d4j-info Mailing List for Dialogic for Java
Brought to you by:
cgm8
You can subscribe to this list here.
2000 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(3) |
Nov
|
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2002 |
Jan
(5) |
Feb
|
Mar
|
Apr
|
May
|
Jun
(2) |
Jul
(4) |
Aug
|
Sep
(1) |
Oct
|
Nov
|
Dec
|
2004 |
Jan
|
Feb
(3) |
Mar
|
Apr
(2) |
May
|
Jun
(3) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Ly T. <ly...@co...> - 2004-06-16 02:19:21
|
Dear Lavasani, We have been using d4j for our IVR projects (with GCChannel and AnalogChannel) since last year and I think this package is stable enough and easy to use. Maybe there are some adjustments needed for country dependent parameters (like hangup freq and deviation, max ANI/DNIS digit buffer). Regards, Ly Tjuan. -----Original Message----- From: d4j...@li... [mailto:d4j...@li...]On Behalf Of Seyed Ehsan Lavasani Sent: Tuesday, June 15, 2004 9:59 PM To: d4j...@li... Subject: [D4J-info] Do you recomend D4J Dear all We're going to write an IVR application and we were wondering if this project is stable enough to count on and build our project on it. regards Lavasani ------------------------------------------------------- This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND _______________________________________________ D4J-info mailing list D4J...@li... https://lists.sourceforge.net/lists/listinfo/d4j-info |
From: Carlos G M. <tr...@hu...> - 2004-06-15 15:31:21
|
Hi, I'm not an unbiased observer, so I will not answer, but bear in mind that the mailing list is not very much used in this project. May be you can consider posting in the open discussion forum. Regards, -Carlos Seyed Ehsan Lavasani wrote: > Dear all > > We're going to write an IVR application and we were wondering if this project is stable enough to count on and build our project on it. > > regards > Lavasani > > > ------------------------------------------------------- > This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference > Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer > Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA > REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND > _______________________________________________ > D4J-info mailing list > D4J...@li... > https://lists.sourceforge.net/lists/listinfo/d4j-info > -- Carlos G Mendioroz <tr...@hu...> LW7 EQI Argentina |
From: Seyed E. L. <eh...@la...> - 2004-06-15 14:57:46
|
Dear all We're going to write an IVR application and we were wondering if this project is stable enough to count on and build our project on it. regards Lavasani |
From: Carlos G M. <tr...@hu...> - 2004-04-13 10:17:54
|
Tony, D4J is thought after a server model, and the "Dialogic" thread (I assume the event polling thread in windows or the work thread in linux) is needed for any activity, so it never ends. It is marked as daemon, so it will be terminated by the VM when/if there is no other non-daemon thread alive. On the other hand, this has no impact on a channel activity. Closing a channel should terminate all activity on the channel and interrupt the service thread, which is normally the one looking for incoming calls. So if you do close all open channels and terminate any other threads you may have started including the main thread, the VM should exit. It works for me :-) -Carlos Tony Seebregts wrote: > Hi, > > Is there an elegant way to shut down an analog channel that is waiting > for an incoming call (i.e. waiting for rings) ? > > channel.close() doesn't seem to terminate the Dialogic thread so the > application doesn't exit. Runtime.exit() does work but its a brute force > solution. > > Tony Seebregts > > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: IBM Linux Tutorials > Free Linux tutorial presented by Daniel Robbins, President and CEO of > GenToo technologies. Learn everything from fundamentals to system > administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click > _______________________________________________ > D4J-info mailing list > D4J...@li... > https://lists.sourceforge.net/lists/listinfo/d4j-info > -- Carlos G Mendioroz <tr...@hu...> LW7 EQI Argentina |
From: Tony S. <ts...@ia...> - 2004-04-13 08:19:12
|
