#Debugging

One of the most important features during application development is the ability to debug your code easily. Because of the asynchrnous nature, a standard Node.js application server works single-threaded by default. When you are developing applications using an IDE like Visual Studio Code, you can very easily debug your Node.js process:

First, download the free Visual Studio Code IDE (@code) and install it on your development machine.

We will start with basic debugging using a very small test.js file and go to the debug view:

We have a large number of processes and operations being applied to outbound HL7 messages from our PAS.

If I select a message, and open its visual trace, I see the service, as well as all processes, and operations. 

My debugging is often about a new operation, and I don't need to see all the other operations that are in functioning as desired.

I am running a ZBREAK trace on a legacy FDBMS application that uses arrays but rather than displaying the array value contents, I am seeing logs similar to the following.

Trace: ZBREAK SET LINE=Array Val at BIOPWR+6^BIOPWR

My ZBREAK trace settings are as follows: 

ZBREAK /TRACE:ON:"/somedir/trace1.log"
ZBREAK *LINE:"T"

Hi, folks!

I need to get the source code of a class installed on a remote server. I have no Studio or Atelier access to it but I can manage REST and Web terminal access.

What is the easiest way for me to make it?

Hi I'm hoping the community will once more help me out with a question!

Right I have a production that takes an attachment from an email and processes it, sending out a hl7 message at the end. This works ok.

Now I just need to take the email and send it back out to a 'backup' email address, basically giving
a) A backup incase there is a problem with the email or something else further down the line.
b) A log of the emails receieved.

So I thought the best thing to do is have an operation with an email adapter and to send that my message.

Cache Studio almost always crashes when I click rows in Call Stack view during debugging. Then Cache has to be restarted because last edited file remains locked.

Looks like I'm the only one who uses the debugger. Or is there a magic spell to make stack view working?

Cache 2016.1.1

In case of system exception (missing property, method etc) Cache Studio's debugger stops at the line throwing it. Unless the line is in try/catch block. 

Is there a way to make it break also in try/catch case?

Cache 2016.1.1

Hello 

I recently spent some time with Atelier, and tried to use it's inherited (from Eclipse) XSL transformation capabilities.

I have installed Exlipse XML Editor and Tools and Orangevolt XSLT - as it provides Saxon XSLT processor.

I have XSLT 2.0 transformation defined, that uses isc:evaluate() callback function to perform some server side code.

And here is my problem:  when I try to run XSL transformation of a file in my project, then as soon as the processor hits isc:evaluate() it fails. 

The article makes an attempt to demonstrate that Atelier is not just repeating the functionality of Caché Studio on a new IDE platform (Eclipse) but goes far beyond. Due to my personal experience, and challenges in former projects I picked first XSLT Debugging. Is it an ordinary task? Not at all. Who is doing XSLT every day? Probably none of us. Than why XSLT Debugging? Simply because there are solutions in our product portfolio which are using XSLT inside and those solutions require customization. Customizing XSLT without some sort of toolkit is more than challenging. The examples of such

Hello, I'm trying to debug web app in Atelier remotely.

In Studio I could use "Set as debug target" option for this purposes. The class launches with CSPDEBUG option in browser and everything goes fine. But I can't find how to do this in Atelier.

Can anyone help me?

Hi,

I have a process that is importing lots of data from an external file, but every now and then I come across a STORE error.

I know that the STORE error is occurring in the %Save method of the class but I know very little else and I think its something to do with a lage amounts of Related Objects.

Is there a method of finding out more information regarding this error i.e. which actual object was being saved etc etc.

Thanks

Jim

This post is meant to provide a quick possible explanation for a very perplexing problem.

Scenario:  You’ve just created your own administrative user in your 2014.1 (or later) instance of Caché.  You gave it every possible security role (including %All), so it should in theory be able to do anything within the instance.

You’ve written a very advanced routine with a break command in it for debugging:

MyTestRoutine
            set ^MyInitGlobal = 1
            write "Hello, my name is..."
            break
            write "Steve"
            quit

Hi,

Thanks a million for taking the time to read this.

I'm calling the functions $ZF(-`1, <path to EXE>) and $ZF(-2, <path to EXE>) from Cache code. The exact code I'm using in Cache Studio is given below.

/// Test the functionality of ZF(-1) and $ZF(-2), 
Class %SourceControl.UnitTest Extends %Persistent
{
Parameter PATH As %String = "E:\VCS\CommandExecutor\bin\Debug\WindowsApplication.exe";
ClassMethod Test()
{
Do $ZF(-2, ##Class(%SourceControl.UnitTest).#PATH)
}

Now when I run the command in the Cache Terminal, as given below,

>do $ZF(-2, "E:\VCS\CommandExecutor\bin\Debug\WindowsApplication.exe")

Hi all,

For custormer support reasons we would like to know if its possible to activate some kind of flag or see/redirect wich is the code (lines) being executed when some misterious problems appear.

Is there any way to view the stack of execution code on a deployed code environment? Is there any other equivalent way to track the execution stack for a certain period of time?

Best regards

Hello everyone,

I have discussed extensively with Andrew Cheshire, the author of the Atelier parser (which I'll be able to hack on and reuse for CachéQuality -- yay!), and while we were at it, we also discussed code coverage.

We ended up talking about the ZBREAK command and its /TRACE option.

It appears that combining this command and the unit test runner, we can indeed do code coverage (Andrew also told me about a way to map "back" routine lines to original lines in the source code).

Hello.

For debugging purposes I  sometimes need to display stack information to the current device or save it (to a global for example).

There is this snippet offered in documentation:

ClassMethod StackToDisplay()
{
    For loop = 0:1:$Stack(-1) {
        Write !, "Context level:", loop, ?25, "Context type: ", $Stack(loop)
        Write !, ?5, "Current place: ", $Stack(loop, "PLACE")
        Write !, ?5, "Current source: ", $Stack(loop, "MCODE")
        Write !
    }
}

And to save stack to a global there's always an exception method:

I am trying to debug a class on Atelier and it is not stopping at the defined breakpoint.

Atelier IDE Version: 1.0.107 - Cloud Connection

I have already taken a look at the Community´s Atelier Debugging Video and followed the steps without success.

Can anybody help me on that?

Tks. 

I'm writing some custom functions for use in a routing rule.  I have a few that are working, but right now I'm trying to use one that has no parameters.  Typically this would be a sub instead of a function, but I'm not familiar enough with Cache to know what I need to do here. 

Here is the code: