eclipse-clp-users — ECLiPSe Users List

Re: [eclipse-clp-users] Eclipse library IC question on reading the intervals

[Joachim S. <jsc...@co...>]

On 22/05/18 12:22, Julio wrote:
> Hi,
> 
> 
> I am using eclipse 7.0 #42 for linux 64 bits, and after loading (in the 
> interpreter) the IC library the answer to the following query appears 
> "erroneous" to me.
> 
> [eclipse 2]: [E1,E2,N1] $:: -1..1, X $= 4 + E1 + E2 + N1, X $= -E1 + 3*E2.
> 
> E1 = E1{-1.0 .. 0.5000000116152874}
> E2 = E2{0.49999998838471316 .. 1.0}
> N1 = N1{-1.0 .. 0.5000000116152874}
> X = X{2.4999999651541396 .. 4.0}
> 
> 
> Delayed goals:...
> 
> It appears to me that the interval for E1 should be -1..-0.5 and that of N1 
> should be -1..0.
> 
> We can, roughly, see that this is the case for E1 when we 'narrow' the domain of 
> E1 as follows.
> 
> [eclipse 3]: E1 $:: -0.4..0.5, [E2,N1] $:: -1..1, X $= 4 + E1 + E2 + N1, X $= 
> -E1 + 3*E2.
> 
> No (0.00s cpu)
> 
> What am I missing in the reading of the displayed intervals?
> 
> Is there an heuristic/method to lead the constraint solver to a tighter answer, 
> in my example?


Hi Julio,

as a general rule, in order to solve a constraint-problem, you have to combine 
the statement of the constraints with a search-procedure.  The constraints (more 
precisely, the implementation's network of "constraint-propagators") help the 
search procedure by pruning domain ranges that definitely do not contain solutions.

If (as in your example) you only state a conjunction of constraints, then you 
only get a "conditional answer" (indicated by the presence of delayed goals). 
The only guarantee you have here is that there are no answers _outside_ the 
resulting domains.  There is no guarantee about tightness, or about the 
existence or number of actual solutions.

When working with integer domains, search is usually done with
http://eclipseclp.org/doc/bips/lib/ic/search-6.htm, which will
locate individual solutions.

With continuous domains, you can use
http://eclipseclp.org/doc/bips/lib/ic/squash-3.html,
http://eclipseclp.org/doc/bips/lib/ic/locate-3.html or
http://eclipseclp.org/doc/bips/lib/ic/locate-4.html.

In your case, where an infinite number of solutions exists, squash/3 is 
appropriate, and narrows the domain bounds to the values you expect:

?- [E1, E2, N1] $:: -1..1,
    X $= 4 + E1 + E2 + N1, X $= -E1 + 3*E2,
    squash([E1, E2, N1, X], 0.001, lin).

E1 = E1{-1.0 .. -0.49908237516423415}
E2 = E2{0.49999998838471316 .. 1.0}
N1 = N1{-1.0 .. 0.0018352496715316469}
X = X{2.4999999651541396 .. 4.0}
There are 2 delayed goals.
Yes (0.00s cpu)



Cheers,
Joachim

[eclipse-clp-users] Eclipse library IC question on reading the intervals

[Julio <jpe...@nu...>]

Hi,


I am using eclipse 7.0 #42 for linux 64 bits, and after loading (in the 
interpreter) the IC library the answer to the following query appears 
"erroneous" to me.

[eclipse 2]: [E1,E2,N1] $:: -1..1, X $= 4 + E1 + E2 + N1, X $= -E1 + 3*E2.

E1 = E1{-1.0 .. 0.5000000116152874}
E2 = E2{0.49999998838471316 .. 1.0}
N1 = N1{-1.0 .. 0.5000000116152874}
X = X{2.4999999651541396 .. 4.0}


Delayed goals:...

It appears to me that the interval for E1 should be -1..-0.5 and that of 
N1 should be -1..0.

We can, roughly, see that this is the case for E1 when we 'narrow' the 
domain of E1 as follows.

