asterisk-java-users — Asterisk-Java User List

Re: [Asterisk-java-users] Originate and multiple variables

[Stefan R. <sr...@re...>]

Eduard Martinez wrote:
> Another possible cause:
> I'm doing Originate to a local channel, like Local@/XXXXX@outcoming
> Originate extension then is s, context [connected]

you can try setting the variables with two leading underscores (e.g.
__callerId) this causes them to be interhited. I am unsure whether this
works with Local/ channels though.

=3DStefan

--=20
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...

[Asterisk-java-users] fail to run the example

[unplug <mai...@gm...>]

Hi,
I follow the website tutorial to run HelloAGIScript.
http://www.asteriskjava.org/latest/tutorial.html
In my directory, I have the following files.
fastagi-mapping.properties
HelloAGIScript.java
HelloAGIScript.class
I have already located asterisk-java.jar in a classpath and I run the
AGIServer as following.
java -cp . net.sf.asterisk.fastagi.DefaultAGIServer
server started

Then I make a call to 1300 (which I have followed the example to
configure the exten).

But there is an error in the console.
Jul 6, 2006 4:41:49 PM net.sf.asterisk.util.impl.JavaLoggingLog info
INFO: Received connection.
Jul 6, 2006 4:41:49 PM net.sf.asterisk.util.impl.JavaLoggingLog error
SEVERE: Resource bundle 'fastagi-mapping' is missing.
Jul 6, 2006 4:41:49 PM net.sf.asterisk.util.impl.JavaLoggingLog error
SEVERE: No script configured for URL 'agi://localhost/HelloAGIScript'
(script 'HelloAGIScript')

Anyone can tell me why AGIServer can't find fastagi-mapping.properties
as it is located in the current directory which is also a classpath?
What's wrong in the above setting?

Re: [Asterisk-java-users] Originate and multiple variables

[Eduard M. <ed...@ba...>]

Maybe, tomorrow I will try upgrading to last bussines edition (I think is=20
equivalent to 1.2)
El Jueves, 6 de Julio de 2006 10:25, Stefan Reuter escribi=F3:
> Edu wrote:
> > According to the logs, the version is Asterisk 1.0:
> >
> > 06-jul-2006 7:58:34 net.sf.asterisk.util.impl.JavaLoggingLog info
> > INFO: Determined Asterisk version: Asterisk 1.0
> >
> > Well, the production machine is not exactly Asterisk 1.0 but Asterisk
> > Bussines Edition ABE-A.1-6
> >
> > I will do more tests
>
> Is it possible that your version of Asterisk is considered to be
> Asterisk 1.0 though it behaves as 1.2 regarding the handling of
> variables in originate?
>
> =3DStefan

Re: [Asterisk-java-users] Originate and multiple variables

[Edu <jav...@ba...>]

Another possible cause:
I'm doing Originate to a local channel, like Local@/XXXXX@outcoming
Originate extension then is s, context [connected]
In extensions I have=20
[outcoming]
exten =3D> _XXXX,1,UserEvent(LocalChannel)
exten =3D> _XXXX,2,Dial(Zap/g2/${EXTEN})
exten =3D> _XXXX,3,Hangup
[connected]

exten =3D> s, 3,UserEvent(Foo| callId: ${callId})

I'm doing this things because with the LocalChannel userEvent I have the re=
al=20
name of the local channel in my java application and I can cancel the=20
originate (hanging up this channel) and when is connected, Foo tells me the=
=20
real zap channel answered.
OriginateSuccesfulEvent doesn't tell me the real channel, and with originat=
e=20
action with a Zap group channel I don't have the channel until is answered=
=20
(so I can't cancel the originate while is ringing)

El Jueves, 6 de Julio de 2006 10:05, Stefan Reuter escribi=F3:
> Edu wrote:
> > As commented in Asterisk wiki, Originate syntax and variables differ in
> > 1.0 and 1.2, so I think that there is the problem so only want to confi=
rm
> > this behaviour :)
>
> Asterisk-Java takes this into account, so it usually works with 1.0 and
> 1.2. Does Asterisk-Java recognize your production server as 1.0?
> (according to the logs)
>
> =3DStefan

Re: [Asterisk-java-users] Originate and multiple variables

[Eduard M. <ed...@ba...>]

