vrt.py (3523B)
1 # coding: utf-8 2 from __future__ import unicode_literals 3 4 import re 5 6 from .common import InfoExtractor 7 from ..utils import ( 8 extract_attributes, 9 float_or_none, 10 get_element_by_class, 11 strip_or_none, 12 unified_timestamp, 13 ) 14 15 16 class VRTIE(InfoExtractor): 17 IE_DESC = 'VRT NWS, Flanders News, Flandern Info and Sporza' 18 _VALID_URL = r'https?://(?:www\.)?(?P<site>vrt\.be/vrtnws|sporza\.be)/[a-z]{2}/\d{4}/\d{2}/\d{2}/(?P<id>[^/?&#]+)' 19 _TESTS = [{ 20 'url': 'https://www.vrt.be/vrtnws/nl/2019/05/15/beelden-van-binnenkant-notre-dame-een-maand-na-de-brand/', 21 'md5': 'e1663accf5cf13f375f3cd0d10476669', 22 'info_dict': { 23 'id': 'pbs-pub-7855fc7b-1448-49bc-b073-316cb60caa71$vid-2ca50305-c38a-4762-9890-65cbd098b7bd', 24 'ext': 'mp4', 25 'title': 'Beelden van binnenkant Notre-Dame, één maand na de brand', 26 'description': 'Op maandagavond 15 april ging een deel van het dakgebinte van de Parijse kathedraal in vlammen op.', 27 'timestamp': 1557924660, 28 'upload_date': '20190515', 29 'duration': 31.2, 30 }, 31 }, { 32 'url': 'https://sporza.be/nl/2019/05/15/de-belgian-cats-zijn-klaar-voor-het-ek/', 33 'md5': '910bba927566e9ab992278f647eb4b75', 34 'info_dict': { 35 'id': 'pbs-pub-f2c86a46-8138-413a-a4b9-a0015a16ce2c$vid-1f112b31-e58e-4379-908d-aca6d80f8818', 36 'ext': 'mp4', 37 'title': 'De Belgian Cats zijn klaar voor het EK mét Ann Wauters', 38 'timestamp': 1557923760, 39 'upload_date': '20190515', 40 'duration': 115.17, 41 }, 42 }, { 43 'url': 'https://www.vrt.be/vrtnws/en/2019/05/15/belgium_s-eurovision-entry-falls-at-the-first-hurdle/', 44 'only_matching': True, 45 }, { 46 'url': 'https://www.vrt.be/vrtnws/de/2019/05/15/aus-fuer-eliott-im-halbfinale-des-eurosongfestivals/', 47 'only_matching': True, 48 }] 49 _CLIENT_MAP = { 50 'vrt.be/vrtnws': 'vrtnieuws', 51 'sporza.be': 'sporza', 52 } 53 54 def _real_extract(self, url): 55 site, display_id = re.match(self._VALID_URL, url).groups() 56 webpage = self._download_webpage(url, display_id) 57 attrs = extract_attributes(self._search_regex( 58 r'(<[^>]+class="vrtvideo"[^>]*>)', webpage, 'vrt video')) 59 60 asset_id = attrs['data-videoid'] 61 publication_id = attrs.get('data-publicationid') 62 if publication_id: 63 asset_id = publication_id + '$' + asset_id 64 client = attrs.get('data-client') or self._CLIENT_MAP[site] 65 66 title = strip_or_none(get_element_by_class( 67 'vrt-title', webpage) or self._html_search_meta( 68 ['og:title', 'twitter:title', 'name'], webpage)) 69 description = self._html_search_meta( 70 ['og:description', 'twitter:description', 'description'], webpage) 71 if description == '…': 72 description = None 73 timestamp = unified_timestamp(self._html_search_meta( 74 'article:published_time', webpage)) 75 76 return { 77 '_type': 'url_transparent', 78 'id': asset_id, 79 'display_id': display_id, 80 'title': title, 81 'description': description, 82 'thumbnail': attrs.get('data-posterimage'), 83 'timestamp': timestamp, 84 'duration': float_or_none(attrs.get('data-duration'), 1000), 85 'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (client, asset_id), 86 'ie_key': 'Canvas', 87 }