yt_dlp/extractor/washingtonpost.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8 from ..utils import traverse_obj
   9
  10
  11 class WashingtonPostIE(InfoExtractor):
  12     IE_NAME = 'washingtonpost'
  13     _VALID_URL = r'(?:washingtonpost:|https?://(?:www\.)?washingtonpost\.com/(?:video|posttv)/(?:[^/]+/)*)(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
  14     _EMBED_URL = r'https?://(?:www\.)?washingtonpost\.com/video/c/embed/[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
  15     _TESTS = [{
  16         'url': 'https://www.washingtonpost.com/video/c/video/480ba4ee-1ec7-11e6-82c2-a7dcb313287d',
  17         'md5': '6f537e1334b714eb15f9563bd4b9cdfa',
  18         'info_dict': {
  19             'id': '480ba4ee-1ec7-11e6-82c2-a7dcb313287d',
  20             'ext': 'mp4',
  21             'title': 'Egypt finds belongings, debris from plane crash',
  22             'description': 'md5:a17ceee432f215a5371388c1f680bd86',
  23             'upload_date': '20160520',
  24             'timestamp': 1463775187,
  25         },
  26     }, {
  27         'url': 'https://www.washingtonpost.com/video/world/egypt-finds-belongings-debris-from-plane-crash/2016/05/20/480ba4ee-1ec7-11e6-82c2-a7dcb313287d_video.html',
  28         'only_matching': True,
  29     }, {
  30         'url': 'https://www.washingtonpost.com/posttv/world/iraq-to-track-down-antiquities-after-islamic-state-museum-rampage/2015/02/28/7c57e916-bf86-11e4-9dfb-03366e719af8_video.html',
  31         'only_matching': True,
  32     }]
  33
  34     @classmethod
  35     def _extract_urls(cls, webpage):
  36         return re.findall(
  37             r'<iframe[^>]+\bsrc=["\'](%s)' % cls._EMBED_URL, webpage)
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41         return self.url_result(
  42             'arcpublishing:wapo:' + video_id, 'ArcPublishing', video_id)
  43
  44
  45 class WashingtonPostArticleIE(InfoExtractor):
  46     IE_NAME = 'washingtonpost:article'
  47     _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/(?:[^/]+/)*(?P<id>[^/?#]+)'
  48     _TESTS = [{
  49         'url': 'http://www.washingtonpost.com/sf/national/2014/03/22/sinkhole-of-bureaucracy/',
  50         'info_dict': {
  51             'id': 'sinkhole-of-bureaucracy',
  52             'title': 'Sinkhole of bureaucracy',
  53         },
  54         'playlist': [{
  55             'md5': '7ccf53ea8cbb77de5f570242b3b21a59',
  56             'info_dict': {
  57                 'id': 'fc433c38-b146-11e3-b8b3-44b1d1cd4c1f',
  58                 'ext': 'mp4',
  59                 'title': 'Breaking Points: The Paper Mine',
  60                 'duration': 1290,
  61                 'description': 'Overly complicated paper pushing is nothing new to government bureaucracy. But the way federal retirement applications are filed may be the most outdated. David Fahrenthold explains.',
  62                 'timestamp': 1395440416,
  63                 'upload_date': '20140321',
  64                 'thumbnail': r're:https://[^\.]+.cloudfront\.net/PAPERMINESplash\.jpg',
  65             },
  66         }, {
  67             'md5': '7ccf53ea8cbb77de5f570242b3b21a59',
  68             'info_dict': {
  69                 'id': '41255e28-b14a-11e3-b8b3-44b1d1cd4c1f',
  70                 'ext': 'mp4',
  71                 'title': 'The town bureaucracy sustains',
  72                 'description': 'Underneath the friendly town of Boyers is a sea of government paperwork. In a disused limestone mine, hundreds of locals now track, file and process retirement applications for the federal government. We set out to find out what it\'s like to do paperwork 230 feet underground.',
  73                 'duration': 2220,
  74                 'timestamp': 1395441819,
  75                 'upload_date': '20140321',
  76                 'thumbnail': r're:https://[^\.]+.cloudfront\.net/BoyersSplash\.jpeg',
  77             },
  78         }],
  79     }, {
  80         'url': 'http://www.washingtonpost.com/blogs/wonkblog/wp/2014/12/31/one-airline-figured-out-how-to-make-sure-its-airplanes-never-disappear/',
  81         'info_dict': {
  82             'id': 'one-airline-figured-out-how-to-make-sure-its-airplanes-never-disappear',
  83             'title': 'One airline figured out how to make sure its airplanes never disappear',
  84         },
  85         'playlist': [{
  86             'md5': 'a7c1b5634ba5e57a6a82cdffa5b1e0d0',
  87             'info_dict': {
  88                 'id': '0e4bb54c-9065-11e4-a66f-0ca5037a597d',
  89                 'ext': 'mp4',
  90                 'description': 'Washington Post transportation reporter Ashley Halsey III explains why a plane\'s black box needs to be recovered from a crash site instead of having its information streamed in real time throughout the flight.',
  91                 'upload_date': '20141230',
  92                 'timestamp': 1419972442,
  93                 'title': 'Why black boxes don’t transmit data in real time',
  94             }
  95         }],
  96         'skip': 'Doesnt have a video anymore',
  97     }, {
  98         'url': 'https://www.washingtonpost.com/nation/2021/08/05/dixie-river-fire-california-climate/',
  99         'only_matching': True,
 100     }]
 101
 102     @classmethod
 103     def suitable(cls, url):
 104         return False if WashingtonPostIE.suitable(url) else super(WashingtonPostArticleIE, cls).suitable(url)
 105
 106     def _real_extract(self, url):
 107         page_id = self._match_id(url)
 108         webpage = self._download_webpage(url, page_id)
 109
 110         title = self._og_search_title(webpage)
 111
 112         uuids = re.findall(r'''(?x)
 113             (?:
 114                 <div\s+class="posttv-video-embed[^>]*?data-uuid=|
 115                 data-video-uuid=
 116             )"([^"]+)"''', webpage)
 117
 118         if not uuids:
 119             json_data = self._search_nextjs_data(webpage, page_id)
 120             for content_element in traverse_obj(json_data, ('props', 'pageProps', 'globalContent', 'content_elements')):
 121                 if content_element.get('type') == 'video':
 122                     uuids.append(content_element.get('_id'))
 123
 124         entries = [self.url_result('washingtonpost:%s' % uuid, 'WashingtonPost', uuid) for uuid in uuids]
 125
 126         return {
 127             '_type': 'playlist',
 128             'entries': entries,
 129             'id': page_id,
 130             'title': title,
 131         }