Another possible cause:
I'm doing Originate to a local channel, like Local@/XXXXX@outcoming
Originate extension then is s, context [connected]
In extensions I have=20
[outcoming]
exten =3D> _XXXX,1,UserEvent(LocalChannel)
exten =3D> _XXXX,2,Dial(Zap/g2/${EXTEN})
exten =3D> _XXXX,3,Hangup
[connected]

exten =3D> s, 3,UserEvent(Foo| callId: ${callId})

I'm doing this things because with the LocalChannel userEvent I have the re=
al=20
name of the local channel in my java application and I can cancel the=20
originate (hanging up this channel) and when is connected, Foo tells me the=
=20
real zap channel answered.
OriginateSuccesfulEvent doesn't tell me the real channel, and with originat=
e=20
action with a Zap group channel I don't have the channel until is answered=
=20
(so I can't cancel the originate while is ringing)

El Jueves, 6 de Julio de 2006 10:05, Stefan Reuter escribi=F3:
> Edu wrote:
> > As commented in Asterisk wiki, Originate syntax and variables differ in
> > 1.0 and 1.2, so I think that there is the problem so only want to confi=
rm
> > this behaviour :)
>
> Asterisk-Java takes this into account, so it usually works with 1.0 and
> 1.2. Does Asterisk-Java recognize your production server as 1.0?
> (according to the logs)
>
> =3DStefan

Re: [Asterisk-java-users] Originate and multiple variables

[Stefan R. <sr...@re...>]

Edu wrote:
> According to the logs, the version is Asterisk 1.0:
>=20
> 06-jul-2006 7:58:34 net.sf.asterisk.util.impl.JavaLoggingLog info
> INFO: Determined Asterisk version: Asterisk 1.0
>=20
> Well, the production machine is not exactly Asterisk 1.0 but Asterisk B=
ussines=20
> Edition ABE-A.1-6
>=20
> I will do more tests

Is it possible that your version of Asterisk is considered to be
Asterisk 1.0 though it behaves as 1.2 regarding the handling of
variables in originate?

=3DStefan


--=20
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...

Re: [Asterisk-java-users] Originate and multiple variables

[Edu <jav...@ba...>]

According to the logs, the version is Asterisk 1.0:

06-jul-2006 7:58:34 net.sf.asterisk.util.impl.JavaLoggingLog info
INFO: Determined Asterisk version: Asterisk 1.0

Well, the production machine is not exactly Asterisk 1.0 but Asterisk Bussi=
nes=20
Edition ABE-A.1-6

I will do more tests

Edu

El Jueves, 6 de Julio de 2006 10:05, Stefan Reuter escribi=F3:
> Edu wrote:
> > As commented in Asterisk wiki, Originate syntax and variables differ in
> > 1.0 and 1.2, so I think that there is the problem so only want to confi=
rm
> > this behaviour :)
>
> Asterisk-Java takes this into account, so it usually works with 1.0 and
> 1.2. Does Asterisk-Java recognize your production server as 1.0?
> (according to the logs)
>
> =3DStefan

Re: [Asterisk-java-users] Originate and multiple variables

[Eduard M. <ed...@ba...>]

According to the logs, the version is Asterisk 1.0:

06-jul-2006 7:58:34 net.sf.asterisk.util.impl.JavaLoggingLog info
INFO: Determined Asterisk version: Asterisk 1.0

Well, the production machine is not exactly Asterisk 1.0 but Asterisk Bussi=
nes=20
Edition ABE-A.1-6

I will do more tests

Edu

El Jueves, 6 de Julio de 2006 10:05, Stefan Reuter escribi=F3:
> Edu wrote:
> > As commented in Asterisk wiki, Originate syntax and variables differ in
> > 1.0 and 1.2, so I think that there is the problem so only want to confi=
rm
> > this behaviour :)
>
> Asterisk-Java takes this into account, so it usually works with 1.0 and
> 1.2. Does Asterisk-Java recognize your production server as 1.0?
> (according to the logs)
>
> =3DStefan

Re: [Asterisk-java-users] Originate and multiple variables

[Stefan R. <sr...@re...>]

Edu wrote:
> As commented in Asterisk wiki, Originate syntax and variables differ in=
 1.0=20
> and 1.2, so I think that there is the problem so only want to confirm t=
his=20
> behaviour :)

Asterisk-Java takes this into account, so it usually works with 1.0 and
1.2. Does Asterisk-Java recognize your production server as 1.0?
(according to the logs)

=3DStefan

--=20
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...

