diff options
Diffstat (limited to 'youtube_dl/extractor/washingtonpost.py')
| -rw-r--r-- | youtube_dl/extractor/washingtonpost.py | 107 |
1 files changed, 46 insertions, 61 deletions
diff --git a/youtube_dl/extractor/washingtonpost.py b/youtube_dl/extractor/washingtonpost.py index ec8b99998..8afb1af83 100644 --- a/youtube_dl/extractor/washingtonpost.py +++ b/youtube_dl/extractor/washingtonpost.py @@ -4,14 +4,45 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import ( - int_or_none, - strip_jsonp, -) class WashingtonPostIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/.*?/(?P<id>[^/]+)/(?:$|[?#])' + IE_NAME = 'washingtonpost' + _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})' + _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}' + _TESTS = [{ + 'url': 'https://www.washingtonpost.com/video/c/video/480ba4ee-1ec7-11e6-82c2-a7dcb313287d', + 'md5': '6f537e1334b714eb15f9563bd4b9cdfa', + 'info_dict': { + 'id': '480ba4ee-1ec7-11e6-82c2-a7dcb313287d', + 'ext': 'mp4', + 'title': 'Egypt finds belongings, debris from plane crash', + 'description': 'md5:a17ceee432f215a5371388c1f680bd86', + 'upload_date': '20160520', + 'timestamp': 1463775187, + }, + }, { + 'url': 'https://www.washingtonpost.com/video/world/egypt-finds-belongings-debris-from-plane-crash/2016/05/20/480ba4ee-1ec7-11e6-82c2-a7dcb313287d_video.html', + 'only_matching': True, + }, { + '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', + 'only_matching': True, + }] + + @classmethod + def _extract_urls(cls, webpage): + return re.findall( + r'<iframe[^>]+\bsrc=["\'](%s)' % cls._EMBED_URL, webpage) + + def _real_extract(self, url): + video_id = self._match_id(url) + return self.url_result( + 'arcpublishing:wapo:' + video_id, 'ArcPublishing', video_id) + + +class WashingtonPostArticleIE(InfoExtractor): + IE_NAME = 'washingtonpost:article' + _VALID_URL = r'https?://(?:www\.)?washingtonpost\.com/(?:[^/]+/)*(?P<id>[^/?#]+)' _TESTS = [{ 'url': 'http://www.washingtonpost.com/sf/national/2014/03/22/sinkhole-of-bureaucracy/', 'info_dict': { @@ -26,9 +57,8 @@ class WashingtonPostIE(InfoExtractor): 'title': 'Breaking Points: The Paper Mine', 'duration': 1290, '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.', - 'uploader': 'The Washington Post', - 'timestamp': 1395527908, - 'upload_date': '20140322', + 'timestamp': 1395440416, + 'upload_date': '20140321', }, }, { 'md5': '1fff6a689d8770966df78c8cb6c8c17c', @@ -38,9 +68,8 @@ class WashingtonPostIE(InfoExtractor): 'title': 'The town bureaucracy sustains', '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.', 'duration': 2220, - 'timestamp': 1395528005, - 'upload_date': '20140322', - 'uploader': 'The Washington Post', + 'timestamp': 1395441819, + 'upload_date': '20140321', }, }], }, { @@ -56,13 +85,16 @@ class WashingtonPostIE(InfoExtractor): 'ext': 'mp4', '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.', 'upload_date': '20141230', - 'uploader': 'The Washington Post', - 'timestamp': 1419974765, + 'timestamp': 1419972442, 'title': 'Why black boxes don’t transmit data in real time', } }] }] + @classmethod + def suitable(cls, url): + return False if WashingtonPostIE.suitable(url) else super(WashingtonPostArticleIE, cls).suitable(url) + def _real_extract(self, url): page_id = self._match_id(url) webpage = self._download_webpage(url, page_id) @@ -74,54 +106,7 @@ class WashingtonPostIE(InfoExtractor): <div\s+class="posttv-video-embed[^>]*?data-uuid=| data-video-uuid= )"([^"]+)"''', webpage) - entries = [] - for i, uuid in enumerate(uuids, start=1): - vinfo_all = self._download_json( - 'http://www.washingtonpost.com/posttv/c/videojson/%s?resType=jsonp' % uuid, - page_id, - transform_source=strip_jsonp, - note='Downloading information of video %d/%d' % (i, len(uuids)) - ) - vinfo = vinfo_all[0]['contentConfig'] - uploader = vinfo.get('credits', {}).get('source') - timestamp = int_or_none( - vinfo.get('dateConfig', {}).get('dateFirstPublished'), 1000) - - formats = [{ - 'format_id': ( - '%s-%s-%s' % (s.get('type'), s.get('width'), s.get('bitrate')) - if s.get('width') - else s.get('type')), - 'vbr': s.get('bitrate') if s.get('width') != 0 else None, - 'width': s.get('width'), - 'height': s.get('height'), - 'acodec': s.get('audioCodec'), - 'vcodec': s.get('videoCodec') if s.get('width') != 0 else 'none', - 'filesize': s.get('fileSize'), - 'url': s.get('url'), - 'ext': 'mp4', - 'preference': -100 if s.get('type') == 'smil' else None, - 'protocol': { - 'MP4': 'http', - 'F4F': 'f4m', - }.get(s.get('type')), - } for s in vinfo.get('streams', [])] - source_media_url = vinfo.get('sourceMediaURL') - if source_media_url: - formats.append({ - 'format_id': 'source_media', - 'url': source_media_url, - }) - self._sort_formats(formats) - entries.append({ - 'id': uuid, - 'title': vinfo['title'], - 'description': vinfo.get('blurb'), - 'uploader': uploader, - 'formats': formats, - 'duration': int_or_none(vinfo.get('videoDuration'), 100), - 'timestamp': timestamp, - }) + entries = [self.url_result('washingtonpost:%s' % uuid, 'WashingtonPost', uuid) for uuid in uuids] return { '_type': 'playlist', |