For testing, I often find myself prioritizing having small data sets because they are easier to comprehend. Also, the small size helps emphasize what is being tested when irrelevant data is absent.
Today, I needed a PDF for testing. Following the same principles, I sought out “the smallest (valid) PDF” and ended up on StackOverflow.
JWT seems broadly used around the internet for all kinds of services and I wanted to use it for a service of my own.
To increase testing speed while interacting with the backend, I wanted to be able to rapidly create new JWTs to see the effect of different payload structures.
As part of a project I needed to handle webhooks from Hubspot (a CRM), and since the implementation turned out nicely, I thought it would be worth blogging about.
The reason why the app ended up in Cloud Run and written ClojureScript was to avoid the hard coupling to the main application.
When running MongoDB in Docker, diagnostics logs are sent to STDOUT by default. Since MongoDB version 4.4 these logs have been in a structured JSON format, which makes it a little cumbersome to use the profiling tool in mtools.
It is impossible for any human to take in all the information available to us. As new technologies emerge, both the amount and pace at which information hits us keep increasing. With an abundance of information, how do we choose which we allow our attention to absorb and which to ignore?
I’ve been programming Clojure for several years, and yet I’ve managed to avoid protocols during all that time (I’ve also avoided macros, but that is another story). I found myself always having a colleague do the “dirty work” or some sad excuse as of why it wasn’t necessary right now. No more… this week I got my hands dirty.
This rant is about a bad habit some developers pick up and seem to have a hard time ditching again… even after gaining lots of experience.
I guess it is to be expected. After having been burned one too many times by missing error handling, in the software they work on, they become overprotective. But it often overcomplicates the code and leaves room (extra lines of code) to place “a fix”, where “the fix” does not belong. Of course, there are plenty of gray areas, murky waters and personal opinions of … exactly where to slice the cake.
Building ClojureScript Single Page Applications (SPA) on Netlify just works. The following instructions require no prior knowledge of neither Netlify nor build tools (like Shadow-cljs, WebPack etc.), but some knowledge about HTML and Git is expected.