l.group('title'))
 106             if m:
 107                 format.update({
 108                     'format_note': m.group('audio_desc'),
 109                     'vcodec': m.group('vcodec'),
 110                     'width': int(m.group('width')),
 111                     'height': int(m.group('height')),
 112                     'abr': int(m.group('abr')),
 113                     'vbr': int(m.group('vbr')),
 114                     'filesize_approx': parse_filesize(m.group('filesize_approx')),
 115                 })
 116             formats.append(format)
 117         self._sort_formats(formats)
 118         return formats
 119
 120     def _real_extract(self, url):
 121         video_id = self._match_id(url)
 122         display_id = video_id.lstrip('-')
 123         webpage = self._download_webpage(url, display_id)
 124
 125         player_url = self._html_search_meta(
 126             'twitter:player', webpage, 'player URL', default=None)
 127         if player_url:
 128             playerpage = self._download_webpage(
 129                 player_url, display_id, 'Downloading player page')
 130
 131             formats = []
 132             for media in re.finditer(
 133                     r'''(?x)
 134                         (?P<q_url>["\'])(?P<url>http://media.+?)(?P=q_url)
 135                         ,\s*type:(?P<q_type>["\'])(?P<type>video|audio)/(?P<ext>.+?)(?P=q_type)
 136                         (?:,\s*quality:(?P<q_quality>["\'])(?P<quality>.+?)(?P=q_quality))?
 137                     ''', playerpage):
 138                 url = media.group('url')
 139                 webpage_type = media.group('type')
 140                 ext = media.group('ext')
 141                 res = media.group('quality')
 142                 f = {
 143                     'format_id': '%s_%s' % (res, ext) if res else ext,
 144                     'url': url,
 145                     'ext': ext,
 146                     'vcodec': 'none' if webpage_type == 'audio' else None,
 147                 }
 148                 f.update(self._FORMATS.get(res, {}))
 149                 formats.append(f)
 150             thumbnail = self._og_search_thumbnail(playerpage)
 151             title = self._og_search_title(webpage).strip()
 152             description = self._og_search_description(webpage).strip()
 153         else:
 154             title = self._html_search_regex(
 155                 r'<span class="headline".*?>(.*?)</span>', webpage, 'title')
 156
 157             DOWNLOAD_REGEX = r'(?s)<p>Wir bieten dieses Video in folgenden Formaten zum Download an:</p>\s*<div class="controls">(.*?)</div>\s*<p>'
 158
 159             webpage_type = self._og_search_property('type', webpage, default=None)
 160             if webpage_type == 'website':  # Article
 161                 entries = []
 162                 for num, (entry_title, download_text) in enumerate(re.findall(
 163                         r'(?s)<p[^>]+class="infotext"[^>]*>.*?<strong>(.+?)</strong>.*?</p>.*?%s' % DOWNLOAD_REGEX,
 164                         webpage)):
 165                     entries.append({
 166                         'id': display_id,
 167                         'title': '%s-%d' % (entry_title, num),
 168                         'formats': self._extract_formats(download_text),
 169                     })
 170                 return self.playlist_result(entries, display_id, title)
 171             else:  # Assume single video
 172                 download_text = self._search_regex(DOWNLOAD_REGEX, webpage, 'download links')
 173                 formats = self._extract_formats(download_text)
 174                 thumbnail = self._og_search_thumbnail(webpage)
 175                 description = self._html_search_regex(
 176                     r'(?s)<p class="teasertext">(.*?)</p>',
 177                     webpage, 'description', default=None)
 178
 179         self._sort_formats(formats)
 180
 181         return {
 182             'id': display_id,
 183             'title': title,
 184             'thumbnail': thumbnail,
 185             'formats': formats,
 186             'description': description,
 187         }