Raphaël G. Git Repositories - youtubedl/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/multimedia/(?:sendung/ts|video/video)(?P<id>-?[0-9]+)\.html'
  12
  13     _TESTS = [{
  14         'url': 'http://www.tagesschau.de/multimedia/video/video1399128.html',
  15         'md5': 'bcdeac2194fb296d599ce7929dfa4009',
  16         'info_dict': {
  17             'id': '1399128',
  18             'ext': 'mp4',
  19             'title': 'Harald Range, Generalbundesanwalt, zu den Ermittlungen',
  20             'description': 'md5:69da3c61275b426426d711bde96463ab',
  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
  35     _FORMATS = {
  36         's': {'width': 256, 'height': 144, 'quality': 1},
  37         'm': {'width': 512, 'height': 288, 'quality': 2},
  38         'l': {'width': 960, 'height': 544, 'quality': 3},
  39     }
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43         display_id = video_id.lstrip('-')
  44         webpage = self._download_webpage(url, display_id)
  45
  46         player_url = self._html_search_meta(
  47             'twitter:player', webpage, 'player URL', default=None)
  48         if player_url:
  49             playerpage = self._download_webpage(
  50                 player_url, display_id, 'Downloading player page')
  51
  52             medias = re.findall(
  53                 r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"',
  54                 playerpage)
  55             formats = []
  56             for url, ext, res in medias:
  57                 f = {
  58                     'format_id': res + '_' + ext,
  59                     'url': url,
  60                     'ext': ext,
  61                 }
  62                 f.update(self._FORMATS.get(res, {}))
  63                 formats.append(f)
  64             thumbnail_fn = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1]
  65             title = self._og_search_title(webpage).strip()
  66             description = self._og_search_description(webpage).strip()
  67         else:
  68             download_text = self._search_regex(
  69                 r'(?s)<p>Wir bieten dieses Video in folgenden Formaten zum Download an:</p>\s*<div class="controls">(.*?)</div>\s*<p>',
  70                 webpage, 'download links')
  71             links = re.finditer(
  72                 r'<div class="button" title="(?P<title>[^"]*)"><a href="(?P<url>[^"]+)">(?P<name>.+?)</a></div>',
  73                 download_text)
  74             formats = []
  75             for l in links:
  76                 format_id = self._search_regex(
  77                     r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID')
  78                 format = {
  79                     'format_id': format_id,
  80                     'url': l.group('url'),
  81                     'format_name': l.group('name'),
  82                 }
  83                 m = re.match(
  84                     r'''(?x)
  85                         Video:\s*(?P<vcodec>[a-zA-Z0-9/._-]+)\s*&\#10;
  86                         (?P<width>[0-9]+)x(?P<height>[0-9]+)px&\#10;
  87                         (?P<vbr>[0-9]+)kbps&\#10;
  88                         Audio:\s*(?P<abr>[0-9]+)kbps,\s*(?P<audio_desc>[A-Za-z\.0-9]+)&\#10;
  89                         Gr&ouml;&szlig;e:\s*(?P<filesize_approx>[0-9.,]+\s+[a-zA-Z]*B)''',
  90                     l.group('title'))
  91                 if m:
  92                     format.update({
  93                         'format_note': m.group('audio_desc'),
  94                         'vcodec': m.group('vcodec'),
  95                         'width': int(m.group('width')),
  96                         'height': int(m.group('height')),
  97                         'abr': int(m.group('abr')),
  98                         'vbr': int(m.group('vbr')),
  99                         'filesize_approx': parse_filesize(m.group('filesize_approx')),
 100                     })
 101                 formats.append(format)
 102             thumbnail_fn = self._search_regex(
 103                 r'(?s)<img alt="Sendungsbild".*?src="([^"]+)"',
 104                 webpage, 'thumbnail', fatal=False)
 105             description = self._html_search_regex(
 106                 r'(?s)<p class="teasertext">(.*?)</p>',
 107                 webpage, 'description', fatal=False)
 108             title = self._html_search_regex(
 109                 r'<span class="headline".*?>(.*?)</span>', webpage, 'title')
 110
 111         self._sort_formats(formats)
 112         thumbnail = 'http://www.tagesschau.de' + thumbnail_fn
 113
 114         return {
 115             'id': display_id,
 116             'title': title,
 117             'thumbnail': thumbnail,
 118             'formats': formats,
 119             'description': description,
 120         }