Home > Archive > Prolog > August 2005 > Prolog compiler for Windows XP.
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Prolog compiler for Windows XP.
|
|
| kurdayon@yahoo.com 2005-08-04, 5:04 pm |
| Hello,
Can anybody recommend me some free prolog compiler for Windows XP?
| |
| Bart Demoen 2005-08-04, 5:04 pm |
| kurdayon@yahoo.com wrote:
> Hello,
>
> Can anybody recommend me some free prolog compiler for Windows XP?
>
SWI-Prolog - you can download it from http://www.swi-prolog.org/
Cheers
Bart Demoen
| |
| Duncan Patton 2005-08-04, 5:04 pm |
| On Thu, 04 Aug 2005 20:45:57 +0200
Bart Demoen <bmd@cs.kuleuven.ac.be> wrote:
> kurdayon@yahoo.com wrote:
>=20
> SWI-Prolog - you can download it from http://www.swi-prolog.org/
>=20
> Cheers
>=20
> Bart Demoen
Does SWI produce native-mode Win32 apps or does it use a RT dll? =20
My own experience is the Diaz's Gprolog does and is the most robust=20
multi-platform Prolog. =20
If the application doesn't need anything fancy...=20
Dhu
--=20
???????????????????????????????????????
Can't get good help? =20
Contact Fubar the Hack: fubar AT neotext.ca
Area code seven eight zero, Exchange four six six, Local zero one zero nine
Highland terms, Canadian workmanship.
All persons named herein are purely fictional victims
of the Canidian Bagle Breeder's Association.
Save the Bagle!=20
Sun =D0hu
???????????????????????????????????????
| |
| Bart Demoen 2005-08-04, 5:04 pm |
| Duncan Patton wrote:
> My own experience is the Diaz's Gprolog does and is the most robust
> multi-platform Prolog.
Which platforms are supported by gprolog and not by SWI-Prolog. Which
robustness problems have you encountered with SWI-Prolog which didn't
show up with gprolog ?
TIA
Bart Demoen
| |
| Duncan Patton 2005-08-04, 5:04 pm |
| On Thu, 04 Aug 2005 21:24:52 +0200
Bart Demoen <bmd@cs.kuleuven.ac.be> wrote:
> Duncan Patton wrote:
>=20
>=20
> Which platforms are supported by gprolog and not by SWI-Prolog. Which
> robustness problems have you encountered with SWI-Prolog which didn't
> show up with gprolog ?
>=20
> TIA
>=20
> Bart Demoen
It's a knee-jerk response that tells me KISS, and SWI is fancier with
more bells'n'whistles. Apps need a fancier install in their simplest
form. =20
Dhu
--=20
???????????????????????????????????????
Can't get good help? =20
Contact Fubar the Hack: fubar AT neotext.ca
Area code seven eight zero, Exchange four six six, Local zero one zero nine
Highland terms, Canadian workmanship.
All persons named herein are purely fictional victims
of the Canidian Bagle Breeder's Association.
Save the Bagle!=20
Sun =D0hu
???????????????????????????????????????
| |
| student 2005-08-04, 10:01 pm |
| Bart Demoen wrote:
> Duncan Patton wrote:
>
>
>
> Which platforms are supported by gprolog and not by SWI-Prolog. Which
> robustness problems have you encountered with SWI-Prolog which didn't
> show up with gprolog ?
>
> TIA
>
> Bart Demoen
For example, I tried to compile Staerk's LPTP on my computer using
SWI-Prolog (which I like very much) but the resulting executable was not
able to replicate the introductory examples from the accompanying
documentation, so I switched to gprolog and was able not only to
reproduce the introductory examples but was able to produce a few
nontrivial proofs on my own.
If anyone has met with complete success compiling LPTP on SWI-Prolog,
I would greatly appreciate seeing the Makefile, etc.
Bill Hogan
logic4sure@yahoo.com
| |
| Richard HAGEN 2005-08-04, 10:01 pm |
| kurdayon@yahoo.com writes:
>Can anybody recommend me some free prolog compiler for Windows XP?
I'd recommend SWI Prolog (http://www.swi-prolog.org). It's got a
lively user community and it's actively supported on multiple platforms.
--
Today's CGI is Tomorrow's Rear Projection
| |
| Richard HAGEN 2005-08-04, 10:01 pm |
| I've just thought of a slogan for SWI:
"It's not the fastest, but it is the friendliest."
--
Today's CGI is Tomorrow's Rear Projection
| |
| Bart Demoen 2005-08-05, 4:01 am |
| student wrote:
> For example, I tried to compile Staerk's LPTP on my computer using
> SWI-Prolog (which I like very much) but the resulting executable was not
> able to replicate the introductory examples from the accompanying
> documentation, so I switched to gprolog and was able not only to
> reproduce the introductory examples but was able to produce a few
> nontrivial proofs on my own.
>
> If anyone has met with complete success compiling LPTP on SWI-Prolog, I
> would greatly appreciate seeing the Makefile, etc.
The fact that LPTP doesn't install properly for SWI-Prolog is a pity,
but as far as I have seen, it is the fault of the LPTP author's and not
because there is something wrong with SWI. The system specific file
swi.pl for instanc contains a definition of initialization/1 which is a
static predicate in SWI. On the other hand, swi.pl does not contain
a definition of atomic_length/2, while it is in gprolog.pl
(atomic_length/2 is not built-in in SWI - there is a atom_length/2
though).
Just doing the test from the LPTP INSTALL file shows many more of such
errors - all because the promised "it runs on SWI" in the LPTP README
is simply not really true.
