I has been zero days since I replied to AI code review claiming that it's claims are complete nonsense.

#programming

Levels of evil for apps:
1. App that should be a website
2. App that is a website under the hood
3. App that is being updated weekly, even though it is a website under the hood

#programming #software

Is it a new trend to provide curl command piped to shell as installation instruction?
What could possibly go wrong?

IDE creators should get some developers filmed working and analyze whether their feature are accelerating them or on the contrary.
I'm so annoyed by IDEs desperately trying to do things for me and failing at such a rate, that me having to undo the things negates quite a good portion of gains made when assistance was successful.

#programming

Never before have I though of this.
In C/C++ signature for main is:
int main(int argc, char **argv);
(yeah, there is also env version).

Note that argv is pointer to pointer where neither pointer is const. You can modify the thing!
Tried using clang on Mac, modifying argv[0] or argv[0][0] works.

#programming #Cpp

Initially impressed by AI code reviews I am lately more and more dissapointed by them. Had multiple reviews lately that quite well demonstrate a thing that is known: these tools can't think.
One review was terrible in that it looked quite detailed description of a problem, but reading and applying logic makes it obvious there is no problem, pure contradiction in the claim.
I was positive in that I can just ignore it, so ok, sometimes good suites. Still true, but deteriorating.

#programming

Apple moving to USB-C is a good thing.
Unfortunately it means suffering for me, as each time I switch device I test stuff on, I have to look which cable to use.
Change is hard...

#programming #iosDev

Oh dear, how tricky is the Gtkmm RefPtr.
Sure, the concept of ref counting is simple. But the fact that you have to use make_refptr_for_instance for your custom classes and no staight forward way of converting instace to RefPtr wrapping it lures you into traps. RefPtr constructor happily taking a pointer is one such trap.
Spent nearly 4 hours chasing a weird crash because of this.

#programming #gtk

While such thing is terrible and don't enjoy saying it, it is foremost your own fault.
Simply:
- do not put yourself in position where all your stuff can be locked in or out simultaneously
- do not keep data you don't want to lose in one place
- do not tie all your digital access to a single item (id, sso etc.)

Yes, all from one provider looks convenient. Until it fails.
https://hey.paris/posts/appleid/

#lockin #apple #BigTech

Fairphone 6 in my hands.
Would still be happy to use Pixel 6 for a couple years, but worn out usb port forced me to change before it's to late.

#Android

Monday morning: Xcode.
I've set it to remove trailing white space, because it's ugly. It does, most of the time.
But I've hit return to have extra like, indented as expected, but after save I moved cursor and shows me expectedly no white space on that empty line.
What unexpected is that git diff does not agree and show a line having indentation.
Does it have to be so damn hard? You could simply remove the space on save and force me hit tab if I want indentation, geez

#rant #programming #xcode

wait, so it is an error to not write "try" when calling function that throws, but fine to write it when calling a function that doesn't throw?

My opinion about Swift keeps deteriorating. Better than Objective-C, sure, but that's not a particularly high bar to beat.

I though the world has learned that exception handling is bad and that Java-style checked exceptions only make it worse.

#programming #SwiftLang

Marketing people should be let anywhere near the software versions!

Windows: 2000, XP, Vista, 7, 8, 8.1, 10
Unity: 5.6, 2017.1, ..., 2022.3, 6.0
iOS: 18, 26

Now one guy at Unity (me) has to change iOS min version check to also include check for non-existent 19-25 range. Hopefully it is not me in the future who'll have to deal with those skipped version somehow coming to existence.

#programming #software #softwareDevelopment

Software translation failure: on Android (vanilla, pixel phone, latest 16) draging app icon gives both remove and uninstalled options as "Pašalinti" in Lithuanian. Good thing icons are still there.
Bad, Google, bad. Worse - it used to be different (or my memory is failing).

#Android #ux

Would be nice that C# compiler handled this case:

float f = 2.5;

I get the idea of protecting against precision loss. But why do you fail to compile for values that can be determined at compile time to not have precision loss?
It's no longer 2005, it's 2025 and changing this will only break the code that does not compile (like someones test suites might have such), which IMO is fine.

#programming #CSharp

Fire UX team in time, 2025 edition.

Using Android Studio (2024.2.2) after a while.
So I have "Run" button at the top and another one with just different color in sidebar. Both tooltip "Run", only differ in shortcut.
Difference?
First one runs the project. Second one opens the Run pane at the bottom, which happens to have yet another button to run the project.
You had good old school UI two years ago!

#programming #ux #Android #rant

Used Firefox web page translation feature for the first time. Up until now it was mostly annoyance due to drop down appearing on it's own (is there a setting to turn that off?).
The German website was translated seemingly well. Of course I can't verify, but since it was just a simple read, it was good.
One thing where it didn't work that well was initial cookie prompt, where it translated text, but not text on buttons, had to guess which one does what.

#Mozilla #Firefox #translation

API docs as comments in comments in code look like a nice idea.
Until someone actually writes a comprehensive documentation in there: several paragraphs with code examples. When doc comments take most of the file and gets difficult to find the code, I'm no longer sure it's a good idea to have the two in the same place.

#programming

Absolutely terrific Google Play feature: you search for app and it is shown to you as second hit!
The first one is some sponsored app, relevant to the search. This is ad-fraud IMO, I guarantee quite a few people install that first hit, artificially boosting the app as well as Gs advertisement "success".

#Android