#!/usr/bin/env python # Record all the UDS streams # Copyright (c) 2010 - 2011, Stefano Rivera # # This program is free software: you can redistribute it and/or modify it under # the terms of the GNU General Public License as published by the Free Software # Foundation, either version 3 of the License, or (at your option) any later # version. # # This program is distributed in the hope that it will be useful, but WITHOUT # ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS # FOR A PARTICULAR PURPOSE. See the GNU General Public License for more # details. # # You should have received a copy of the GNU General Public License along with # this program. If not, see . import csv import datetime import itertools import logging import os import re import urlparse import dateutil.parser from twisted.internet import protocol, reactor, stdio, task from twisted.protocols import basic from twisted.protocols.shoutcast import ShoutcastClient from twisted.web import client CSV_URL = 'http://summit.ubuntu.com/uds-p.csv' CSV_REFRESH = 300 OVERLAP = 300 SCHED_INTERVAL = 1 URLS = { 'Antigua 1': 'http://icecast.ubuntu.com:8000/antigua1.ogg', 'Antigua 2': 'http://icecast.ubuntu.com:8000/antigua2.ogg', 'Antigua 3': 'http://icecast.ubuntu.com:8000/antigua3.ogg', 'Antigua 4': 'http://icecast.ubuntu.com:8000/antigua4.ogg', 'Bonaire 1': 'http://icecast.ubuntu.com:8000/bonaire1.ogg', 'Bonaire 2': 'http://icecast.ubuntu.com:8000/bonaire2.ogg', 'Bonaire 3': 'http://icecast.ubuntu.com:8000/bonaire3.ogg', 'Bonaire 4': 'http://icecast.ubuntu.com:8000/bonaire4.ogg', 'Bonaire 5': 'http://icecast.ubuntu.com:8000/bonaire5.ogg', 'Bonaire 6': 'http://icecast.ubuntu.com:8000/bonaire6.ogg', 'Bonaire 7': 'http://icecast.ubuntu.com:8000/bonaire7.ogg', 'Bonaire 8': 'http://icecast.ubuntu.com:8000/bonaire8.ogg', 'Curacao 1': 'http://icecast.ubuntu.com:8000/curacao1.ogg', 'Curacao 2': 'http://icecast.ubuntu.com:8000/curacao2.ogg', 'Curacao 3': 'http://icecast.ubuntu.com:8000/curacao3.ogg', 'Curacao 4': 'http://icecast.ubuntu.com:8000/curacao4.ogg', 'Curacao 5': 'http://icecast.ubuntu.com:8000/curacao5.ogg', 'Curacao 6': 'http://icecast.ubuntu.com:8000/curacao6.ogg', 'Curacao 7': 'http://icecast.ubuntu.com:8000/curacao7.ogg', 'Curacao 8': 'http://icecast.ubuntu.com:8000/curacao8.ogg', 'Grand Sierra D': 'http://icecast.ubuntu.com:8000/grandsierra-d.ogg', 'Grand Sierra F': 'http://icecast.ubuntu.com:8000/grandsierra-f.ogg', 'Grand Sierra G': 'http://icecast.ubuntu.com:8000/grandsierra-g.ogg', 'Grand Sierra H': 'http://icecast.ubuntu.com:8000/grandsierra-h.ogg', 'Grand Sierra I': 'http://icecast.ubuntu.com:8000/grandsierra-i.ogg', } EXT = 'ogg' # Debugging :) if False: CSV_URL = 'http://mirrors.tumbleweed.org.za/uds-n/test.csv' URLS.update({ 'Classic Rock': 'http://ogg2.as34763.net:80/vc160.ogg', 'Absolute Radio': 'http://ogg2.as34763.net/vr160.ogg', 'Absolute 80s': 'http://ogg2.as34763.net/a8160.ogg', }) schedule = None recording = {} class StreamRecorder(ShoutcastClient): def gotMetaData(self, data): pass def gotMP3Data(self, data): self.factory.resetDelay() self.factory.outstream.write(data) #self.factory.outstream.flush() def handleResponseEnd(self): pass class StreamRecorderFactory(protocol.ReconnectingClientFactory): protocol = StreamRecorder maxDelay = 60 def __init__(self, filename, path): self.filename = filename self.path = path self.log = logging.getLogger(path) def doStart(self): for i in itertools.count(): fn = '%s.