This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
Nmeri17 said I Used to scribble my thoughts on paper. It's haphazard yet handy. And even though I can't make corrections without crossing out or drawing arrows to transfer the reader to continuation of the thought on another page, I have this liberty to express myself and glance at a panoramic view before putting them in their final resting place –soft copy
Maybe my thought process became more efficient but I no longer need to flesh things out with ink. Database designs, implementation logic. Everything goes to a special file I create on every project for odds and ends
Until today
I have something to think about. I will miss connecting the dots if they appear in fancy fonts. I need to gradually build upon each outline, pursuing it in an exploratory manner until its possibilities are exhausted. I will draw a conclusion from their character arcs
For some reason, I see parallels between this scenario and sql vs nosql. This is one of those extreme cases where structured data storage is not sufficient. I sincerely doubt nosql should be used as a main database, but instead an intermediary for an aggregator to treat each row/record as a unique blob, extract necessary information into a sql for the actual system to work with
Sql is more sane and recommended for when you know the exact end goal but need help arriving there. Today, I'm confused and need to weigh options. I need to actually cross things out, not press the back button. It's a bit of a stretch but if this were data, it feels like what nosql would excel at,Those jsfiddle/codepen snippets on stackoverflow back in the day used to replicate the same traumatic effect as the threat on injection at the hospital when I was a child,A few days back, I was battling to setup a spring app the old fashioned way. By that, I mean the way I'm used to securing routes in laravel and indicating model relationships. For the most part, I'd gotten used to the authentication config more than my first experience with it
But the jpa/hibernate relationship mess? Honestly, it's insane. If I can only access the relationship if I let them be defined eagerly (which stupidly breaks due to circular dependencies during deserialization btw), what's the point even bothering to define them there? Might as well pretend the relationship doesn't exist and strictly use IDs on belongsTo. That's it. At the time, it felt like Very sick, amateurish stuff but I'm calmer now so it's just disappointing. How come the framework is so popular??
They haven't figured out migrations, lazy and eager load, hydrating models while evading the circular dependency trap, etc. These seem so rudimentary in eloquent, I don't even have to think about it,I fucking hate spring boot. I can go unwind and nobody would hear this but boy, does it suck. Every single thing about it is a pita. I spend 98% of time I should have used for feature implementation in JAVA, struggling and battling MUNDANE functionality of the framework that ought to be nobrainers
Today started out with a project I inherited. I don't even know whether to blame its original author but he installed a couple of funny libraries for logging. The spring app doesn't build yet gradle build completes successfully. No errors are logged to the terminal, just reams and bundles of json. WHY IS THE APP NOT BUILDING??? You want me wrangling json through that pinhole console window?
I struggle with the yml settings, none works. Eventually get rid of the package (hint, it wasn't the slf4j one). I'm able to debug app not starting now, but now live reload doesn't work
I copy configs from a previous project where it worked into this. Nope, doesn't budge. Eventually enable an ide setting but now server restarts twice after file changes. The implication is that request argument annotation no longer works! So the server just restarts and has amnesia about argument type resolution
I've been sitting here for hours, without implementing a single new feature. Everything is a painstaking, avoidable aggravation VS the "framework". Never seen anything as horrendous. No line of java code yet. I just want to send a request, retrieve parameters and verify live reload is "up and running". You'd think something as low level as this shouldn't take more than two minutes. Alas, welcome to the incredible world of spring development,It's shockingly bad, Jesus 😣
https://stackoverflow.com/questions...
I know I'm just whining atp instead of moving on something else. For lack of anything else, I blame capitalism for limiting my options. I would never subject myself to this torture of my own volition. Yet, you have to wonder how a joke like this. Even a routing library developed over a code kata or aboard a flight would at least, include a regex capturing possibility of trailing slashes. How would one ever imagine that all clients of this api have to strictly either end their url with slashes, otherwise they get slammed with a 404
Unbelievable,Disclaimer: This joke will be lost in translation but I can't help but make it
In my country, you can use the phrase "it's not beans" to describe something difficult eg "that ielts exam no be beans". Given my recent woes with spring boot, I can also say "spring webflux no be beans". But it's funny to me because the framework calls objects bound to the container "beans" so it's a nerdy, don't leave me joke,That time you would have used to test that code in postman, bravely muster the werewithal to write automated tests instead. It's a onetime investment that keeps malfunction in check until code is altered
I acknowledge the fact that it's not always possible. You may have gotten thrown in headfirst into unfamiliar territory ie tech stack, or inherit a monolith where no tests were pioneered. Or you may be strongly constrained for time. But in events that you can, it's worthwhile
Whether automated or manual, Testing your work the least professional thing to do before handover. Might as well swallow the bitter pill of avoiding the gui shortcut, and write those certifications once and get it over with
My preference is to write a boilerplate that gets generated each time I create a new module/resource management classes. Another strategy is to write them immediately after completing implementation of each endpoint/user story/feature, even if they're not run immediately. That way, they don't pile up in the end
Or you could try the tdd that everyone else cherishes. Whatever works for you, the end justifies the means,I would batter this nigga if I could
My database migrations are failing after setting a 3 character max length on an integer field with default value 200. I attempt to investigate why and after numerous back and forth, this dude reveals 200 is in fact greater than 998```