Hi, Is there an elegant way to shut down an analog channel that is waiting for an incoming call (i.e. waiting for rings) ? channel.close() doesn't seem to terminate the Dialogic thread so the application doesn't exit. Runtime.exit() does work but its a brute force solution. Tony Seebregts |
From: Joe P. <joe...@in...> - 2004-02-17 15:35:34
|
On Tue, 2004-02-17 at 04:48, Carlos G Mendioroz wrote: > The list never "engaged" as the comm link, may be because the project > is (still ?) mainly developed by 1 person. > (There have been contributions though :-) > > Forums seem to fit better the communication needs. > > Matt Darnell wrote: > > Is this project still being developed? There hasn't been any postings since > > 2002. I don't know how long I've been on this list and this is the first discussion I've seen come across it. Personally, I prefer mail lists over forums but as Carlos says, maybe forums are better for the needs of the user base? -joe -- Innovation Software Group, LLC - http://www.innovationsw.com Custom Internet and Computer Solutions Linux, UNIX, Java Training |
From: Carlos G M. <tr...@ac...> - 2004-02-17 09:53:13
|
The list never "engaged" as the comm link, may be because the project is (still ?) mainly developed by 1 person. (There have been contributions though :-) Forums seem to fit better the communication needs. Matt Darnell wrote: > Is this project still being developed? There hasn't been any postings since > 2002. > > -Matt > > > > ------------------------------------------------------- > SF.Net is sponsored by: Speed Start Your Linux Apps Now. > Build and deploy apps & Web services for Linux with > a free DVD software kit from IBM. Click Now! > http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click > _______________________________________________ > D4J-info mailing list > D4J...@li... > https://lists.sourceforge.net/lists/listinfo/d4j-info > -- Carlos G Mendioroz <tr...@ac...> |
From: Matt D. <mda...@se...> - 2004-02-17 04:42:33
|
Is this project still being developed? There hasn't been any postings since 2002. -Matt |
From: berserksangr <ber...@po...> - 2002-09-06 09:16:15
|
Hi. We HAVE to use Sun's JDK. But when we tried to start dialogic under Linux (demo programs from Dialogic works fine) we got error. Looks like JVM totally abort executing right after Dialogic$WorkThread. Please help! What we should do? How can we deal with this problem? It's only few days to show off at client. Platform: Linux RH 7.2, Dialogic SR 5.1, SUN JDK 1.3.1_04, Dialogic d/4pci board. -- Best regards, berserksangr mailto:ber...@po... |
From: <viv...@he...> - 2002-07-15 18:07:02
|
------------------ Virus Warning Message (on mail.hellorainbow.com) Found virus WORM_FRETHEM.K in file decrypt-password.exe The uncleanable file decrypt-password.exe is moved to /etc/iscan/virus/virGaGjFQ. --------------------------------------------------------- |
From: Carlos G M. <tr...@hu...> - 2002-07-12 10:19:59
|
Vijay, D4J is not an implementation of JTapi. D4J implements in java the native API of Dialogic plus some extensions. The best way to understand it is by reading dialogic docs, and that's not that good :-( But that's the way it is, docs are available on-line. To get an easy start, you have to understand that you'll be dealing with the telephony hardware asynchronously. You have to open a device and then command it to do things and wait for things to be done. Only 1 thing at a time. Also, some external events can be thrown to you by the telephony layer (like incoming call, call terminated, or even digit pressed) Voice "resources", a term dialogic uses for the part of the hardware that deals with play/record/digit detection, do extensive use of a structure called TPT, wich is used to determine what conditions can terminate a play/record/ digits collection. And that is one of the forms to get a key press. (Ypu can specify you are after a particular key, or any, or some number of them, and then retrieve them). The other way is to enable "digits as events" in which case, each digit is sent to you asynchronously. This is just a small snapshot. There are more things to it, but I would expect to have a small functional program working in 2/3 days if you would like to. Regards, vijay wrote: > > Hi there, > I am developing an application using