%i.%s' % (self.filename, i, EXT) if not os.path.exists(fn) or os.path.getsize(fn) < 10240: break self.outstream = file(fn, 'wb') def doStop(self): self.outstream.close() def stop(self): self.log.info('Shutting down cleanly') self.stopTrying() self.connector.disconnect() def buildProtocol(self, addr): self.log.info('Connected') p = self.protocol(self.path) p.factory = self return p def clientConnectionLost(self, connector, reason): self.log.info('Connection lost') protocol.ReconnectingClientFactory.clientConnectionLost(self, connector, reason) def start_recordings(): log = logging.getLogger('loop') log.debug('Checking schedule...') if schedule is None: return now = datetime.datetime.utcnow() for talk in schedule: start, end, room, track, name, title = talk if not (start <= now < end): continue if (room, start) in recording: continue if room not in URLS: log.error('Unknown room: %s', room) continue log.info('Recording: "%s" in %s', title, room) url = URLS[room] fn = '%s-%s' % ( start.strftime('%Y-%m-%d-%H-%M'), name or title.replace(' ', '_'), ) url = urlparse.urlparse(url) factory = StreamRecorderFactory(fn, url[2]) hostname = url[1] port = 80 if ':' in hostname: hostname, port = hostname.split(':') port = int(port) factory.connector = reactor.connectTCP(hostname, port, factory) recording[(room, start)] = factory reactor.callLater((end - now).seconds, end_stream, title, room, factory, log, start) def end_stream(title, room, factory, log, start): log.info('Stopping Recording: "%s" in %s', title, room) factory.stop() del recording[(room, start)] def refresh_schedule(): client.getPage(CSV_URL).addCallback(do_refresh_schedule) def do_refresh_schedule(data): global schedule log = logging.getLogger('loop') log.info('Updating Schedule...') temp_sched = [] reader = csv.reader(data.split('\n')) # Skip header: reader.next() for row in reader: if not row: continue start, end, room, track, name, title = row start = dateutil.parser.parse(start).replace(tzinfo=None) start -= datetime.timedelta(seconds=OVERLAP) end = dateutil.parser.parse(end).replace(tzinfo=None) end += datetime.timedelta(seconds=OVERLAP) # At UDS-o, we had 'name location' #m = re.match(r'^(.+?)\s*\d+$', room) #if m: # room = m.group(1) if title.lower().strip() == 'private meeting': continue if room not in URLS: log.warning('Unknown room: %s', room) temp_sched.append((start, end, room, track, name, title)) schedule = temp_sched log.info('Updated Schedule') class CLI(basic.LineReceiver): delimiter = os.linesep def lineReceived(self, line): log = logging.getLogger('cli') if line == 'h': log.info("CLI Help:\n[r]eload\n[n]ext start\n[c]urrent recordings") elif line == 'r': refresh_schedule() elif line == 'c': if not recording: log.info("Not currently recording") for room, start in recording.iterkeys(): log.info("Recording %s since %s", room, start.strftime('%Y-%m-%d-%H-%M')) elif line == 'n': now = datetime.datetime.utcnow() next_start = sorted(talk for talk in schedule if now < talk[0]) if len(next_start) > 0: next_start = next_start[0][0] log.info('Next recording starts in %i minutes, at %s', (next_start - now).seconds // 60, next_start.strftime('%Y-%m-%d-%H-%M')) else: log.warning("Looks like we're done") if __name__ == '__main__': logging.basicConfig(level=logging.INFO) task.LoopingCall(refresh_schedule).start(CSV_REFRESH) task.LoopingCall(start_recordings).start(SCHED_INTERVAL) stdio.StandardIO(CLI()) reactor.run()