Conclusion: PLTP installing in GNU and not in SWI, is related to the
effort Staerk has put into making that true; not to the relative
strength of the two Prolog systems involved.
Cheers
Bart Demoen
| |
| Jan Wielemaker 2005-08-05, 4:01 am |
| On 2005-08-04, Richard HAGEN <zzrhagen@uq.net.au> wrote:
> I've just thought of a slogan for SWI:
>
> "It's not the fastest, but it is the friendliest."
I like it :-) I'd be even more happy to find the resources to be close to
the fastest though ...
Cheers --- Jan
| |
| student 2005-08-05, 9:01 am |
| Bart Demoen wrote:
>
> The fact that LPTP doesn't install properly for SWI-Prolog is a pity,
> but as far as I have seen, it is the fault of the LPTP author's and not
> because there is something wrong with SWI. The system specific file
> swi.pl for instanc contains a definition of initialization/1 which is a
> static predicate in SWI. On the other hand, swi.pl does not contain
> a definition of atomic_length/2, while it is in gprolog.pl
> (atomic_length/2 is not built-in in SWI - there is a atom_length/2
> though).
> Just doing the test from the LPTP INSTALL file shows many more of such
> errors - all because the promised "it runs on SWI" in the LPTP README
> is simply not really true.
>
> Conclusion: PLTP installing in GNU and not in SWI, is related to the
> effort Staerk has put into making that true; not to the relative
> strength of the two Prolog systems involved.
>
That is a cheap shot, not a conclusion, at least not if by
"conclusion" you mean the last sentence in a proof.
Granting the same facts, one might equally "conclude" that
LPTP's installing in GNU and not in SWI is the result of
negligence on the part of SWI in hewing to the same high standard
excellence as Gnu Prolog. I am not sure it is the responsibility of Mr
Staerk to accommodate every Prolog compiler in existence.
To the contrary, when it comes to applications as important
as LPTP, I think that at least 50% of the synergy required
to make LPTP and SWI-Prolog work together smoothly should come from the
SWI-Prolog side.
BH
| |
| Torkel Franzen 2005-08-05, 9:01 am |
| student <not@anywhere.invalid> writes:
> To the contrary, when it comes to applications as important
> as LPTP, I think that at least 50% of the synergy required
> to make LPTP and SWI-Prolog work together smoothly should come from the
> SWI-Prolog side.
This makes no sense whatever. The importance of various applications
- however we measure this - is completely irrelevant to the quality
and design of a programming language. Naturally it is the
responsibility of those who produce these important applications to
decide which platforms they want them to run on, and to make sure they
do.
| |
| Jan Wielemaker 2005-08-05, 5:02 pm |
| On 2005-08-05, Torkel Franzen <torkel@sm.luth.se> wrote:
> student <not@anywhere.invalid> writes:
>
>
> This makes no sense whatever. The importance of various applications
> - however we measure this - is completely irrelevant to the quality
> and design of a programming language. Naturally it is the
> responsibility of those who produce these important applications to
> decide which platforms they want them to run on, and to make sure they
> do.
SWI-Prolog has always strived to be compatible with standards as well as
de-facto standards. It is however completely unrealistic to expect
people developing a language to start hunting for applications and make
sure they run. It is upto users or developers of such applications to
signal problems. I'm more than happy to cooperate with application and
library developers to find solutions for portability issues. First thing
to look for is whether there is a good solution that satisfies standards
or de-facto standards. If there is I generally make sure SWI-Prolog will
be compatible to these. If not, it is sometimes a good idea to add or
extend a xyz_prolog compatibility library.
Cheers --- Jan
| |
| Bart Demoen 2005-08-05, 5:02 pm |
| student wrote:
> Bart Demoen wrote:
>
> That is a cheap shot, not a conclusion, at least not if by "conclusion"
> you mean the last sentence in a proof.
I meant neither. The author of LPTP writes in the INSTALL file:
- LPTP runs under SICStus Prolog, GNU Prolog, SWI Prolog,
Quintus Prolog, ECLiPSe, or CProlog.
I assume that at the time he wrote that, it was true (also for SWI), but
he does not mention the version of any of these systems. Systems have no
obligation for backward compatibility for every application ever
written. Possibly the reason why LPTP "still" runs under GNU-Prolog is
because GNU-Prolog has not "evolved" since the moment the above INSTALL
notes were written, while SWI has.
> Granting the same facts, one might equally "conclude" that
> LPTP's installing in GNU and not in SWI is the result of
> negligence on the part of SWI in hewing to the same high standard
> excellence as Gnu Prolog. I am not sure it is the responsibility of Mr
> Staerk to accommodate every Prolog compiler in existence.
I agree that it is not the responsability of Mr Staerk to accomodate
every Prolog compiler. But I do not think that Jan Wielemaker has the
responsability of making sure that LPTP runs unchanged in all future
versions of SWI. [Torkel and Jan have argued that also in the mean time,
I think].
> To the contrary, when it comes to applications as important
> as LPTP, I think that at least 50% of the synergy required
> to make LPTP and SWI-Prolog work together smoothly should come from the
> SWI-Prolog side.
Perspective please ... in Flemmish we have this saying: one fool can ask
more questions than ten wise men can answer. [just to make sure: I am
not comparing Staerk or you to either of the fool or wise men]
But one Prolog system's implementation team against the rest of the
world who is implementing important applications ... it doesn't seem a
fair game, neither possible for Jan or any other larger Prolog
development team to keep up the standards you require.
Or maybe I should ask: which other applications are important enough in
you opinion that Jan should make sure they run forever in SWI ?
Cheers
Bart Demoen
|
|
|
|
|