Re: Re (tilly) 2: Frivolous function names

I admit I didnt take a lot of time to write my code snippet as my point was to avoid naming problems and use a flag. The choice of name for the flag variable would be the authors call. Although on second thought I wouldnt use $will_create but probably $create_on_demand

Anyway, point well taken though, in future Ill put more thought into variable names in my snippets...

:-)

UPDATE
<bad_excuse>
On second thought there is a point here actually. You were talking about externally available names, whereas $flag while poorly named is internal and thus less of an issue. This isnt to say that $flag was a good name, it is not. But there is more room for error in internal use code than external use code, especially if said variables/code are properly scoped. </bad_excuse>

Yves / DeMerphq
--
This space for rent.

Comment on Re: Re (tilly) 2: Frivolous function names Select or Download Code

Replies are listed 'Best First'.
Re (tilly) 4: Frivolous function names by tilly (Archbishop) on Dec 07, 2001 at 00:50 UTC
I call that a really bad excuse. Flipping flags in internal code is an incredibly easy and common error to make. There are some estimates suggesting that it is actually more common than off-by-one errors - but not as often caught because it is much harder to see the error. (According to legend I heard the very first Fortran compiler made this error - the flag determining whether to compile carefully or not was reversed. No idea whether that is true.) As far as I am concerned the entire question of whether summon/get is OK is much less important than getting flags reversed. As matters stand, I don't think people would accidentally call summon, and most of the time get is right. But flipping flags is a basic error that you will make again and again...	[reply]

Replies are listed 'Best First'.

Re (tilly) 4: Frivolous function names
by tilly (Archbishop) on Dec 07, 2001 at 00:50 UTC

Flipping flags in internal code is an incredibly easy and common error to make. There are some estimates suggesting that it is actually more common than off-by-one errors - but not as often caught because it is much harder to see the error. (According to legend I heard the very first Fortran compiler made this error - the flag determining whether to compile carefully or not was reversed. No idea whether that is true.)

As far as I am concerned the entire question of whether summon/get is OK is much less important than getting flags reversed. As matters stand, I don't think people would accidentally call summon, and most of the time get is right. But flipping flags is a basic error that you will make again and again...

[reply]