[eclipse 3]: E1 $:: -0.4..0.5, [E2,N1] $:: -1..1, X $= 4 + E1 + E2 + N1, 
X $= -E1 + 3*E2.

No (0.00s cpu)

What am I missing in the reading of the displayed intervals?

Is there an heuristic/method to lead the constraint solver to a tighter 
answer, in my example?

Kind regards,

Julio

Re: [eclipse-clp-users] element/3 constraint in the ic library

[Joachim S. <jsc...@co...>]

On 15/04/18 20:57, Panagiotis Stamatopoulos wrote:
> Hello everybody,
> 
> The second argument of the element/3 constraint of the ic library
> (version 7.0) is supposed to be "a non-empty collection of numbers
> or variables", according to the documentation. However, the following
> example included in the description of element/3 does not work as
> expected:
> 
> [eclipse 2]: A#::2..4, B#::8..9, C#::4..5, element(I, [A,B,C], X), X#=<7.
> instantiation fault in element(I, [A{2 .. 4}, B{[8, 9]}, C{[4, 5]}], X)
> Abort
> 
> Any idea what went wrong?


Apologies, this is a functionality enhancement that I let slip in between 
version 7.0#41 and 7.0#42 (normally we don't change functionality without 
changing the x.y version number).

Previously, the second argument of ic:element/3 had to be fully instantiated. 
If you need the new functionality, please update to 7.0#42.


Regards,
Joachim

[eclipse-clp-users] element/3 constraint in the ic library

[Panagiotis S. <ta...@di...>]

Hello everybody,

The second argument of the element/3 constraint of the ic library
(version 7.0) is supposed to be "a non-empty collection of numbers
or variables", according to the documentation. However, the following
example included in the description of element/3 does not work as
expected:

[eclipse 2]: A#::2..4, B#::8..9, C#::4..5, element(I, [A,B,C], X), X#=<7.
instantiation fault in element(I, [A{2 .. 4}, B{[8, 9]}, C{[4, 5]}], X)
Abort

Any idea what went wrong?

Regards,
Panagiotis Stamatopoulos

Re: [eclipse-clp-users] Debugging in embedded eclipse

[Victor M. <ma...@fh...>]

On Mittwoch, 4. April 2018 22:31:58 CEST Joachim Schimpf wrote:
> On 04/04/18 15:04, Victor Mataré wrote:
> > Hello,
> > 
> > I'm trying to debug/trace eclipse code that is being called from an
> > instance embedded in C++ code, but I can't seem to trigger interactive
> > tracing. trace(p) does not trigger tracing for p, and trace/0 or debug/0
> > cannot even be called. I.e. in C++ i do:
> > 
> > post_goal(EC_atom("trace"));
> > EC_resume();
> > 
> > Which results in:
> > 
> > calling an undefined procedure profile : trace in module eclipse
> > 
> > However I'm not even sure EC_atom is the correct way of referencing a
> > predicate with arity 0. Can't find anything covering this in the
> > documentation either. trace/0 also seems to be in none of the standard
> > modules, for example when I get a list of all available modules in a
> > plain interactive eclipse:
> > 
> > [eclipse 14]: findall(M, current_module(M), L), writeln(L).
> > [eclipse, sepia_kernel, dfid, eclipse_language, profile, toplevel,
> > ecl_compiler, tracer_tty, suspend, development_support, lists]
> > 
> > Then trace/0 doesn't seem to be in any of them, i.e. any of eclipse/trace,
> > sepia_kernel/trace, ... yield the same error as above.
> > 
> > So what's going on here? How do I trace embedded code?
> 
> Hi Victor,
> 
> trace/0 and debug/0 are not actually predicates, they are commands
> interpreted by the interactive toplevel.  trace/1 is in principle the
> correct predicate to use, but in an embedded situation you have to load any
> development tools explicitly beforehand.
> 
> You can use either the command line tools or the graphical tools.  For
> reference, I have summarized the necessary steps on our wiki at
> 
> http://eclipseclp.org/wiki/Examples/DebuggingEmbeddedECLiPSeCode
> 
> Let me know if anything is unclear!
> 
> 
> Cheers,
> Joachim
> 
> ----------------------------------------------------------------------------
> -- Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> ECLiPSe-CLP-Users mailing list
> ECL...@li...
> https://lists.sourceforge.net/lists/listinfo/eclipse-clp-users


Thanks Joachim, that looks awesome. Didn't even know about this wiki before. From a quick glance it looks very informative. The only thing that's unclear to me is why you're recommending against using interactive debugging in an embedded application in general.

