What's most commonly used and what do you prefer?

C-Style (this_is_my_function), Camelcase 1 (thisIsMyFunction), Camelcase 2 (ThisIsMyFunction), or other (please explain if other)?

Thanks!

I like to use lowerCamelCase for functions and vars, and CamelCase for classes/data types.

Hi, Golmschenk -

Here's a great "readme" written by Linus Torvalds himself (you know - the guy who created Linux):

Cool cool. Thanks to both of you for your input.

Quoth Mr. Torvalds:
No. No no no. No.

If you call something "i", and you use your text editor to find all occurrences of that variable, you'll also stop at every variable whose name inclues "i", and every "if", and who knows how many places in your comments.

I like to follow the rule that no variable's name should be found as part of any other name. It makes searching for all occurrences of a variable far, far easier.

Nonsense! "i" is a *perfect* name for a temporary index ... for a loop ... that's only going to span a couple of lines anyway.

If your loop is out-of-control to the point you *need* a text editor to find the index ... then you've got bigger problems than naming conventions for variables.

IMHO .. PSM

I prefer ThisIsAStruct, THIS_IS_A_CONSTANT, this_is_a_variable, and ThisIs_aPseudoMemberFunctionOf(ThisIs*).

At work, CThisIsAClass, ThisIsAFunction, THIS_IS_A_STRUCT_s, THIS_IS_AN_ENUM_e, conjAnd pnThese vAre nVariables, because our code used to be Windows-only and that's what MFC does.

I don't agree, but you have an interesting point. I believe that this question is a minor example of this.

I tend to follow the rule "When in Rome, do as the Romans do." If I'm working in a project started by someone else or as part of a team, I'll follow their naming conventions (and brace placement, and tabs vs. spaces, etc.). But on this I have to chime in.

This is a very weak excuse at best for not using "i" as a loop variable name. If you're searching occurrences of variable names, typically you're searching for global variables (or member variables, in C++) which should have longer, more descriptive names anyway. If you want to find the loops themselves, look for "for" or "while". If you want to find every occurrence of "i" within that loop code, you should be able to eyeball it. If you can't eyeball it, there is too much code in the loop and it needs to be broken up or re-written.

Hi, mjones490 -
I agree completely!

Let's face it: the whole POINT of "coding conventions" is to instill "consistency". A code base that's "consistent" is easier to read; a code base that's easier to read helps the team and helps the project. An all around good thing.

If I've got a "better idea" and start violating those conventions (position my braces differently, name my variables differently - whatever), then I'm injecting INCONSISTENCIES. Regardless of how one feels about names or braces ... doing it DIFFERENTLY ... is worse than no coding conventions at all.

IMHO .. PSM

In general, this is a good idea, and I guess in principle I'll have to agree with it. (You'll note that my dissent in this thread had to do with an entirely different issue.)

I would just point out that there have been times where I've looked at some modification to some C code and asked myself, "What's going on here?" And then I looked at the braces, the commenting style, and so forth, and thought, "Wait, that's Bryan's handwriting, I'll go ask him."

This might make you gag, and it makes me gag, too, but it's helped from time to time.