[Asterisk-java-users] Originate and multiple variables

[Edu <jav...@ba...>]

I'm using Asterisk-java 0.2 and I want to put 3 variables in a originate 
action. In my devel machine with asterisk 1.2 works Ok, but in the production 
machine with asterisk 1.0 doesn't work.
Example code:

(...)
Map<String,String> variables=new LinkedHashMap<String,String>(2);
variables.put("callId",call.getIdCall());
while(i.hasNext())//Locuciones
variables.put(LOCUTIONSNAME+(++n),i.next());
originate.setVariables(variables);

And later, in extensions.conf I have an UserEvent like this:

exten => s, 3,UserEvent(Foo| callId: ${callId})

As commented in Asterisk wiki, Originate syntax and variables differ in 1.0 
and 1.2, so I think that there is the problem so only want to confirm this 
behaviour :)

Regards,
Edu

Re: [Asterisk-java-users] asterisk-java modify the sip.conf?

[Stefan R. <sr...@re...>]

Haiyin Qiu wrote:
> the user running the java program to connect asterisk server and then m=
odify=20
> the sip.conf file

No. You can use Realtime and modify the sip user database table
programmatically though.

> is there any difference between using AGI to connect the server and usi=
ng=20
> manager API?=20

FastAGI is usually used to implement IVR like applications, the Manager
API is used to get Asterisk's state, originate new calls, etc.

> I check a little the asterisk-java package, some class are=20
> some implementation of fastAGI. Is it mean that if I program a java usi=
ng=20
> the manager API to connect the server, then I can't use those classes?

that's correct.

> Does asterisk-java package only support the command of AGI and manager =
AGI?

Yes, what else could do you think it could/should support?

=3DStefan

--=20
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...
Jabber:  sr...@ja...

[Asterisk-java-users] asterisk-java modify the sip.conf?

[Haiyin Q. <yan...@ho...>]

In the asterisk-java package. it there any class or method that supports the 
function like below:

the user running the java program to connect asterisk server and then modify 
the sip.conf file

is there any difference between using AGI to connect the server and using 
manager API?  I check a little the asterisk-java package, some class are 
some implementation of fastAGI. Is it mean that if I program a java using 
the manager API to connect the server, then I can't use those classes?

Does asterisk-java package only support the command of AGI and manager AGI?


Thanks

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

Re: [Asterisk-java-users] How to test a FastAGI script without any Asterisk server ?

[Stefan R. <sr...@re...>]

Hi Oliver,

Olivier wrote:
> From reading Asterisk, the Future of Telephony, I understood that an
> Asterisk server simply uses stdin, stdout and stderr files when calling=

> an AGI script :
> - Asterisk sends line of text like :
> agi_request: test.py
> agi_channel: Zap/1-1
>=20
> - Asterisk receives:
> SET EXTENTION 122
> SET VARIABLE var1 150

It's not that easy.

what goes on is more like:

AGI Tx >> agi_network_script: test.agi
AGI Tx >> agi_request: agi://endorsed/test.agi
AGI Tx >> agi_channel: SIP/1310-f620
AGI Tx >>
AGI Rx << ANSWER
AGI Tx >> 200 result=3D0
AGI Rx << STREAM FILE "vm-instructions" "1"
AGI Tx >> 200 result=3D49 endpos=3D45760
AGI Rx << HANGUP
AGI Tx >> 200 result=3D1

So whenever you send a command asterisk sends a reply, this must be
handled by your test driver.
Therefore the far easier option is to use mock objects for testing and
not dealing with I/O at all. I'll show you what this might look like for
a simple example:

