But what a happy and strange coincidence. In its new version (4.26) released yesterday (on my birthday), Docker adds support for PHP with Apache to its docker init instruction. Let's see how this translates into a practical case.
Imagine you're a web developer or you work in a web agency and you'd like to keep an eye on the sites you've developed for your clients, free of charge. Are these sites online now, or are they down?
This work is carried out by site monitoring platforms. Could you install one on your computer? Of course!
For 1,000 reasons or more, you want to encrypt a file containing text. Which software should you install? Well ... none other than Docker!
By using a Docker Alpine/OpenSSL image, it's so easy to encrypt/decrypt files using OpenSSL.
Good old memories... During my studies (in 1991-1993), I was a huge fan of Turbo Pascal 7.0. It was the first language that I really learned and loves it so much. I used to spend dozens of hours behind my computer writing anything and everything.
If you good remember that time, Turbo Pascal 7.0 was used to create executables for MS-DOS.
The idea for this article came after the one written on assembly language: is it possible in 2023 to run Pascal code written 30 years earlier?
When I was young...er than today, during my studies, I was playing with the x86 Assembly language on my 386DX40 computer; this was in the years 1993-1995. The beginning of Windows 3.1 that I didn't like. If someone had asked me whether graphical interfaces would be successful, I would have said no, of course not.
I remember that I wrote a resident program (TSR for Terminate and Stay Resident) which, once loaded into memory, monitored the keyboard and recorded the keys that had been pressed in a file on the hard disk. And yes, it was a password stealer. Just for fun and, above all, for the challenge it represented.
And now, in 2023, I was wondering whether it was still possible to run .asm files on my computer. It was an excellent pretext for this article.
Remember a few years ago when you wanted to learn a new programming language such as Python? What did you need before you could even start programming your first Hello World? You had to install the language on your computer; maybe you also had to get a lot of libraries/dependencies; you had to spend time configuring your machine before you could even start writing your first script.
That's all over with the advent of Docker and the concept of containers.
This time, we'll play with Python for this article. I'm discovering Python at the same time as writing this article: at this precise moment, I've never written .py scripts.
In this post, we'll play with Docker and Java. Since there are ready-to-use Java images for Docker you don't need to install or configure anything other than Docker.
You just need to know, I've absolutely no skills in Java. Which software should be installed, how to run a script and so on. I'll just rely on a very few Docker commands and, about the installation, yeah, using Docker, it's easy: nothing to install, nothing to configure.
But what a joy it is to be able to draw diagrams by, well, just writing text. Some tools are better known than others, e.g. Mermaid.
Did you know https://diagrams.mingrammer.com/? Let's explore it using, of course, a Docker ready-to-use image.
When working with a Docker container, data can be persistent or not. Imagine you're creating a localhost website with Joomla, WordPress or any other tool (Laravel, Symfony, etc.).
You've perfectly created the various Docker files needed to run the local site, you've run the command docker compose up --detach to start the containers and now you're busy installing the site. After a few moments, your local site is up and you can start developing its functionalities.
By default, if you haven't taken any precautions, the moment you stop the container (docker compose down), you'll kill your site, i.e. by default, having not taken care to save your data (your site, your database), everything will be lost and reset the next time you run docker compose up --detach. Well... Maybe that was your wish (something totally ephemeral); maybe not;
Version updated on November 21, 2023 after a discussion with Alexandre about making his Docker image available on hub.docker.com
Based on their documentation, FrankenPHP is 3.5 faster than PHP FPM.
It is still fairly young for use on production sites, but because it's so promising, it's certainly worth playing with when developing locally.
Alexandre Elis茅 has written a script to use FrankenPHP with Joomla. You can find the source here: https://github.com/alexandreelise/frankenphp-joomla.