RE: Does the recommendation against variable names starting with SQL apply to constants?

-----Message d'origine-----
De : rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx] De la part de Charles Wilt

Bad idea IMHO, you should have either taken the time to
change all references or used the existing but poorly named one.

Charles, normally I would have, but then I didn't tell you about the other constants in the source, like

D cD1 C 'D1'
D cP2 C 'P2'

and all the others. What I feel really needed to be done was to enlighten the guy who wrote it. But I don't have the freedom to do so.

Having two names for the same thing makes things more
confusing for the next guy, not less. You've went from one
WTF to two WTFs.

I don't agree. Now I have

IF MyProgrammer = cXO;

and then,

IF MyProgrammer = CantankerousOldGit;


Surely that's still only 1 WTF!!

WDSCi makes it easy to see every line with a X0, IIRC so does
SEU. It shouldn't have been hard to change them.

If you're doing proper regression testing, it shouldn't
matter if you changed one line or fifty.

Even if you're not doing good regression testing, if you
can't change the name of a constant and feel good about your
changes you're probably in the wrong line of work.

That's a bit harsh. As I already said, the rules don't always let me proceed as I would like to do.