1
0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-05 02:32:44 +01:00
yt-dlp/youtube_dl/extractor/democracynow.py
2015-10-31 22:21:52 +08:00

85 lines
3.4 KiB
Python

# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
class DemocracynowIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?democracynow.org/?(?P<id>[^\?]*)'
IE_NAME = 'democracynow'
_TESTS = [{
'url': 'http://www.democracynow.org/shows/2015/7/3',
'info_dict': {
'id': '2015-0703-001',
'ext': 'mp4',
'title': 'July 03, 2015 - Democracy Now!',
'description': 'A daily independent global news hour with Amy Goodman & Juan Gonz\xe1lez "What to the Slave is 4th of July?": James Earl Jones Reads Frederick Douglass\u2019 Historic Speech : "This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag : "We Shall Overcome": Remembering Folk Icon, Activist Pete Seeger in His Own Words & Songs',
'uploader': 'Democracy Now',
'upload_date': None,
},
}, {
'url': 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree',
'info_dict': {
'id': '2015-0703-001',
'ext': 'mp4',
'title': '"This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag',
'description': 'md5:4d2bc4f0d29f5553c2210a4bc7761a21',
'uploader': 'Democracy Now',
'upload_date': None,
},
}]
def _real_extract(self, url):
display_id = self._match_id(url)
base_host = re.search(r'^(.+?://[^/]+)', url).group(1)
if display_id == '':
display_id = 'home'
webpage = self._download_webpage(url, display_id)
description = self._og_search_description(webpage)
jstr = self._search_regex(r'<script[^>]+type="text/json"[^>]*>\s*({[^>]+})', webpage, 'json')
js = self._parse_json(jstr, display_id)
video_id = None
formats = []
subtitles = {}
for key in ('caption_file', '.......'):
# ....... = pending vtt support that doesn't clobber srt 'chapter_file':
url = js.get(key, '')
if url == '' or url is None:
continue
if not re.match(r'^https?://', url):
url = base_host + url
ext = re.search(r'\.([^\.]+)$', url).group(1)
subtitles['eng'] = [{
'ext': ext,
'url': url,
}]
for key in ('file', 'audio', 'video'):
url = js.get(key, '')
if url == '' or url is None:
continue
if not re.match(r'^https?://', url):
url = base_host + url
purl = re.search(r'/(?P<dir>[^/]+)/(?:dn)?(?P<fn>[^/]+?)\.(?P<ext>[^\.\?]+)(?P<hasparams>\?|$)', url)
if video_id is None:
video_id = purl.group('fn')
if js.get('start') is not None:
url += '&' if purl.group('hasparams') == '?' else '?'
url = url + 'start=' + str(js.get('start'))
formats.append({
'format_id': purl.group('dir'),
'ext': purl.group('ext'),
'url': url,
})
self._sort_formats(formats)
ret = {
'id': video_id,
'title': js.get('title'),
'description': description,
'uploader': 'Democracy Now',
'subtitles': subtitles,
'formats': formats,
}
return ret