Best regards,
Victor

Re: [eclipse-clp-users] Debugging in embedded eclipse

[Joachim S. <jsc...@co...>]

On 04/04/18 15:04, Victor Mataré wrote:
> Hello,
> 
> I'm trying to debug/trace eclipse code that is being called from an instance embedded in C++ code, but I can't seem to trigger interactive tracing. trace(p) does not trigger tracing for p, and trace/0 or debug/0 cannot even be called. I.e. in C++ i do:
> 
> post_goal(EC_atom("trace"));
> EC_resume();
> 
> Which results in:
> 
> calling an undefined procedure profile : trace in module eclipse
> 
> However I'm not even sure EC_atom is the correct way of referencing a predicate with arity 0. Can't find anything covering this in the documentation either. trace/0 also seems to be in none of the standard modules, for example when I get a list of all available modules in a plain interactive eclipse:
> 
> [eclipse 14]: findall(M, current_module(M), L), writeln(L).
> [eclipse, sepia_kernel, dfid, eclipse_language, profile, toplevel, ecl_compiler, tracer_tty, suspend, development_support, lists]
> 
> Then trace/0 doesn't seem to be in any of them, i.e. any of eclipse/trace, sepia_kernel/trace, ... yield the same error as above.
> 
> So what's going on here? How do I trace embedded code?


Hi Victor,

trace/0 and debug/0 are not actually predicates, they are commands interpreted 
by the interactive toplevel.  trace/1 is in principle the correct predicate to 
use, but in an embedded situation you have to load any development tools 
explicitly beforehand.

You can use either the command line tools or the graphical tools.  For 
reference, I have summarized the necessary steps on our wiki at

http://eclipseclp.org/wiki/Examples/DebuggingEmbeddedECLiPSeCode

Let me know if anything is unclear!


Cheers,
Joachim

[eclipse-clp-users] Debugging in embedded eclipse

[Victor M. <ma...@fh...>]

Hello,

I'm trying to debug/trace eclipse code that is being called from an instance embedded in C++ code, but I can't seem to trigger interactive tracing. trace(p) does not trigger tracing for p, and trace/0 or debug/0 cannot even be called. I.e. in C++ i do:

post_goal(EC_atom("trace"));
EC_resume();

Which results in:

calling an undefined procedure profile : trace in module eclipse

However I'm not even sure EC_atom is the correct way of referencing a predicate with arity 0. Can't find anything covering this in the documentation either. trace/0 also seems to be in none of the standard modules, for example when I get a list of all available modules in a plain interactive eclipse:

[eclipse 14]: findall(M, current_module(M), L), writeln(L).
[eclipse, sepia_kernel, dfid, eclipse_language, profile, toplevel, ecl_compiler, tracer_tty, suspend, development_support, lists]

Then trace/0 doesn't seem to be in any of them, i.e. any of eclipse/trace, sepia_kernel/trace, ... yield the same error as above.

So what's going on here? How do I trace embedded code?

Thanks in advance & regards,
Victor

[eclipse-clp-users] missing files

[Giuseppe C. <ct...@un...>]

Dear ECLiPSe developers,

I was reading your guide "Embedding and Interfacing Manual" about 
Java-ECLiPSe interface. In particular, I was reading Section 8.5 
"Communicating between Java and ECL^i PS^e using queues". Unfortunately, 
the prolog examples in here 
http://eclipseclp.org/doc/examples/JavaInterface/ are not reachable (I 
suppose there is something wrong with permissions). Could you please 
make queue_example_1.pl 
<http://eclipseclp.org/doc/examples/JavaInterface/queue_example_1.pl> 
and queue_example_2.pl 
<http://eclipseclp.org/doc/examples/JavaInterface/queue_example_2.pl> 
available again?

