What are the worst programming habits?

R Erasmus

Neglecting the design phase: - reverse engineering the design requirements because the design was neglected.

"Program testing can be used to show the presence of bugs, but never to show their absence." << please vote!! >>

Martin Hart Turner

;) 3) This can be a project killer, good observation. 4) There is no excuse nowadays, there are plenty of tools available to help with refactoring. 5) :mad: 6.1) Unforgivable! Good observations!

Peter Adam

2. Short variable names. But usually using my own conventions: Array processing: i for rows, j for columns, v for current value; read/written from/to file: d for data, c for counting, etc. Why? Because I try to keep 80 columns of code. C# constants are superfluous (MessageBoxDefaultButton.Button1 FTW!), no way to keep the 80 columns, so I use more descriptive variable names there.

Stefan Bogdan

Never ending methods.

Herbie Mountjoy

Re-using code blocks in different applications without checking that the pre-existing comments are relevant in the latest incarnation. I keep finding examples of this in my archive. Stupid boy...

I may not last forever but the mess I leave behind certainly will.

greldak

Nish Sivakumar wrote:

Sometimes, it is though.

not really - the comments that are there are irrelevant but the ones that are needed are missing. The user is valid so why does it need updating?

Member 9063556

Leaving Edits in the code (Edits are messages that often pop up in developmental purposes for our in-house testing) 2) Bad tabbing. Don't blame me, really. I use a different tabbing structure due to the program we use doesn't automatically tab things well.

if (Broken) then fix.this else !fix.this end-if

Rosenne

Wrong or obsolete comments.

Mel Padden

I've seen the best and worst of stuff, but I've grown so used to it I just edit it out. There are modern tools to autoformat code in any case, I think the only things I would seriously have a problem with on that list are 4 and 5. Worst coding pet peeve? Code that isn't written defensively, with a mind to robustness, or that is not fully tested.

I too dabbled in pacifism once.

SortaCore

Some code I write includes the "sneaky minus".

someFunc(300, 250 * abc, -(500-otherVar * (3+abc), 592.3f);
// ^

I generally document that, unless the logic shouldn't ever need changing.

SortaCore

pfld_ is a pointer to a fixed long double and it ranks under other variables in testing. Hence, the underscore notation.

SortaCore

class uhoh
{
bool yay;
uhoh(bool yay)
{
this.yay = yay;
}
}

... is usually the only time I want to use "this". Unless there was another "uhoh" being involved in an uhoh function. Which is probably not good practice.

Lost User

I'm not into saving keystrokes; but it does convey the same information using less symbols. For your comparison:

// Delphi style;
procedure Test()
begin
end

// C#
void Test()
{
}

Would you like to imply that we use "{" and "}" merely to save keystrokes? You cannot deny that C# is a bit more readable than COBOL. Still, feel free to state the obvious if you feel like you have to :) It's a non-discussion. Try

11 + 2 = 13
Eleven plus two is thirteen

Would we prefer the first version, just to save keystrokes? And which of the two explains the fastest what is going on?

Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^]

Lost User

PIEBALDconsult wrote:

the class is private? :confused: How does that work?

Create a new console-application. Look at the access modifier for "Program" and it's entrypoint. Both are private. Surprised? :)

Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^]

Lost User

Can you explain the reason? :) There is no good argumentation. "This" is used for the nut-cases who don't want to prefix with an underscore, and it is one of the most abused keywords, littering code without adding ANY value whatsoever.

Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^]

Stefan_Lang

Eddy Vluggen wrote:

Should I prefix each class with a complete namespace

Maybe not each, but I've found that some namespaces use way too simple names to be safe to use without! E. g. in C++ I never use using namespace std, since it clutters the global namespace with many symbols that are common enough that they may clash with just about every nontrivial application code.

GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)

Stefan_Lang

One of the common errors of OO design is making everything public. Being private by default means that this won't happen by accident, or out of sloppyness. i rather have the compiler complain that I forgot an accessor than having other programmers modify my internal state because I forgot to explicitely make them private!

GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)

Stefan_Lang

harold aptroot wrote:

- Unreachable code. Often mistaken for "defensive programming". Code that provably can't run is provably useless.

Dunno about that one - I once inserted a check that I was 100% sure couldn't possibly fail, so I inserted a message saying this shouldn't be happening, and to please contact me. Thank god, it was a beta tester eventually seeing said message, not an actual user in production code - it turned out I was wrong on my 100% assumption... :-O

GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)

Lost User

I hope it wasn't provably unreachable, that would have bad implications for the fabric of reality.

Peter Adam

Then please show me the way parametrizing a table name, or field names in a query. Of course, you can keep hacking[^].