#Development Environment

How to develop an interoperability solution in one code base and then use it to generate many individual interfaces

Overview

I wrote a step by step tutorial in the qewd-howtos repository how you can write state of the art multi-page web apps with Node.js using a QEWD-Up WebSocket/REST api back-end integrated with a mainstream web framework like NuxtJS & Vue.js.

The latest release of Serenji focuses on the debug experience. Not only can you now debug on the spot, no matter where you are or what you're doing, but we've also added some smaller features that will make debugging more straightforward. 

• Run and Debug CodeLenses - clickable links above each class method, procedure, subroutine or extrinsic function.
• Intuitive prompting for entrypoints and arguments - when debugging a class method, produce, subroutine or extrinsic function. 
• Shaded read-only background - to clearly differentiate between documents which are editable and those that are read-only. 
• Program output in debug console - output is shown in the debug console. 
• Serenji commands on Server Manager's namespace trees - allowing you to launch Serenji through Server Manager. 

Find out more in our release notes here. 

The most recent release of Serenji features our innovative gj::locate technology. It was a standalone tool we originally created for a Developer Community contest earlier this year, but we've incorporated it into our debugger after some great feedback from developers. 

It works by navigating you directly to the source of your server-side errors in just a couple of clicks - enabling you to quickly fix errors without the need to count tedious lines of code... and let's be real, who has got time for that when you're under pressure to fix this bug? 

It's simple and straightforward to use:

1. Click on the gj::locate panel in the status bar

2. Enter the ObjectScript error message or line reference from a class/.mac routine..

3. gj::locate then does the work for you by taking you straight to the corresponding line in your source code.

Easy peasy... and with time to spare to make yourself a coffee before your deadline! 

The video below shows it in action - let us know if you've already given it a go. Or, if you're interested in trying it out we offer a free 30 day trial license, just drop me a message through the Developer Community or email us at info@georgejames.com.

The latest release of Serenji by George James Software introduces zero configuration debugging, so it's ready to go in just one click - no matter where you are or what you're doing, your on-the-spot debugger is ready to go. 

We have introduced a host of new features to enhance the debugging experience so you can focus on identifying and fixing errors, without losing focus by spending time setting up a launch configuration. 

Check out the new features in our release notes. 

If you've already started using Serenji 3.2.0 let us know how you're getting on! 

Hi developers,

We have just published an update to the Deltanji version control extension for Visual Studio Code.

When used in certain Deltanji workflows, triggering an automatic checkout on first keystroke could result in the developer working on an outdated copy of the code and only being notified of this later when saving their changes. In order to prevent this from happening checkout now reloads the document immediately.

Error messaging has been improved in two areas - when attempting to cancel code that is not checked out, and when trying to register new code onto a withdraw system.

We have also improved interoperability between the Deltanji extension and the InterSystems ObjectScript one, as well as its handling of multi-component objects.

If you're already a user of Deltanji then your extension with automatically update. You can find out more information on Deltanji and the latest release here.

I have been working within Objectscript for a few months and I am seeking assistance today because all of the help files within the docs.intersystems.com domain are currently down and they wont be back up until next week. (I raised a special ticket for this)

gj :: locate was the entry from George James Software for the Developer Tools competition back in April 2021. When an ObjectScript error is reported, gj :: locate takes the .int line location and navigates you directly to the line in the source file where the error originated. Over the past few months we've found it a very useful tool when it comes to debugging, especially when used alongside the Serenji debugger in Visual Studio Code.

Our first update, version 1.1.0, now includes support for multi-root workspaces. This enables gj :: locate to be used with every workspace folder. If your VS Code workspace has more than one root level folder, connecting to different namespaces or perhaps different servers, then gj :: locate will determine which workspace folder is current and locate the source code in that folder. If your workspace doesn't have any files open then gj :: locate will prompt you to select a folder from a pick-list.

This will particularly benefit VS Code power users who have workspaces with simultaneous connections to multiple namespaces, on multiple different IRIS and Caché servers.

If you would like to try it out then you can download it from the VS Code marketplace. Or if you’re already a user, then gj :: locate will automatically update. Let us know how you're getting on with it!

This time I want to talk about something not specific to InterSystems IRIS, but that I think is important if you want to work with Docker and your server at work is a PC or laptop with Windows 10 Pro or Enterprise.

As you likely know, containers technology comes basically from Linux world and, nowadays, is on Linux hosts were it shows maximum potential. Those who use Windows on a normal basis see that both, Microsoft and Docker, have done important efforts during these last years that allow us to run containers based on Linux images on our Windows system in a really easy way... but it's something not supported for production systems and, this is the big problem, is not reliable if we want to keep persistent data outside of containers, in the host system,... mostly due to the big differences between Windows and Linux file systems. In the end, Docker for Windows itself uses a small linux virtual machine (MobiLinux) to run the containers... it does it transparently for the windows user... and it works perfectly well if, as I said, you don't require that your databases survive longer than the container...

Well,...let's get to the point,... the point is that many times, to avoid issues and simplify, we need a full Linux system and, if our server is based on Windows, the only way of having it is through a virtual machine. At least till WSL2 in Windows is released, but that will be another story and sure it'll take a bit of time to become robust enough.

In this article, I'll tell you, step by step, how to install an environment where you'll be able to work, if you need it, with Docker containers on an Ubuntu system in your Windows server. Let's go...

Why SSH ?

If you do not have direct access to the server that runs your IRIS Docker container
you still may require access to the container outside "iris session" or "WebTerminal".
With an SSH terminal (PuTTY, KiTTY,.. ) you get access inside Docker, and then, depending
on your needs you run "iris session iris" or display/manipulate files directly.

Note:

Let me introduce the support of InterSystems IRIS in IntelliJ IDEA. This plugin adds Syntax Highlighting support for ObjectScript, and auto import and compile on the server after saving a changed file. It uses LanguageServer written in Rust, where was added an ability to import and compile code.

You may think it isn’t too difficult to get from label+offset^routine to the actual source line responsible for the error. For an expert it isn't that hard... most of the time. But there are enough oddities and special rules that even an expert can get it wrong, whilst spending a lot of time trying to get there.

gj :: locate is the latest tool from George James Software – it debugs any error, class or routine by converting the location of an error in compiled .int code to the corresponding location in your source, and then taking you right there.

Image this scenario…

We released a new version of ZPM (Package Manager)

New in ZPM 0.2.14 release:

• Publishing timeout
• Embedded vars usage in module parameters
• Package installation from Github repo
• Transaction support for install, load and publish.

See the details below.

In Studio you could open a class directly via it's name, without having to traverse the package tree with multiple clicks until arriving at the desired class.

You would Ctrl + O or (File -> Open) and be able to simply type in the class name, for example:

You press Enter, and viola - the class is opened.

How do you achieve this in VSCode?