You want to test this script:

import org.asteriskjava.fastagi.AgiChannel;
import org.asteriskjava.fastagi.AgiException;
import org.asteriskjava.fastagi.AgiRequest;
import org.asteriskjava.fastagi.AgiScript;

public class MyAgiScript implements AgiScript
{
    public void service(AgiRequest request, AgiChannel channel) throws
AgiException
    {
        char choice;
        choice =3D channel.getOption("welcome-please-press-1", "1");
        if (choice =3D=3D '1')
        {
            channel.streamFile("thank-you");
        }
        else
        {
            channel.streamFile("you-didnt-do-what-i-said");
            channel.hangup();
        }
    }
}

so this is the JUnit test case:


import org.asteriskjava.fastagi.AgiChannel;
import org.asteriskjava.fastagi.AgiRequest;

import junit.framework.TestCase;
import static org.easymock.EasyMock.*;

public class MyAgiScriptTest extends TestCase
{
    AgiRequest request;
    AgiChannel channel;
    MyAgiScript myAgiScript =3D new MyAgiScript();

    public void setUp()
    {
        request =3D createMock(AgiRequest.class);
        channel =3D createMock(AgiChannel.class);
    }

    public void testServiceUserPresses1() throws Exception
    {
        // we expect the script to play 'welcome-please-press-1' and the
        // user presses 1
        expect(channel.getOption("welcome-please-press-1",
"1")).andReturn('1');
        // next the script should say 'thank-you'
        channel.streamFile("thank-you");

        // replay the mock object
        replay(channel);
        myAgiScript.service(request, channel);

        // and verify all expected methods have been called
        verify(channel);
    }

    public void testServiceUserPresses2() throws Exception
    {
        // we expect the script to play 'welcome-please-press-1' and the
        // user presses 2
        expect(channel.getOption("welcome-please-press-1",
"1")).andReturn('2');
        // next the script should say 'you-didnt-do-what-i-said'...
        channel.streamFile("you-didnt-do-what-i-said");
        // and hangup
        channel.hangup();

        // replay the mock object
        replay(channel);
        myAgiScript.service(request, channel);

        // and verify all expected methods have been called
        verify(channel);
    }
}

You'll find more information about using mock objects in general and
easymock in particular on easymocks homepage at http://www.easymock.org/.=


=3DStefan

--=20
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...

[Asterisk-java-users] How to test a FastAGI script without any Asterisk server ?

[Olivier <oza...@my...>]

Hi,

(I'm new to the list and I couldn't find in the archive, anything answering
my question.
Please, forgive me if this question has already been answered)

My question is "How can I test a FastAGI script without any Asterisk server
?".

I would like to develop a CTI application.
Ultimately, this CTI application will run on a dedicated server and exhange
data with an Asterisk server and a database server on the same LAN.

For the sake of simplicity, I would like to develop this CTI application
independently from any Asterisk server :
- I will develop a dummy call server
- as long as the target CTI application receives from the dummy call server,
the same input, it should reply to the dummy server, the same output,
- I will just have to substitute the dummy server with a running Asterisk
one to complete the setup.

>From reading Asterisk, the Future of Telephony, I understood that an
Asterisk server simply uses stdin, stdout and stderr files when calling an
AGI script :
- Asterisk sends line of text like :
agi_request: test.py
agi_channel: Zap/1-1


- Asterisk receives:
SET EXTENTION 122
SET VARIABLE var1 150

So, from an AGI script perspective, a simple set of echo or cat commands
behaves the same as a fully fledged Asterisk :
echo "agi_request: test.py\nagi_channel: Zap/1-1\n\n" | myagiscript     will
print "SET EXTENTION 122\nSET VARIABLE var1 150\n\n" on stdout

I'm searching the FastAGI equivalent.
>From Asterisk the Future of Telephony, I know FastAGI uses TCP instead of
stdin, stdout.

Which program shall I use to simply output AGI data from a server through
TCP ?
Something like :
mycmd "agi_request: test.py\nagi_channel: Zap/1-1\n\n" | myagiscript   which
would print "SET EXTENTION 122\nSET VARIABLE var1 150\n\n" somewhere.

Regards

Re: [Asterisk-java-users] dial number then pause then dial extension

[Stefan R. <sr...@re...>]

You can probably originate to the Dial application and use its D option:

    D([called][:calling]) - Send the specified DTMF strings *after* the
called
           party has answered, but before the call gets bridged. The
'called'
           DTMF string is sent to the called party, and the 'calling' DTM=
F
           string is sent to the calling party. Both parameters can be us=
ed
           alone.

