[theintercept] improve extraction

This commit is contained in:
remitamine 2015-12-23 22:36:53 +01:00
parent c153bd8b2f
commit 96db61ffb8

View File

@ -1,24 +1,28 @@
# encoding: utf-8 # encoding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import compat_str
from ..utils import ( from ..utils import (
parse_iso8601,
int_or_none,
ExtractorError, ExtractorError,
) )
class TheInterceptIE(InfoExtractor): class TheInterceptIE(InfoExtractor):
_VALID_URL = r'https://theintercept.com/fieldofvision/(?P<id>.+?)/' _VALID_URL = r'https://theintercept.com/fieldofvision/(?P<id>[^/?#]+)'
_TESTS = [{ _TESTS = [{
'url': 'https://theintercept.com/fieldofvision/thisisacoup-episode-four-surrender-or-die/', 'url': 'https://theintercept.com/fieldofvision/thisisacoup-episode-four-surrender-or-die/',
'md5': '145f28b41d44aab2f87c0a4ac8ec95bd',
'info_dict': { 'info_dict': {
'id': 'thisisacoup-episode-four-surrender-or-die', 'id': '46214',
'ext': 'mp4', 'ext': 'mp4',
'title': '#ThisIsACoup Episode Four: Surrender or Die', 'title': '#ThisIsACoup Episode Four: Surrender or Die',
'upload_date': '20151218',
'description': 'md5:74dd27f0e2fbd50817829f97eaa33140', 'description': 'md5:74dd27f0e2fbd50817829f97eaa33140',
'timestamp': 1450429239,
'upload_date': '20151218',
'comment_count': int,
} }
}] }]
@ -26,43 +30,20 @@ def _real_extract(self, url):
display_id = self._match_id(url) display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id) webpage = self._download_webpage(url, display_id)
mobj = re.search(r'initialStoreTree =(?P<json_data>.+})', webpage) json_data = self._parse_json(self._search_regex(
if mobj is None: r'initialStoreTree\s*=\s*(?P<json_data>{.+})', webpage,
raise ExtractorError('Unable to extract initialStoreTree') 'initialStoreTree'), display_id)
json_data = self._parse_json(mobj.group('json_data'), display_id)
info = None
for post in json_data['resources']['posts'].values(): for post in json_data['resources']['posts'].values():
if post['slug'] == display_id: if post['slug'] == display_id:
info = post return {
break '_type': 'url_transparent',
if info is None: 'url': 'jwplatform:%s' % post['fov_videoid'],
raise ExtractorError('Unable to find info for %s'%display_id) 'id': compat_str(post['ID']),
'display_id': display_id,
title = info['title'] 'title': post['title'],
description = info['excerpt'] 'description': post.get('excerpt'),
upload_date = info['date'][:10].replace('-', '') 'timestamp': parse_iso8601(post.get('date')),
video_id = info['fov_videoid'] 'comment_count': int_or_none(post.get('comments_number')),
creator = ','.join([a['display_name'] for a in info['authors']]) }
thumbnail = self._og_search_property('image', webpage) raise ExtractorError('Unable to find the current post')
content_id = thumbnail.split('/')[-1].split('.')[0]
content_url = 'https://content.jwplatform.com/jw6/{content_id}.xml'.format(content_id=content_id)
content = self._download_xml(content_url, video_id)
formats = []
for source in content.findall('.//{http://rss.jwpcdn.com/}source'):
if source.attrib['file'].endswith('.m3u8'):
formats.extend(self._extract_m3u8_formats(
source.attrib['file'], video_id, 'mp4', preference=1, m3u8_id='hls'))
return {
'creator': creator,
'description': description,
'display_id': display_id,
'formats': formats,
'id': video_id,
'id': video_id,
'thumbnail': thumbnail,
'title': title,
'upload_date': upload_date,
}