[retroforth] Re: Request for Linux specific words
- From: "Helmar Wodtke" <helmwo@xxxxxx>
- To: retroforth@xxxxxxxxxxxxx
- Date: Thu, 20 Jan 2005 21:55:23 +0100
retroforth@xxxxxxxxxxxxx schrieb am 20.01.05 21:41:29:
>
>
> On Thu, January 20, 2005 12:14, aprice@xxxxxxxxxxxxxx said:
> I wrote:
> >> I agree. I also mentioned that appending-a-zero thing to Charles a while
> >> ago, I don't think he was receptive to the idea (but I'll be doing it in
> >> Reva).<<
>
> > It seems to me that appending a zero is adding a tax (and complexity) to
> > every operation on strings. zstrings are a historical anomoly -- an
> > inefficient aberration that should be forgotten: therefore embedding
> > support for them more deeply than absolutely neccessary is a mistake. So I
> > concur with Charles here. :)
>
> I strongly disagree. One who wishes to interact with the OS (Linux or
> Windows, anyway), needs to pass a NUL terminated string. Requiring the
> programmer to do that manually every time is simply wrong, and inefficient.
I'm not so sure. The problem is that in the case you want to construct strings
you always will have to take care about the zero at end. Or do you want to make
cmove always moving one zero character more? I would prefer words like
z" This is a zero-terminated string that passes a pointer to stack only"
for the library interfacing. The original problem was argument passing - and
there I'm not sure. I think not to touch the system (that gives asciiz-strings)
is best option. But this means it's not so nice to access the things from FORTH.
Bis dann,
Helmar
helmwo@xxxxxx
-- Binary/unsupported file stripped by Ecartis --
-- Type: application/x-pkcs7-signature
-- File: smime.p7s
-- Desc: S/MIME Cryptographic Signature
Other related posts:
