[futurebasic] Re: [FB] FB^3 Toolbox HANDANDHAND

Message: < previous - next > : Reply : Subscribe : Cleanse
Home   : September 1999 : Group Archive : Group : All Groups

From: "Dr. Steven J. Stratford" <sstratford@...>
Date: Sun, 19 Sep 1999 07:59:00 -0500
> >Line 308 in "Tlbx Standard.INCL" says
>>toolbox fn HandAndHand(long) = long `0x205F,0xA9E4,0xC188
>>However Think Reference says
>>iErr := HandAndHand ( srcHandle, destHandle );
>I'm going from memory here, this was discussed _real_ early in the beta
>The problem is that Apple _changed_ the definition on us! (Actually
>giving two.) The new FB^3 way is the PPC method, the old FBII way is the
>68K method. C programmers have to put a COMPILE LONG IF (equivalent)
>around every HANDANDHAND. Andy & Staz figured it was better to just have
>one method, matching the "new" way, for future (Carbon, etc.)
>compatibility, with a minimum of "relearning". I _think_ I covered this
>in the conversion docs... I know I did for GESTALT...

How does the new definition work? The old one I 'get'--append the 
contents of the first handle onto the second. Since I haven't been 
involved in FB^3 beta, could someone forward me copies of the 
relevant beta emails that discussed this? Or are they archived 

You covered HandToHand but not HandAndHand, as far as I can see.

>>Can I just change the line in Tlbx Standard.INCL to make it precisely
>>compatible with the Mac Toolbox? The trapword is 0xA9E4. iErr is a
>>SHORT. Would the 100% compatible toolbox call be this?
>>toolbox fn HandAndHand(long, long) = short `0xA9E4
>No. You could _add_ a function that would look like this, maybe naming it
>"myHandAndHand", and maybe get a TBALIAS to work, but (I think) when you
>try to call it from PPC, you'll get an error. (But you can certainly try
>it! That's one nice thing about being able to edit the runtimes and
>headers... just keep a "clean" copy handy!)

Has anyone written down what to watch out for in the FBII->FB^3 
converting process with regard to 68K->PPC? I think I vaguely but not 
completely get the register-based variables stuff. What else do we 
have to watch out for?


>To unsubscribe, send ANY message to <futurebasic-unsubscribe@...>