Printable View
Probably many of you use Hungarian notation for naming variables. But if you don't want to indicate what type of the variable it is, how do you name the variable?
Say, MyVariable, for example.
I noticed that many put the first word low case, and capitalize other words.
For example, myVariable.
Why is that? What is the most common way of naming variables and functions?
I name them to help describe what the variable is for. I only use Hungarian notation when I'm doing windows programming, and then only to identify handles and such.
For variables that I use and then reassign a lot I give them simple names. My most common ones are tmp#, ch, c#, count, etc (note # means a number not the actual symbol).
When you write a variable, you want to have it describe what the variable is. for example, if you wanted a variable to hold the value of some widgets made in 1986, you might write:Quote:
Originally posted by Ariod
I noticed that many put the first word low case, and capitalize other words.
For example, myVariable.
Why is that? What is the most common way of naming variables and functions? [/B]
numberofwidgetsmadein1986
Of course that is kind of hard to read, so you might write it with capital letters to begin each new word, like this:
numberOfWidgetsMadeIn1986
Now it is slightly easier to read. I did not capitalise the letter 'n' at the beginning of the variable, as I am lazy, and I don't like pushing shift more than I have to...actually I am lying, as you will see, I like doing more work than a lot of people when I name my variables... No sense in capitalising the first letter though, as its not needing any emphasis, as opposed to the words 'buried' within the variable, like 'widgets' and 'made.'
Another way to make your variable easier to read (and someone might scream when they see this..lol) is to use an underscore.
number_of_widgets_made_in_1986
This is my personal preference (even though it is slower to type with underscores) over against
...over against the 'camel hump' style with the capital letter for the start of each word. I just never found it easy to read camel hump style. Lots of people like it though.
One other thing, though the variable name should be quite descriptive, it should not be excessively long. A better version of my example variable name might be:
number_of_widgets_1986
In this way, you still know what it is. As far as naming functions, you might want to take more care and describe what the function does precisely - in that light, a function name will tend to be longer (if necessary - don't make a name longer than you need to, but don't make it so sparse you forget what the thing does)
Hope this helps...
kermit
I name all my variables one or two character names.
Err... no, not really, but I did when I was programming in VB. That got QUITE difficult to follow ;) a,b,c,d,e,f,g through m or so, and no comments... lol.
Keep 'em short and descriptive. I'd change Kermit's
numberOfWidgetsMadeIn1986 to numWidgetsMade or WidgetsMade. Short; descriptive.
typically variables begin with lower case letters and each additional word in the variable is capitalized as mentioned above. Words that start with an upper case letter are generally used for classes. All caps typically denotes a constant. These are commonly accepted convetions but are not rules. There are other styles.
My style is generally
myVariableName or myvar
MyFunctionName
MYDEFINENAME
breaking this convention at programmer's whim...
In all but a couple of special cases, I call my variable Variable or if two or more word SecondVariable, i.e. with each word capitalised. I shorten some words where this does not introduce ambiguity.
Exceptions are loop counters, I always start with a lower case i, then j, and so on - hang over from many years of coding in Fortran. Pointers I call pSomething, hang over from an early database system I used whereby that was a requirement, and for some reason, I call handles hSomething.
Functions follow the same rule as variables.
With todays IDE's, Hungarian is a relic, if I have forgotten what type a variable is, I put my cursor on it and the IDE tells me and offers to take me to the declaration.
Often the software house you're working for has installation standards which dictate you must use so-and-so convention for naming.
*** EDIT ***
Although I'll agree with WaltP
>>> breaking this convention at programmer's whim...
Exceptions are loop counters, I always start with a lower case i, then j, and so on - hang over from many years of coding in Fortran. Pointers I call pSomething, hang over from an early database system I used whereby that was a requirement, and for some reason, I call handles hSomething.
Just curious... did you have to call loop counters i, j, k in Fortran?
>>> did you have to call loop counters i, j, k in Fortran?
No. The thing with Fortran is that unless you specifically tell the compiler otherwise, any variable name that begins with an i,j,k,l,m,n or o is implicitly typed to be an integer, thus it is not necessary to declare them.
In fact, for production work, I always insisted all of my coders started their programs with an Implicit None statement, which turned all of the auto declarations off. Thus it was necessary to both type and size all of your variables.