Here we go again...

[Hanns Ackermann]

Just curious: Is there a problem with using A and A% in the first and B%()
and B% in the second part of my example, respectively? I know, DIM ALL will
be the best choice, but I'm just porting an old DOS program to 8.04 and it
would be very easy for me to proceed as I did in my example program below.

ADDED: my example program works as intended, all results are ok

Thanks and Regards!
Hanns

#COMPILE EXE
DEFSNG A-B  ' <-- cave!!
FUNCTION PBMAIN () AS LONG
  A=1
  A%=2
  MSGBOX STR$(A)+STR$(A%)
  DIM B%(1:1)
  B%(1)=3
  B%=4
  MSGBOX STR$(B%(1))+STR$(B%)
END FUNCTION

[Russ Srole]

Hanns,

When I first started using PBWin, someone suggested Dim All and at first I thought it was a big pain, but stuck with it, just to see. The advice was good, it's a good way to head off certain kinds of bugs (typos) at the pass.

[Michael Mattias]

>and B%() and B%

This I know does not work: a scalar variable name may not conflict with an array variable name.

"Work" or "Not Work" aside, coding like this is IMNSHO totally silly. As long as you have to port it, clean it up while you are at it.

[Hanns Ackermann]

Russ, Michael,

many thanks for your kind response. Indeed, it was not my intention to get gray hairs
while coding as I did in my example, it was just to learn somewhat more about the logic
of the compiler. Nevertheless, following your advice, I'll keep hands off things like this
and stick to #dim all as I usually used to do.

BTW, in my example prog the B% and B%() thing works well - AISIMVHO The same is
with B%(B%), B(B%) and B%(B) etc. - I agree, Michael, may be that's really silly, just
a question out of curiosity ...

Regards, Hanns.

[Michael Mattias]

Maybe 'duplicate' variable names don't work for me because I always use #DIM ALL.

Regardless, I don't think I've ever even been tempted to use B% (scalar) and B%() (array) in the same procedure.

#SILLY OFF   ' Save me from myself!