=3DStefan

--=20
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...
Jabber:  sr...@ja...

[Asterisk-java-users] dial number then pause then dial extension

[Forte, G. \(GFORTE\) <gf...@ar...>]

Hello all,
I was wondering if there was a way to do the following:
=20
=20
I currently use the following method to join a conference
=20
[code]
{
userID =3D "8505551212";
    OriginateAction oAction;
    ManagerResponse mResponse;
    oAction =3D new OriginateAction();
    oAction.setApplication("MeetMe");
    oAction.setData(confID);
    oAction.setChannel(userID);
    oAction.setCallerId(callerID);
    oAction.setTimeout(new Long(30000));
    oAction.setAsync(Boolean.TRUE);
    if(! managerConnection.isConnected())
    {
        managerConnection.login();
    }
        mResponse =3D managerConnection.sendAction(oAction,30000);
        managerConnection.logoff(); =20
  }
[/code]
=20
This works fine.
Now I have a device that interfaces with radios, it has a 9 digit number =
8505551211. Dial this number and you get the option to dial an extension =
100,101,102,103,etc.. for each radio. So if you dial the extension you =
are connected to a radio.
=20
I wanted to be able to use the method above but also dial the extension =
so I need to dial 8505551212 wait N seconds and then dial the extension =
101.
=20
My question is: is the setExten() method designed for this?
If so how is this accomplished?
If not is there a way to do this in *-J?
If not I guess I can do this in the asterisk conf files with a extension =
or something.
=20
Thanks!

Re: [Asterisk-java-users] Problems releasing connections

[Stefan R. <sr...@re...>]

Hi Andrew,

yes there is an issue with the reconnect logic that sometimes triggers
the behavior you described.
It is fixed in trunk and will be fixed in 0.3.

=3DStefan

--=20
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...
Jabber:  sr...@ja...

[Asterisk-java-users] Problems releasing connections

[Andrew W. <an...@ji...>]

Stefan,

It looks like we be having problems disconnecting from asterisk in  
Asterisk-IM. The other day our wildfire instance crashed do to the  
following:

"ManagerReader" prio=1 tid=0x085f2ae0 nid=0x1888 waiting for monitor  
entry [0x43deb000..0x43deb7f0]
    at net.sf.asterisk.manager.DefaultManagerConnection.disconnect 
(DefaultManagerConnection.java:582)
    - waiting to lock <0x72549e00> (a  
net.sf.asterisk.manager.DefaultManagerConnection)
    at net.sf.asterisk.manager.DefaultManagerConnection.reconnect 
(DefaultManagerConnection.java:1044)
    at net.sf.asterisk.manager.DefaultManagerConnection.dispatchEvent 
(DefaultManagerConnection.java:1001)
    at net.sf.asterisk.manager.impl.ManagerReaderImpl.run 
(ManagerReaderImpl.java:282)
    at java.lang.Thread.run(Thread.java:595)


Asterisk-IM had more than 1K of threads at this spot, probably do to  
the plugin being reloaded. Do you have any idea why the connection  
isn't closing and the lock being released?

thanks,

Andrew

[Asterisk-java-users] CommandResponse result problem

[<mar...@my...>]

Hi there!
When I sent more than one CommandAction, every List I recently got from 
getResult() will "point" to a last Result. 
That's a problem since it is not documented and can cause strange 
behaviour of some applications.
I send a simple patch I did.

Whoops.  I forgot to send it last time ;)

Here it is:

Index: DefaultManagerConnection.java
===================================================================
RCS file: 
/cvsroot/asterisk-java/asterisk-java/src/java/net/sf/asterisk/manager/DefaultManagerConnection.java,v
retrieving revision 1.36
diff -u -r1.36 DefaultManagerConnection.java
--- DefaultManagerConnection.java       10 Jan 2006 23:09:11 -0000 1.36
+++ DefaultManagerConnection.java       14 Jun 2006 13:09:13 -0000
@@ -24,6 +24,7 @@
 import java.util.HashMap;
 import java.util.Iterator;
 import java.util.List;
+import java.util.LinkedList;
 import java.util.Map;
 
 import net.sf.asterisk.AsteriskVersion;
@@ -635,7 +636,10 @@
                         + action.getAction());
             }
         }
-
+        if (result.getResponse() instanceof CommandResponse) {
+          CommandResponse res = (CommandResponse) result.getResponse();
+          res.setResult(new LinkedList(res.getResult()));
+        }
         return result.getResponse();
     }
 

[Asterisk-java-users] CommandResponse result problem

[<mar...@my...>]

Hi there!
When I sent more than one CommandAction, every List I recently got from 
getResult() will "point" to a last Result. 
That's a problem since it is not documented and can cause strange 
behaviour of some applications.
I send a simple patch I did.

