Hi Paul



 RM> Shall I raise this with Borland, or would you like to?



 PE> I don't get on with Borland very well, so I'll leave that to you.

 PE> You will have seen that there are a couple of messages saying how

 PE> to get around the problem.



Yes, saw that and checked - the 4.02 help file has that

explanation, but the 3.0 help file doesn't.



 PE> However, what is unforgivable is that

 PE> there is no compile-time option to override the situation where

 PE> the compiler gets it wrong.  BFN.  Paul.



Agreed - finding the float in the &s->x might be a bit harder than

normal (is it possible to declare the structure baldly and use it

before defining its contents?), but the "%f" is a bit of a

giveaway, surely the compiler can work it out from that...



Cheers



--- PPoint 1.88