jtapi with > Dialogic card and the implementation i use is d4j. > I want to receive button presses from the caller, so > that my application could respond audio files for each > different key press. > How to detect which key the caller presses. Will any > one pls help me . > Thank u > vijay > > ________________________________________________________________________ > Want to sell your car? advertise on Yahoo Autos Classifieds. It's Free!! > visit http://in.autos.yahoo.com > > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Gadgets, caffeine, t-shirts, fun stuff. > http://thinkgeek.com/sf > _______________________________________________ > D4J-info mailing list > D4J...@li... > https://lists.sourceforge.net/lists/listinfo/d4j-info -- Carlos G Mendioroz <tr...@hu...> LW7 EQI Argentina |
From: <jvi...@ya...> - 2002-07-12 05:17:39
|
Hi there, I am developing an application using jtapi with Dialogic card and the implementation i use is d4j. I want to receive button presses from the caller, so that my application could respond audio files for each different key press. How to detect which key the caller presses. Will any one pls help me . Thank u vijay ________________________________________________________________________ Want to sell your car? advertise on Yahoo Autos Classifieds. It's Free!! visit http://in.autos.yahoo.com |
From: Carlos G M. <tr...@hu...> - 2002-06-20 11:14:47
|
Vivek, I'm afraid there's no much currently to choose from. There's a free provider called FreeJTapi at http://free-jtapi.westhawk.co.uk, that's based on an earlier version of D4J. Core JTapi should be functional. There are 2 SF projects also for generic JTapi support, but they both lack (for the time being) dialogic drivers. We are "talking" about doing them though. HTH, Vivek Mangala wrote: > > Hi, > I am interested in JTAPI implementation > that can work on D300SC-E1 Card. > Where can I find it ? > Please Inform. > Regards > Vivek -- Carlos G Mendioroz <tr...@hu...> LW7 EQI Argentina |
From: Vivek M. <viv...@he...> - 2002-06-20 07:25:32
|
Hi, I am interested in JTAPI implementation that can work on D300SC-E1 Card. Where can I find it ? Please Inform. Regards Vivek --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.370 / Virus Database: 205 - Release Date: Jun/5/2002 |
From: Carlos G M. <tr...@hu...> - 2002-01-25 21:40:57
|
Hi, I've been bitten by a former feature of D4J, namely that passing a null value of IOTT in a record will bring the whole JVM down at once, at least in linux. So now, both play() and record() check for this and throw an exception if that's the case. Fix already at the CVS repository. -- Carlos G Mendioroz <tr...@hu...> LW7 EQI Argentina |
From: Carlos G M. <tr...@hu...> - 2002-01-17 18:18:02
|
Thomas McGuire wrote: > > This question is two fold one is it possible with a dialogic card and second > is there a way in D4J. > > Can I play a G711 stream into a Dialogic timeslot and have it play out a T1 > port? Is this possible to do in full duplex? > Well, kind of. AFAIK, G711 is good old PCM 16 bits @ 8Khz sampling, i.e., standard phone coding, available in the 2 companding forms (A-law and u-law). So, can you play a stream to a timeslot ? yes. Realtime ? Not quite. D4J supports playing from InputStream. ANY InputStream, and does so via a JNI implementation of dialogic User I/O. There are 2 issues involved: 1- half-duplex vs full-duplex. each voice resource is half-duplex, so you need to dedicate 2 of them for full-duplex. 2- latency using non-DM3 cards, you can not get rid of the buffers, so you'll have some 2 secs latency added. It should be possible to cut this some, but not too much. Dialogic has the DM3 line of cards targetted to this kind of applications. Unfortunatelly, none of those have come to my hands :-( So D4J has no provision to work with DM3 cards... -- Carlos G Mendioroz <tr...@hu...> LW7 EQI Argentina |
From: Thomas M. <tcm...@us...> - 2002-01-17 17:44:25
|
This question is two fold one is it possible with a dialogic card and sec= ond is there a way in D4J. = Can I play a G711 stream into a Dialogic timeslot and have it play out a = T1 port? Is this possible to do in full duplex? ____________________________________________________________________ Get free e-mail and a permanent address at http://www.amexmail.com/?A=3D1= |
