On Thu, Nov 9, 2017 at 11:52 AM, Ralf Quint <freedos...@gmail.com> wrote:
> On 11/8/2017 6:07 PM, Rugxulo wrote:
>> On Wed, Nov 8, 2017 at 1:23 PM, Ralf Quint <freedos...@gmail.com> wrote:
>>> On 11/8/2017 10:26 AM, Jim Hall wrote:
>>>> Not sure why DeSmet wasn't well received long ago, but I think this is
>>>> a Good Thing to see. I'd love to see DeSmet included in the next FreeDOS.
>> It's very minimal, maybe too minimal. It wasn't packaged well.
The main page of the website at http://www.desmet-c.com/ hasn't been
updated (apparently) since 2009.
You say it's as good as DJGPP (as far as maintenance), but that's not
true at all. 2009 would be GCC 4.4.2, and they've had many releases
since then. Even in the 4.x series, we got 4.9.4 back in Aug. 2016.
The DeSmet website lists (in order):
* DeSmet C, version 2.40
* DeSmet C, version 2.51
* PCC version 1.2D
* DeSmet C, version 3.03
* DeSmet C, version 3.1h
* DeSmet C, version 3.1L
* DeSmet C, version 3.1N
* DeSmet C Tools disk
* SEE Version 4.2 [editor]
* SEE Version 3.2
* O88, version 3.06D [optimizer]
(and a few other links)
It's very confusing. Which .ZIPs do you recommend? (Presumably 3.1N
plus ... ??) Which one does O88 work with (if any)?
Even locally, I have nine .ZIPs:
Okay, so I found INCLUDE.ZIP (among a billion other files) which has
"some" (but apparently not all 15) C89 includes. Even unzipping all
archives, I can't seem to find these required headers:
Again, I'm not trying to be rude or complain, but it's somewhat incomplete.
>> The popular "shareware" version (PCC) back in the day (e.g. on Simtel) was
>> lacking even more:
> I stopped reading your reply at that point, as it clearly shows that you
> haven't really looked at it at all.
Yes, I have looked. That link was only for historical context, to
prove that it really started out as even more minimal. Yes, I know
that "old" version is pre-ANSI, and my point was that it lacked many
headers (and only comes with these four: stdio.h, ctype.h, setjmp.h,
Do you see what I mean? I'm sympathetic, but it's hard to convince
people to use solutions that don't even provide full C89 (ANSI)
> The shareware version is equivalent to the pre-ANSI 2.51 version.
Yes, I know that, and it's literally dated circa 1989, so that's
understandable. However, even the modern "release" has .EXEs that are
no newer (apparently) than 1992. That is not what I'd call
I don't blame anybody, free time is always short. But let's not
pretend this is the perfect solution that we've been waiting for.
> You are comparing apples and oranges,
No, I'm comparing DeSmet C, in its latest release, with ancient
__STDC__ from 1989 (even before __STDC_VERSION__ existed). It comes up
lacking. It may work for some things, and I'm not complaining.
Whatever works for you is good enough.
But let's not tell people that it can replace OpenWatcom or DJGPP,
which are literally hundreds of times better.
> and contrary to your statement in another reply, all you do is complaining.
If you want to waste time making a package for this, go ahead. Nobody
else cares. I'm still sympathetic, but it is not really as good as you
pretend (at least not "as is", maybe we can all improve it,
Again, I think SmallerC is better in every way (now also with OS X
support). That doesn't mean we can have both, but don't overhype it.
P.S. Read this:
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
Freedos-user mailing list