Re: [Asterisk-java-users] GetVarAction() problems...

[idir a. <idi...@ho...>]

Thank you for your reply Stefan.
I understand now why i can't get anything from the SIP_CODEC variable....
So please just tell me if you know how to get information about the nature 
of a channel (video/audio).
In brief, when a channel is set, i just want to know if there is video on 
this one.
For the moment i'm interested in both ways: with or without the Manager 
API...
Thank you very much dear Stefan Reuter

[Asterisk-java-users] Originate failure two calls in a row

[Tony M. <to...@am...>]

This is perhaps an asterisk question instead of an asterisk-java, but maybe
someone here will know the answer.

If I make two calls in a row to the same ZAP extension the second call fails
with a "Channel Unavailable Exception".  If I add in a delay of half a
second between calls then the call will succeed.

If the extension is a SIP extension then there is no problem.

Below is my test code (ext 100 is my ZAP phone and 101 my SIP phone):

package com.pbxworkbench;

import java.beans.PropertyChangeEvent;
import java.beans.PropertyChangeListener;
import java.util.concurrent.Semaphore;
import java.util.concurrent.TimeUnit;

import junit.framework.TestCase;

import org.asteriskjava.live.AsteriskChannel;
import org.asteriskjava.live.ChannelState;
import org.asteriskjava.live.DefaultAsteriskServer;
import org.asteriskjava.live.LiveException;
import org.asteriskjava.live.OriginateCallback;
import org.asteriskjava.manager.ManagerConnection;
import org.asteriskjava.manager.ManagerConnectionFactory;

public class AsteriskServerTest extends TestCase {
	private DefaultAsteriskServer server;

	protected void setUp() throws Exception {
		super.setUp();
		ManagerConnectionFactory factory = new
ManagerConnectionFactory(
				"192.168.2.150", "username", "password");
		ManagerConnection connection =
factory.createManagerConnection();
		server = new DefaultAsteriskServer(connection);
		server.initialize();

	}

	protected void tearDown() throws Exception {
		super.tearDown();
	}

	/*
	 * Test method for
	 * 
	 */
	public void testOriginate ()
			throws Exception {
		MyOriginateCallback callback = new MyOriginateCallback();
		server.originateToExtensionAsync(
"local/100@default", 
"default", 
"101",				
1, 30000, 
callback);
		callback.waitForCallEnd();

		Thread.sleep(200);

		server.originateToExtensionAsync(
"local/100@default", 
"default", 
"101",
1, 
30000, callback);
		callback.waitForCallEnd();
	}

	private class MyOriginateCallback implements OriginateCallback,
			PropertyChangeListener {
		Semaphore semaphore = new Semaphore(1);

		public MyOriginateCallback() {
			semaphore.drainPermits();
		}

		public void waitForCallEnd() throws Exception {
			semaphore.tryAcquire(30, TimeUnit.SECONDS);
			semaphore.drainPermits();
		}

		public void onSuccess(AsteriskChannel channel) {
			channel.addPropertyChangeListener(this);
		}

		public void onCallEnd() {
			System.out.println("call ended");
			semaphore.release();
		}

		public void onNoAnswer(AsteriskChannel channel) {
			System.out.println("no answer");
			onCallEnd();

		}

		public void onBusy(AsteriskChannel channel) {
			System.out.println("busy");
			onCallEnd();

		}

		public void onFailure(LiveException cause) {
			System.out.println("failure " + cause);
			onCallEnd();
		}

		public void propertyChange(PropertyChangeEvent evt) {
			System.out.println(evt.getPropertyName() + "=" +
evt.getNewValue());
			if (ChannelState.HUNGUP.equals(evt.getNewValue())) {
				onCallEnd();
			}

		}

	}
}

Re: [Asterisk-java-users] originateCall in AsteriskManager

[Stefan R. <sr...@re...>]

Tony Mowers wrote:
> I'm not sure if it is working as it should yet.  If I call someone, who=
 is
> using a SJPhone for example, and they hit their "ignore call" button th=
en I
> get an "onFailure" callback.  I'd expect an "onBusy".=20

I have to check this with a softphone, and probably with a sip client
that is not registered at all. It seems like this is treated the same
way as a failure to a completely invalid channel. Hopefully we can see a
difference for the two cases.