From: Carlos G M. <tr...@hu...> - 2002-01-17 11:36:59
|
Thomas McGuire wrote: > > Can you use D4J to operate D240-T1 cards in a clearchannel mode. No signalling > but just play, record and collect dtmf. In this case, you can use D4J via AnalogChannel and it should work. > Is this something you accomplish > programmatically with D4J? or do you have to configure the Dialogic card > directly with the driver tools? The DTI-DX association is done automatically by the dialogic driver initialization. If you are going to use switching functions (i.e. connect one channel to another) then there would be a need for a specific channel logic (because you need to be aware of the DTI frontend for switching). Actually I guess that a stripped down R2CASChannel may actually work... we can call it DTIClearChannel :-) No big deal. -- Carlos G Mendioroz <tr...@hu...> LW7 EQI Argentina |
From: Thomas M. <tcm...@us...> - 2002-01-17 01:10:21
|
Can you use D4J to operate D240-T1 cards in a clearchannel mode. No signa= lling but just play, record and collect dtmf. Is this something you accomplish programmatically with D4J? or do you have to configure the Dialogic card directly with the driver tools? ____________________________________________________________________ Get free e-mail and a permanent address at http://www.amexmail.com/?A=3D1= |
From: <tr...@hu...> - 2000-10-17 00:44:46
|
> So it is possible, but it could be a problem ! Yes... > > (Note: D4J is not mono threaded, but there is only one thread > > that talks to dialogic) > > Which is good, since we wouldn't have a hope on Unix otherwise. > There are a couple of (old) unix tricks we may need to employ. > > My favorite is using a pipe to decouple 2 bits of code. > > Unix Signals run on a separate stack, so the only 'safe' thing to > do in a signal handler is to set a flag and return. Well, we are not running the handler, but instead we are being called by the handler. D4J is just puting an event in a queue at signal time. (In signal callback mode, which is not the default mode the distributed code works, but it is way to easy to move to by changing a couple of lines) The problem is that in signal callback mode, the dialogic thread should be normally blocked at a wait, and it seems to not get signals in that state... One thing I want to test is to mask the dialogic handler, just by starting in signal callback mode and seting up our own SIGPOLL handler, to flag the poller thread and do waitevt(). > > Often this isn't enough, so the hack was to open a pipe, and let > the signal handler compose a (small) struct (say an event), which it > wrote to the pipe. The main program would be sitting in a select() call > which included the other end of the pipe (this is SunOs). So when > an event came in the select would wake up and process it. > This tended to work pretty well unless you had > 4k in the pipe, > when things would block, or events would be dropped, depending on > how you set up the write to the pipe. Well, it gets tough to implement some functions like getdigits() that need common buffer space between caller and called function... But nothing impossible... -tron -- Carlos G Mendioroz <tr...@hu...> +54 11 4790-5093 D.F.Sarmiento 4165 - Olivos BA 1636 // LW7 EQI Argentina |
From: Tim P. <th...@we...> - 2000-10-16 12:06:14
|
tr...@hu... wrote: > > > Hey, great news! > > What's the feeling about dialogic > > on linux? Have they done a > > decent job? > > Don't know yet. I've been only fighting to get it working, and by > working I mean just a test to answer, couple of plays and digits > as events running on two lines. That's where we are. > > I just can't seem to understand why a lot of things are done the > way they are, and I have no experience with the Unix API of > Dialogic, so we'll see. There's a guy (Johan Lithander) that says > it works ok (the dialogic driver). I have actually spent some time in the UNIX API, it was my first intro to dialogic ! > > I'm worried about the funnel thing. Any call that blocks and we > have the whole thing frozen. On the other side, I have built a > DOS IVR framework (ha, new name for old stuff ;-) that is > also monothreaded and has run OK for millions