youtube_dl/extractor/vier.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import itertools
   6
   7 from .common import InfoExtractor
   8
   9
  10 class VierIE(InfoExtractor):
  11     IE_NAME = 'vier'
  12     IE_DESC = 'vier.be and vijf.be'
  13     _VALID_URL = r'https?://(?:www\.)?(?P<site>vier|vijf)\.be/(?:[^/]+/videos/(?P<display_id>[^/]+)(?:/(?P<id>\d+))?|video/v3/embed/(?P<embed_id>\d+))'
  14     _TESTS = [{
  15         'url': 'http://www.vier.be/planb/videos/het-wordt-warm-de-moestuin/16129',
  16         'info_dict': {
  17             'id': '16129',
  18             'display_id': 'het-wordt-warm-de-moestuin',
  19             'ext': 'mp4',
  20             'title': 'Het wordt warm in De Moestuin',
  21             'description': 'De vele uren werk eisen hun tol. Wim droomt van assistentie...',
  22         },
  23         'params': {
  24             # m3u8 download
  25             'skip_download': True,
  26         },
  27     }, {
  28         'url': 'http://www.vijf.be/temptationisland/videos/zo-grappig-temptation-island-hosts-moeten-kiezen-tussen-onmogelijke-dilemmas/2561614',
  29         'info_dict': {
  30             'id': '2561614',
  31             'display_id': 'zo-grappig-temptation-island-hosts-moeten-kiezen-tussen-onmogelijke-dilemmas',
  32             'ext': 'mp4',
  33             'title': 'ZO grappig: Temptation Island hosts moeten kiezen tussen onmogelijke dilemma\'s',
  34             'description': 'Het spel is simpel: Annelien Coorevits en Rick Brandsteder krijgen telkens 2 dilemma\'s voorgeschoteld en ze MOETEN een keuze maken.',
  35         },
  36         'params': {
  37             # m3u8 download
  38             'skip_download': True,
  39         },
  40     }, {
  41         'url': 'http://www.vier.be/planb/videos/mieren-herders-van-de-bladluizen',
  42         'only_matching': True,
  43     }, {
  44         'url': 'http://www.vier.be/video/v3/embed/16129',
  45         'only_matching': True,
  46     }]
  47
  48     def _real_extract(self, url):
  49         mobj = re.match(self._VALID_URL, url)
  50         embed_id = mobj.group('embed_id')
  51         display_id = mobj.group('display_id') or embed_id
  52         site = mobj.group('site')
  53
  54         webpage = self._download_webpage(url, display_id)
  55
  56         video_id = self._search_regex(
  57             [r'data-nid="(\d+)"', r'"nid"\s*:\s*"(\d+)"'],
  58             webpage, 'video id')
  59         application = self._search_regex(
  60             [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'],
  61             webpage, 'application', default=site + '_vod')
  62         filename = self._search_regex(
  63             [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'],
  64             webpage, 'filename')
  65
  66         playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename)
  67         formats = self._extract_wowza_formats(playlist_url, display_id, skip_protocols=['dash'])
  68         self._sort_formats(formats)
  69
  70         title = self._og_search_title(webpage, default=display_id)
  71         description = self._og_search_description(webpage, default=None)
  72         thumbnail = self._og_search_thumbnail(webpage, default=None)
  73
  74         return {
  75             'id': video_id,
  76             'display_id': display_id,
  77             'title': title,
  78             'description': description,
  79             'thumbnail': thumbnail,
  80             'formats': formats,
  81         }
  82
  83
  84 class VierVideosIE(InfoExtractor):
  85     IE_NAME = 'vier:videos'
  86     _VALID_URL = r'https?://(?:www\.)?(?P<site>vier|vijf)\.be/(?P<program>[^/]+)/videos(?:\?.*\bpage=(?P<page>\d+)|$)'
  87     _TESTS = [{
  88         'url': 'http://www.vier.be/demoestuin/videos',
  89         'info_dict': {
  90             'id': 'demoestuin',
  91         },
  92         'playlist_mincount': 153,
  93     }, {
  94         'url': 'http://www.vijf.be/temptationisland/videos',
  95         'info_dict': {
  96             'id': 'temptationisland',
  97         },
  98         'playlist_mincount': 159,
  99     }, {
 100         'url': 'http://www.vier.be/demoestuin/videos?page=6',
 101         'info_dict': {
 102             'id': 'demoestuin-page6',
 103         },
 104         'playlist_mincount': 20,
 105     }, {
 106         'url': 'http://www.vier.be/demoestuin/videos?page=7',
 107         'info_dict': {
 108             'id': 'demoestuin-page7',
 109         },
 110         'playlist_mincount': 13,
 111     }]
 112
 113     def _real_extract(self, url):
 114         mobj = re.match(self._VALID_URL, url)
 115         program = mobj.group('program')
 116         site = mobj.group('site')
 117
 118         page_id = mobj.group('page')
 119         if page_id:
 120             page_id = int(page_id)
 121             start_page = page_id
 122             playlist_id = '%s-page%d' % (program, page_id)
 123         else:
 124             start_page = 0
 125             playlist_id = program
 126
 127         entries = []
 128         for current_page_id in itertools.count(start_page):
 129             current_page = self._download_webpage(
 130                 'http://www.%s.be/%s/videos?page=%d' % (site, program, current_page_id),
 131                 program,
 132                 'Downloading page %d' % (current_page_id + 1))
 133             page_entries = [
 134                 self.url_result('http://www.' + site + '.be' + video_url, 'Vier')
 135                 for video_url in re.findall(
 136                     r'<h[23]><a href="(/[^/]+/videos/[^/]+(?:/\d+)?)">', current_page)]
 137             entries.extend(page_entries)
 138             if page_id or '>Meer<' not in current_page:
 139                 break
 140
 141         return self.playlist_result(entries, playlist_id)