> An "onFailure" makes
> me think there is something wrong with the phone number, the trunk, or =
some
> such thing.

Yes the only reason for onFailure i am aware of is an invalid channel.

> I'd also kind of expect the onFailure to have a cause if
> possible.

Yes, I added a cause parameter (up to now this is always a
NoSuchChannelException).


> The property change listener also works for me. It seemed a little stra=
nge
> at first because I was expecting the natural listener with an onHangup =
and
> such. After I saw that there were several properties that could be obse=
rved
> I understood the reason for your choosing the property change listener.=
  I
> think most other people will understand too.

ok. plus its always the same for observing channels, meetme users and so
on. will need a good example ;)

=3DStefan

--=20
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...

Re: [Asterisk-java-users] Asterisk Live

[Tony M. <to...@am...>]

This change makes sense to me.

-----Original Message-----
From: ast...@li...
[mailto:ast...@li...] On Behalf Of
Stefan Reuter
Sent: Friday, June 09, 2006 4:12 AM
To: ast...@li...
Subject: Re: [Asterisk-java-users] Asterisk Live

Stefan Reuter wrote:
> Tony Mowers wrote:
>> I am guessing that in theory it would now be possible for you to throw an
>> Exception with the origination failure cause contained in the exception.
In
>> that case you'd never have to return AsteriskChannel = null.
>>
>> Is that the plan?
> 
> Yes that might be done. - Probably a good idea!

I changed it to throw a NoSuchChannelException instead of returning
null. A ManagerCommunictionException indicates a lower level problem,
i.e. something is wrong with the connection to the Manager API like not
being connected.

=Stefan

-- 
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...
Jabber:  sr...@ja...

Re: [Asterisk-java-users] originateCall in AsteriskManager

[Tony M. <to...@am...>]

The new async OriginateCallback method will work for me.  

I'm not sure if it is working as it should yet.  If I call someone, who is
using a SJPhone for example, and they hit their "ignore call" button then I
get an "onFailure" callback.  I'd expect an "onBusy".  An "onFailure" makes
me think there is something wrong with the phone number, the trunk, or some
such thing.  I'd also kind of expect the onFailure to have a cause if
possible.

The property change listener also works for me. It seemed a little strange
at first because I was expecting the natural listener with an onHangup and
such. After I saw that there were several properties that could be observed
I understood the reason for your choosing the property change listener.  I
think most other people will understand too.



-----Original Message-----
From: ast...@li...
[mailto:ast...@li...] On Behalf Of
Stefan Reuter
Sent: Friday, June 09, 2006 4:28 AM
To: ast...@li...
Subject: Re: [Asterisk-java-users] originateCall in AsteriskManager

> The types of things I'd like to watch for are:
> - Originating channel was created (I didn't screw up with channel data)
> - Origination information was incorrect and call ended
> - Originating channel was congested and so call aborted
> - Alerting is happening on originating channel
> - Origination alerting timed out
> - Originator has answered
> - Another channel has been linked to call
> - A channel in the call is being alerted
> - A channel in the call has answered
> - One of the channels has hung-up or failed
> - Call has ended

With the new async originate methods, do you think things are better now?
Do you like the OriginateCallback and the PropertyChangeListener or does
it feel strange to register a PropertyChangeListener and you would
expect to have a more "natural" listener for channel changes with
methods like onNewState or even onHangup and so on?

=Stefan

-- 
reuter network consulting
Neusser Str. 110
50760 Koeln
Germany
Telefon: +49 221 1305699-0
Telefax: +49 221 1305699-90
E-Mail:  sr...@re...
Jabber:  sr...@ja...