of calls. So it is possible, but it could be a problem ! The whole mono-thread thing spooked me too, I ended up writing a layer that talked to the card, and then communicated over sockets to the java, to ensure that the java never deadlocked. (I also wanted to split the application over 2 OS's so it turned out to be handy). I've still got the code, basically it just polled sr_enbhdlr( dip->chdev, EV_ANYEVT, dx_handler ); .... if( sr_waitevt( 20 ) == -1 ){ } > > E.g., I've spotted an issue with the playing of java streams. > This is done via call backs from the dialogic driver, but the > driver insists on doing the open and the first read from the > control thread (the one calling dx_play) and I don't know > why, the RandomAccessFile.read() takes forever to return > (>1s !) and this does hog the whole thing. > > (Note: D4J is not mono threaded, but there is only one thread > that talks to dialogic) Which is good, since we wouldn't have a hope on Unix otherwise. There are a couple of (old) unix tricks we may need to employ. My favorite is using a pipe to decouple 2 bits of code. Unix Signals run on a separate stack, so the only 'safe' thing to do in a signal handler is to set a flag and return. Often this isn't enough, so the hack was to open a pipe, and let the signal handler compose a (small) struct (say an event), which it wrote to the pipe. The main program would be sitting in a select() call which included the other end of the pipe (this is SunOs). So when an event came in the select would wake up and process it. This tended to work pretty well unless you had > 4k in the pipe, when things would block, or events would be dropped, depending on how you set up the write to the pipe. > > > Would I (or someone) else be able to > > do the global call stuff, or is it non-obvious? > > Was not that hard on win32, although it's not that clean either. > (No easy way to clean a messy implementation ;-) > Basically, when you get an event, you have to give control to the > GC handler that has a chance to preprocess it. > I've not looked into it. Should not be that hard, but you'll have to > get aquainted with the workings of the d4J/JNI model. > > Be my guest ! (I'll help you out) I'm busy this week, but I'll have a try at the end of the week, or early next... T. |
From: <tr...@hu...> - 2000-10-16 10:13:54
|
> Hey, great news! > What's the feeling about dialogic > on linux? Have they done a > decent job? Don't know yet. I've been only fighting to get it working, and by working I mean just a test to answer, couple of plays and digits as events running on two lines. That's where we are. I just can't seem to understand why a lot of things are done the way they are, and I have no experience with the Unix API of Dialogic, so we'll see. There's a guy (Johan Lithander) that says it works ok (the dialogic driver). I'm worried about the funnel thing. Any call that blocks and we have the whole thing frozen. On the other side, I have built a DOS IVR framework (ha, new name for old stuff ;-) that is also monothreaded and has run OK for millions of calls. E.g., I've spotted an issue with the playing of java streams. This is done via call backs from the dialogic driver, but the driver insists on doing the open and the first read from the control thread (the one calling dx_play) and I don't know why, the RandomAccessFile.read() takes forever to return (>1s !) and this does hog the whole thing. (Note: D4J is not mono threaded, but there is only one thread that talks to dialogic) > Would I (or someone) else be able to > do the global call stuff, or is it non-obvious? Was not that hard on win32, although it's not that clean either. (No easy way to clean a messy implementation ;-) Basically, when you get an event, you have to give control to the GC handler that has a chance to preprocess it. I've not looked into it. Should not be that hard, but you'll have to get aquainted with the workings of the d4J/JNI model. Be my guest ! (I'll help you out) > Tim. > http://www.westhawk.co.uk/ -tron P.S. I'm CCing this to a D4J list I've created on sourceforge, for the archives... (I guess nobody is reding that yet) -- Carlos G Mendioroz <tr...@hu...> +54 11 4790-5093 D.F.Sarmiento 4165 - Olivos BA 1636 // LW7 EQI Argentina |