How to declare function as LPFLOAT?

[Doug Deming]

I need to declare and call a DLL and don't know how to declare
the return type as LPFLOAT (long pointer float). Please help.

Here is the correct C++ definition:
LPFLOAT FindAddress_Array( LPFLOAT lpArray[ArrayPosition],INT nSpaceOfs, INT nOfs, DWORD wStartAddr, DWORD wArraySize );

Here is what I've come up with for PBDLL:
DECLARE FUNCTION FINDADDRESS_ARRAY LIB "elkit32.dll" (BYREF lpArray AS SINGLE, BYVAL nSpaceOfs AS INTEGER, BYVAL nOfs AS INTEGER, BYVAL wStartAddr AS DWORD, BYVAL wArraySize AS DWORD) AS DWORD

I think it's all correct, EXCEPT for the Function type as DWORD.
It's supposed to be a Long Pointer Float (aka Long Pointer Single).
How do I do this?

Thank you,
Doug

------------------

[Florent Heyworth]

The DWORD function return should then be a pointer to a float
which means that you can assign the pointer to a SINGLE PTR
as in:

LOCAL dwReturn AS DWORD
LOCAL sngPtr AS SINGLE PTR

dwReturn = FindAddress_Array()
sngPtr = dwReturn

or more succintly:

sngPtr = FindAddress_Array()
IF sngPtr <> %NULL THEN ...

Have you tried this?

Cheers

Florent




[This message has been edited by Florent Heyworth (edited September 13, 2000).]