[franceinter] Fix extraction and modernize

This commit is contained in:
Sergey M․ 2014-09-19 20:58:50 +07:00 committed by Sergey M
parent f566d9f1d5
commit 532f5bff70

View File

@ -4,16 +4,21 @@
import re
from .common import InfoExtractor
from ..utils import int_or_none
class FranceInterIE(InfoExtractor):
_VALID_URL = r'http://(?:www\.)?franceinter\.fr/player/reecouter\?play=(?P<id>[0-9]{6})'
_VALID_URL = r'http://(?:www\.)?franceinter\.fr/player/reecouter\?play=(?P<id>[0-9]+)'
_TEST = {
'url': 'http://www.franceinter.fr/player/reecouter?play=793962',
'file': '793962.mp3',
'md5': '4764932e466e6f6c79c317d2e74f6884',
"info_dict": {
"title": "LHistoire dans les jeux vidéo",
'id': '793962',
'ext': 'mp3',
'title': 'LHistoire dans les jeux vidéo',
'description': 'md5:7e93ddb4451e7530022792240a3049c7',
'timestamp': 1387369800,
'upload_date': '20131218',
},
}
@ -22,17 +27,26 @@ def _real_extract(self, url):
video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id)
title = self._html_search_regex(
r'<span class="roll_overflow">(.*?)</span></h1>', webpage, 'title')
path = self._search_regex(
r'&urlAOD=(.*?)&startTime', webpage, 'video url')
r'<a id="player".+?href="([^"]+)"', webpage, 'video url')
video_url = 'http://www.franceinter.fr/' + path
title = self._html_search_regex(
r'<span class="title">(.+?)</span>', webpage, 'title')
description = self._html_search_regex(
r'<span class="description">(.*?)</span>',
webpage, 'description', fatal=False)
timestamp = int_or_none(self._search_regex(
r'data-date="(\d+)"', webpage, 'upload date', fatal=False))
return {
'id': video_id,
'title': title,
'description': description,
'timestamp': timestamp,
'formats': [{
'url': video_url,
'vcodec': 'none',
}],
'title': title,
}