My last topic was resident control add-ins. In my video and written blog I’ve explained what they are and how you can quickly have them up and running. However, there was one particular thing I said I generally don’t like – single-instance codeunits. So my second blog on this topic focuses precisely on that: how to make your control add-in available to your AL code from everywhere, without having to rely on a single-instance codeunit.
One of the most common questions I get asked about Control Add-ins is whether you can make a control add-in be always present and able to respond to your calls from AL. In other words: can you have a resident control add-in that you can invoke from anywhere in your AL code.
I’ve done a lot about control add-ins. However, I’ve never done resident control add-ins for real; all I know about them is pure theory and then some playing I did at various points of time. Still, this is an interesting topic that I wanted to address some way or other.
So, yesterday I did a live session about it.
If you missed my From JavaScript to AL and back: SkipIfBusy live session on January 8, or if you simply prefer to read rather than watch, then this post is for you. As promised, I’ll follow up in writing on all my video blogs, so here we go.
Thanks to everyone who watched my live stream today! The audience wasn’t big, but it’s a very narrow topic, not of broad interest. Still, I am glad I got a few interesting questions that actually introduced my future topics quite nicely.
You can find the recording here:
As promised, I have published the GetImageResource.js file on my GitHub. You can find it here:
https://github.com/vjekob/RandomBits/tree/main/src/js
(There you can also find some usage notes and explanation why you may want to use that file in your control add-ins)
Once again – thanks, and see you next week in a topic that should draw a lot more audience: AL interfaces!
I’ve been absent for a while – sometimes it’s difficult to find time for blogging even when I want it. Anyway… here’s the last post in the introductory series of posts that explain my NAV TechDays 2019 demos. In this code example, I explain the difference between typical development and production build configurations, and I show how to configure gulp accordingly.
To check the code example, and to read the detailed explanations of it, jump over to https://github.com/vjekob/supercharged_01/tree/08-gulp-build-prod
After a short absence, I am back. The next lesson in the Control Add-ins Supercharged series I am introducing Babel.
“Babel is a JavaScript compiler”, says babeljs.io. So, what’s the big deal, you may ask? The big deal is that Babel will take your super-fancy high-end future-generation JavaScript code that barely any browser supports, and turn it into something that even Internet Explorer can run. In theory, at least. In practice, it’s a bit more complicated.
To learn about theory and practice of Babel in control add-in development workflows, check this branch: https://github.com/vjekob/supercharged_01/tree/07-gulp-babel
And that’s it. The Readme.md file contains all the theory and practice you need. So, navigate to my github and check it out.
So far I’ve shown how to configure gulp to help you with some basic JavaScript development tasks, such as bundling and minifying, and how to automate builds on every source file change. Let’s now take it to an entirely new level: debugging.
If you ever tried to debug your control add-in source code using Visual Studio Code, you might have realized two things: first, that it’s actually pretty simple to do; and second, that it’s not the kind of debugging that you want.
Let’s see why that is.
First, using VS Code with a browser such as Chrome, Edge (especially “Canary”, mmmm) or Firefox is easy. There are debugging extensions built for that purpose that allow you to either launch a browser with VS Code attached as a debugger, or to attach VS Code to a running instance of a browser (some browser configuration needed beforehand). However, when you use VS Code as a debugger like this, it’s hardly more than a glorified browser built-in developer tool – the only benefit of using VS Code is that, well, you are in VS Code, rather than inside the browser. However, you are not really debugging your source files – you are debugging scripts currently loaded in the browser. You can’t set debug breakpoints in your source files – you can only set them in copies of files that are loaded by the browser. If you catch a bug (kind of a point with a debugger) you can’t stop the process and immediately change the line where the bug is, but you have to first locate the matching source script, and then the offending line, and then fix it there. Not exactly a fun pastime.
And that’s annoying enough if you have a single script to work with, but if you have several scripts, bundled together, then minified, well, good luck with that. Debugging minified code is as close as you can get to inquisition-level torture in this century.
So, let’s fix this mess, and let’s enable debugging as it should be – let’s enable debugging of actual source files with full breakpoint integration and all.
That’s what my next branch is about. In it, I introduce source maps and how you can use gulp to build them. And then I explain how you can configure your VS Code to debug your source files.
So, head over to https://github.com/vjekob/supercharged_01/tree/06-gulp-debug and check out some pretty cool stuff.
Yesterday you’ve seen how to use gulp to make it do something truly useful: bundling your JavaScript and CSS files. Introducing gulp for the purpose of bundling your files allows you to split your functionality into as many files as you want, but use only a single script and a single stylesheet file from your control add-in. That way you not only optimized how your browser loads the files, but you also eliminated the need to maintain your controladdin object file every single time a new JavaScript or CSS file is added (or an existing one is removed).
However, you still need to run gulp build every single time you want to build your bundles. If you forget that, your controladdin will not be updated with your changes.
Thankfully, gulp has an answer to this: the gulp.watch feature. That’s the topic of my next demo, that I’ve explained in the 05-gulp-watch branch of my supercharged-01 repository.
The Readme.md file of that branch explains how to configure the watch task in the gulpfile.js file, and how to configure VS Code to automatically run the watch task every time you open the workspace. Once you follow those instructions, you can work on individual JavaScript or CSS files without the need to run gulp build or to maintain your controladdin object file. The former is taken care of by gulp, and the latter is taken care by… well, gulp too. Gulp is cool, really 😎
My last demo introduces gulp, an amazingly simple task automation tool. You’ve seen what it is, how to set it up, and how to create a simple hello-world task. Now it’s time to teach gulp some useful tricks and make it automate the “build” process for the JavaScript and CSS files.
The first demo in this series uses a single-file approach, and there I stated that it’s a bad thing to develop like that. It’s bad to put all the functionality in one single file. However, that rule applies for development, not for runtime. Your browser actually prefers using a smaller number of JavaScript (or CSS) files. That’s why typical runtime JavaScript libraries are bundled. On top of bundling, you’ll also want to minify your JavaScript and CSS files to make them as small as possible, to further improve the download and initialization speed.
These are the things that today’s demo uses. So, head over to the https://github.com/vjekob/supercharged_01/tree/04-gulp-useful branch, and check out how to configure your gulp to build your JavaScript and CSS bundles from your source files.
Get ready for the next big step in taking the control add-in development to the next level. Today, I am (re)introducing gulp.
I’ve blogged about gulp in the past, but I’ve never really finished that series. Life and work got in the way. But while I may not have blogged about it much, it’s been my companion for quite some time now, and now I want it to become yours, too.
In previous “Control Add-ins Supercharged” examples we’ve mostly explored how not to do things. Now we are ready to take a look in the opposite direction and start talking about how to do do things.
For this example, I’ll use my next demo branch 04-gulp-hello-world that you can check out here: https://github.com/vjekob/supercharged_01/tree/03-gulp-hello-world.
I don’t intend to copy/paste all the theory explained in the branch’s Readme file, but I’ll just say what you can find there:
And with this small example that introduces gulp I conclude the introduction into the “Control Add-ins Supercharged” series. All the next steps will deliver ready-to-use stuff that you can plug into your own control add-in development process, and we’ll start building our toolchain step by step.
Have a great weekend!