[washingtonpost] Fix extractor (#2796)

Closes #2778
Authored by: Bricio
This commit is contained in:
Bricio 2022-02-17 14:38:58 -03:00 committed by GitHub
parent 6bb608d055
commit 5b804e3906
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -5,6 +5,8 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import traverse_obj
class WashingtonPostIE(InfoExtractor): class WashingtonPostIE(InfoExtractor):
IE_NAME = 'washingtonpost' IE_NAME = 'washingtonpost'
@ -50,7 +52,7 @@ class WashingtonPostArticleIE(InfoExtractor):
'title': 'Sinkhole of bureaucracy', 'title': 'Sinkhole of bureaucracy',
}, },
'playlist': [{ 'playlist': [{
'md5': 'b9be794ceb56c7267d410a13f99d801a', 'md5': '7ccf53ea8cbb77de5f570242b3b21a59',
'info_dict': { 'info_dict': {
'id': 'fc433c38-b146-11e3-b8b3-44b1d1cd4c1f', 'id': 'fc433c38-b146-11e3-b8b3-44b1d1cd4c1f',
'ext': 'mp4', 'ext': 'mp4',
@ -59,9 +61,10 @@ class WashingtonPostArticleIE(InfoExtractor):
'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 But the way federal retirement applications are filed may be the most outdated.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.', '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 But the way federal retirement applications are filed may be the most outdated.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.',
'timestamp': 1395440416, 'timestamp': 1395440416,
'upload_date': '20140321', 'upload_date': '20140321',
'thumbnail': r're:https://[^\.]+.cloudfront\.net/PAPERMINESplash\.jpg',
}, },
}, { }, {
'md5': '1fff6a689d8770966df78c8cb6c8c17c', 'md5': '7ccf53ea8cbb77de5f570242b3b21a59',
'info_dict': { 'info_dict': {
'id': '41255e28-b14a-11e3-b8b3-44b1d1cd4c1f', 'id': '41255e28-b14a-11e3-b8b3-44b1d1cd4c1f',
'ext': 'mp4', 'ext': 'mp4',
@ -70,6 +73,7 @@ class WashingtonPostArticleIE(InfoExtractor):
'duration': 2220, 'duration': 2220,
'timestamp': 1395441819, 'timestamp': 1395441819,
'upload_date': '20140321', 'upload_date': '20140321',
'thumbnail': r're:https://[^\.]+.cloudfront\.net/BoyersSplash\.jpeg',
}, },
}], }],
}, { }, {
@ -88,7 +92,11 @@ class WashingtonPostArticleIE(InfoExtractor):
'timestamp': 1419972442, 'timestamp': 1419972442,
'title': 'Why black boxes dont transmit data in real time', 'title': 'Why black boxes dont transmit data in real time',
} }
}] }],
'skip': 'Doesnt have a video anymore',
}, {
'url': 'https://www.washingtonpost.com/nation/2021/08/05/dixie-river-fire-california-climate/',
'only_matching': True,
}] }]
@classmethod @classmethod
@ -106,6 +114,13 @@ def _real_extract(self, url):
<div\s+class="posttv-video-embed[^>]*?data-uuid=| <div\s+class="posttv-video-embed[^>]*?data-uuid=|
data-video-uuid= data-video-uuid=
)"([^"]+)"''', webpage) )"([^"]+)"''', webpage)
if not uuids:
json_data = self._search_nextjs_data(webpage, page_id)
for content_element in traverse_obj(json_data, ('props', 'pageProps', 'globalContent', 'content_elements')):
if content_element.get('type') == 'video':
uuids.append(content_element.get('_id'))
entries = [self.url_result('washingtonpost:%s' % uuid, 'WashingtonPost', uuid) for uuid in uuids] entries = [self.url_result('washingtonpost:%s' % uuid, 'WashingtonPost', uuid) for uuid in uuids]
return { return {