Hi Paul,

On Dec 14 13:23, 1996, Paul Edwards of 3:711/934.9 wrote:

PE>>> The *.MSG header has no place in it for "date
arrived".  It could

PE> I have done the mods to both MSGED and Tobruk to handle date 
PE> arrived. I don't think I should send you the changes though, 
PE> because that will just confuse your code with mine.

Fine.

PE> BTW, did you get the message I wrote to you in PUBLIC_DOMAIN
PE> about the Lattice makefile?

Not originally, because I used the wrong directory name for
PUBLIC_DOMAIN in my config file.  :( Fortunately, I still have all
the mail packets that I've received & sent, so when I noticed the
problem the other day, I rebuilt my message base.

PE> Also, I think Andrew is likely to be merging all the changes in
PE> to MSGED over Christmas.  Can you get your mods in to him soon?
PE> Especially the non-Lattice-5.1 mods.

I'll do my best; the current form is probably ok.  When I first
started to work on msged, I just looked at sasc.c & "fixed" it so
that it used event-driven methods.  Then, when I looked at ansi.c, I
realized that this was not really appropriate.  Anyway, I got it
working, but it required the user to press a key on exit to "flush"
the pending read request packet.  (No, I can't see how to abort the
request.) I finally got sick of this sloppy behaviour, so today I
changed the getch() routine back to what it was.

PE> I'm not sure what the philosophy on the Lattice 5.1 mods should
PE> be.  I think you should have an independent archive containing
PE> 5.1 fixups, which can be used in all products expecting an ISO
PE> environment etc.  Then again, I included a PACIFIC.C in MSGED
PE> last time.  But then I also made available a patch to Pacific
PE> C's header files as a separate thing.

At present, I use a set of include files which patch the originals
(lc*.h) and a library file (lcext.lib).  I'm not exactly sure how to
do this properly, though.  Because I need to specify a math library
to compile difftime(), it makes lcext.lib specific to that math
library.

PE> The thing is, it's likely to be me, not you, that will be doing
PE> the Amiga compile, since I have the more advanced compiler.  So
PE> it's only people who are stuck with an old Lattice compiler and
PE> want to do development that need the fixups.
PE> I think I'd start with a separate Lattice C thing.

Understood.

PE> BTW, in order to get quick.c to compile, you must have done
PE> something to make up for the fact that fstat() isn't defined.
PE> What did you do?  I hope you created a LATTICE etc instead of
PE> using SASC, because the current SASC has an fstat().

Yes, all is well - see another msg in this packet for details.

PE> BTW, the message in PUBLIC_DOMAIN was trying to get a Lattice C
PE> makefile as just one file, instead of needing the separate link
PE> file.

Oh, I did that ages ago.

Michael Stapleton of Graphic Bits
@EOT:

--- Msged/AM 4.00