🚀 go-pugleaf

On Friday, in article <881886281snz@wash.demon.co.uk>
     john$@wash.demon.co.uk "John Washington" wrote:

> On 11 Dec, in article
>      <slrn68ujte.qs4.robm@desade.noc.demon.net> robm@demon.net
>      "Robert Mortimer" wrote:
> 
> RM> It has been for a while, and suffers from the slight drawback, as with
> RM> anything compiled by DJGPP, in that it won't run on sub 386's.
> 
> OK, so that's a (dis)advantage of that compiler.  I've not seen
> that source, but is it reasonably std C?  If so, would it be
> a good base to take and add another 'make' file for BCC?
> No doubt there'd need to be some condit compile, e.g. ksubr.c
> is presumably rather different.  But do people think it would
> be much more than that?

Reasonably std C ? How "std C" do you think Textwin is ?

So far as I know, nobody does DJGPP >> BC conversions, and the reasons 
are probably (a) it'd be seen as a retrograde step and (b) it's not easy 
at all.
I think that I've tried all of the main KA9Q flavours and the Karn/TNOS 
versions don't stand out. In fact none of them *function* that much 
better than the bog standard DIS 2.16d and most are well behind GEK's 
latest and greatest. Most have bells and whistles lacking in Demon's, 
and even Graham's, finest, but none actually work better, ie no faster 
or less flaky.
Why not just give in and track gremNOS ? It works fine, is a real 
advance and the code appears to be "std C", which is more than you can 
say for TW :(

> The way I'd like to go with TextWin is (regardless of whether we
> use the Karn source or not) to abandon the present method of
> issuing a set of replacement source modules.  Instead provide
> a utility (plus script files) that would take a given ka9q
> source and attempt to generate the extensions for TW.

I'd like that, *if* I could compile TW, but as it stands BC 2.0 and 3.0 
can't. Whatever "std C" TW's written in, I don't think anything earlier 
than BC 3.1 can talk it and I'd be amazed if any non-Borland compiler 
could either.

> The intended benefit is that release of a new ka9q could then
> have TW immediately bolted in without brain-work.  Of course,
> it would rely on ka9q upgrades not being too major.

Hmmm, I think that's where it dies. TW is essentially just an interface 
and that is one of the areas in which the later Karn efforts have 
changed rather dramatically. The one thing I did like was the screen 
scroll-back, but the surgery looked beyond my motivation and, probably, 
ability.

> What is the 'Karn' source?  Is it really by the great man himself,
> and does he still have the time to make a good job of it?

Source written by Karn ? The "export" version, I'd imagine.
As for the other question; yes and probably. Any the wiser ?
The better question might be, has he done a useful job ?  For my 
purposes, no.

> Is it responsively maintained, amalgamating best ideas from
> elsewhere?  Or are there other versions around that would be
> better to track?

Oh dear, who decides what's best ? Responsive to who ?
Face it, all of the main versions are Ham Radio driven. I expect that 
the best Ham ideas are there and in that sense it's responsively 
maintained, but my, and probably your, needs are not the same as Ham 
Radio operators.

Why worry ? My impression is that GEK has already been there and done 
that. His efforts may not exactly fit your requirements, but riding on 
his back has far more advantages than disadvantages. The main gain (for 
me at least) being that he's a far better programmer than I am.
My advice stands; take Graham's latest and bolt on your goodies.

> Hey, if it doesn't have AMcC's TextWin in it, how can it
> possibly be any good?  ;-)

It could have all the features that the DIS KA9Q and TW lack :)
This, presumably, is the point of the exercise and it begs the question;
what do you think is missing from DIS/TW ?

Ian.
-- 

  |  Ian Smith             |  "The Moving Finger writes;
  |  ian@isis.demon.co.uk  |   and, having writ, Moves on."