2016-10-06 05:09:53 +00:00
|
|
|
# coding: utf-8
|
|
|
|
|
from __future__ import unicode_literals
|
|
|
|
|
|
|
|
|
|
import re
|
|
|
|
|
|
|
|
|
|
from .common import InfoExtractor
|
2016-10-06 05:38:25 +00:00
|
|
|
from ..compat import compat_urlparse
|
2016-10-06 05:09:53 +00:00
|
|
|
from .generic import GenericIE
|
|
|
|
|
|
|
|
|
|
# IQM2 aka Accela is a municipal meeting management platform that
|
|
|
|
|
# (among other things) stores livestreamed video from municipal
|
|
|
|
|
# meetings. After a hefty (several-hour) processing time, that video
|
2016-10-06 05:51:06 +00:00
|
|
|
# is available in easily downloadable form from their web portal, but
|
2016-10-06 05:09:53 +00:00
|
|
|
# prior to that, the video can only be watched in realtime through
|
|
|
|
|
# JWPlayer. This extractor is designed to download the realtime video
|
2016-10-06 05:51:06 +00:00
|
|
|
# prior to download links being available. For more info on Accela, see:
|
2016-10-06 05:09:53 +00:00
|
|
|
# http://www.iqm2.com/About/Accela.aspx
|
|
|
|
|
# http://www.accela.com/
|
|
|
|
|
|
2016-10-08 23:38:28 +00:00
|
|
|
# This processing makes hard to test since there's only a narrow
|
|
|
|
|
# window when it matters. After that the extractor finds links to
|
|
|
|
|
# the processed video intead.
|
2016-10-06 05:09:53 +00:00
|
|
|
|
2016-10-08 23:38:28 +00:00
|
|
|
# No metadata is retrieved, as that would require finding a metadata
|
|
|
|
|
# URL and retreiving a 3rd HTTP resource.
|
2016-10-06 05:29:58 +00:00
|
|
|
|
2016-10-06 05:09:53 +00:00
|
|
|
# Contributed by John Hawkinson <jhawk@mit.edu>, 6 Oct 2016.
|
|
|
|
|
|
|
|
|
|
class IQM2IE(InfoExtractor):
|
2016-10-06 05:29:58 +00:00
|
|
|
|
2016-10-08 23:21:33 +00:00
|
|
|
# We commonly see both iqm2.com and IQM2.com.
|
|
|
|
|
_VALID_URL = r'(?i)https?://(?:\w+\.)?iqm2\.com/Citizens/\w+.aspx\?.*MeetingID=(?P<id>[0-9]+)'
|
2016-10-08 23:17:32 +00:00
|
|
|
_TESTS = [ {
|
2016-10-06 05:09:53 +00:00
|
|
|
'url': 'http://cambridgema.iqm2.com/Citizens/SplitView.aspx?Mode=Video&MeetingID=1679#',
|
|
|
|
|
'md5': '478ea30eee1966f7be0d8dd623122148',
|
|
|
|
|
'info_dict': {
|
|
|
|
|
'id': '1563_720',
|
|
|
|
|
'ext': 'mp4',
|
|
|
|
|
'title': 'Cambridge, MA (2)',
|
|
|
|
|
'uploader': 'cambridgema.iqm2.com',
|
2016-10-08 23:17:32 +00:00
|
|
|
}}, {
|
|
|
|
|
'url': 'https://CambridgeMA.IQM2.com/Citizens/VideoMain.aspx?MeetingID=1679',
|
|
|
|
|
'only_matching': True,
|
|
|
|
|
}]
|
2016-10-06 05:09:53 +00:00
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
2016-10-08 23:21:33 +00:00
|
|
|
mobj = re.match(self._VALID_URL, url)
|
2016-10-06 05:09:53 +00:00
|
|
|
video_id = mobj.group('id')
|
|
|
|
|
|
|
|
|
|
webpage = self._download_webpage(url, video_id)
|
|
|
|
|
|
2016-10-06 05:42:30 +00:00
|
|
|
# print "Original URL is", url
|
|
|
|
|
|
2016-10-08 23:38:28 +00:00
|
|
|
# Simple extractor: take, e.g.
|
|
|
|
|
# http://cambridgema.iqm2.com/Citizens/SplitView.aspx?Mode=Video&MeetingID=1679
|
|
|
|
|
# and look for
|
|
|
|
|
# <div id="VideoPanel" class="LeftTopContent">
|
|
|
|
|
# <div id="VideoPanelInner" ... src="/Citizens/VideoScreen.aspx?MediaID=1563&Frame=SplitView">
|
|
|
|
|
# and feed the canonicalized src element to the generic extractor
|
2016-10-06 05:38:25 +00:00
|
|
|
inner_url_rel = self._html_search_regex(
|
|
|
|
|
r'<div id="VideoPanelInner".*src="([^"]+)"',
|
|
|
|
|
webpage, 'url');
|
2016-10-06 05:42:30 +00:00
|
|
|
# print "inner_URL is", inner_url_rel
|
2016-10-06 05:38:25 +00:00
|
|
|
|
|
|
|
|
inner_url = compat_urlparse.urljoin(url, inner_url_rel)
|
2016-10-06 05:42:30 +00:00
|
|
|
# print "Joined URL is", inner_url
|
2016-10-06 05:38:25 +00:00
|
|
|
|
2016-10-08 23:31:50 +00:00
|
|
|
# Generic extractor matches this under the "Broaden the
|
|
|
|
|
# findall a little bit: JWPlayer JS loader" (line 2372 as of 6
|
|
|
|
|
# Oct 2016, dcdb292fddc82ae11f4c0b647815a45c88a6b6d5).
|
|
|
|
|
return self.url_result(inner_url, 'Generic')
|
