Re: picforth-1.2.3 problems

Samuel Tardieu <[email protected]> 27 Nov 2004 15:34:00 +0100
Newsgroups gmane.comp.lang.forth.picforth
Organization Avian Carrier & Friends
Message-ID <[email protected]>
>>>>> "Jamie" == Jamie Lawson <[email protected]> writes:

Jamie> Thanks for the response. Is this a new requirement? I have not
Jamie> had problems with this before. I assume the stack referred to
Jamie> is the stack in GForth not PicForth. Is this correct?

This is correct.

However, this is not a new requirement: this convention has been
introduced in PicForth 0.14, released in September 2002 :)

Jamie> I see how it follows the PicForth convention but, when in
Jamie> assembler mode, I don't think the syntax should be differ from
Jamie> the standards published in the PIC data sheets (apart from the
Jamie> postfix notation).

Well, you can define your own constants if you wish or use numerical
values. But I prefer PicForth to act consistently between its Forth
and its assembler mode.

Jamie> As I tend to resort to assembler, probably more often than
Jamie> necessary, I am concerned about backwards compatibility with
Jamie> existing code.

Those checks will make PicForth stop when it encounters improper
code. However, once you make the modification in the source, it will
still compile with older versions of PicForth, so you should be safe,
in the sense that in no case will incorrect code be generated.

  Sam
-- 
Samuel Tardieu -- [email protected] -- http://www.rfc1149.net/sam