Best regards,
Giuseppe Cota

Re: [eclipse-clp-users] Errors when trying to build 7.0#37 from sources on macOS 10.13.3 (High Sierra)

[Joachim S. <jsc...@co...>]

On 03/02/18 23:48, Paulo Moura wrote:
> Hi,
> 
>> On 3 Feb 2018, at 19:42, Joachim Schimpf <jsc...@co...> wrote:
>>
>> On 03/02/18 15:26, Paulo Moura wrote:
>>> Hi,
>>> I'm trying to build 7.0#37 from sources on macOS 10.13.3 (High Sierra):
>> ...
>>> unimplemented functionality in integer_list(9007474139725823, 31, _404977)
>>
>> This is caused by building without bignum support (GMP or MPIR library):
>> the compiler can then not process clauses with too many environment variables.
>>
>>
>> The current configure script will look for gmp/mpir in
>>
>> $ECLIPSETHIRDPARTY/{gmp4.2,gmp4.1,mpir1.3.1}/$ECLIPSEARCH/{lib,include}
> 
> Why such a odd location for third-party software? macOS users usually use either MacPorts or Homebrew for installing third-party software like GMP. MacPorts install on /opt/local. Homebrew install on /usr/local. I use MacPorts and tried several variants:
> 
> export ECLIPSETHIRDPARTY=/opt/local
> export ECLIPSETHIRDPARTY=/opt/local/lib
> export ECLIPSETHIRDPARTY=/opt/local/include
> 
> each after a make -f Makefile.$ECLIPSEARCH clean but got the same errors.

ECLIPSETHIRDPARTY is a directory structure you set up yourself.
It has a fixed structure, independent of operating system whims:
useful for cross-compilation (where we can't configure) and for
cases where configure doesn't do the right thing.


>  I'm also at:
> 
> $ port info gmp
> gmp @6.1.2_1 (devel, math)
> 
> Are the old 4.1/4.2 versions required?

No -- as long as gmp sticks to its APIs.


> 
>> or, failing this, try to use a system-wide installed one by linking with -lgmp
> 
> How exactly?

Well, you tell me!  How _do_ you link something with gmp on macOS?

Apparently our configure check
	AC_CHECK_LIB(gmp, __gmpz_import)
fails...


-- Joachim

Re: [eclipse-clp-users] Errors when trying to build 7.0#37 from sources on macOS 10.13.3 (High Sierra)

[Paulo M. <pm...@lo...>]

Hi,

> On 3 Feb 2018, at 19:42, Joachim Schimpf <jsc...@co...> wrote:
> 
> On 03/02/18 15:26, Paulo Moura wrote:
>> Hi,
>> I'm trying to build 7.0#37 from sources on macOS 10.13.3 (High Sierra):
> ...
>> unimplemented functionality in integer_list(9007474139725823, 31, _404977)
> 
> This is caused by building without bignum support (GMP or MPIR library):
> the compiler can then not process clauses with too many environment variables.
> 
> 
> The current configure script will look for gmp/mpir in
> 
> $ECLIPSETHIRDPARTY/{gmp4.2,gmp4.1,mpir1.3.1}/$ECLIPSEARCH/{lib,include}

Why such a odd location for third-party software? macOS users usually use either MacPorts or Homebrew for installing third-party software like GMP. MacPorts install on /opt/local. Homebrew install on /usr/local. I use MacPorts and tried several variants:

export ECLIPSETHIRDPARTY=/opt/local
export ECLIPSETHIRDPARTY=/opt/local/lib
export ECLIPSETHIRDPARTY=/opt/local/include

each after a make -f Makefile.$ECLIPSEARCH clean but got the same errors.  I'm also at:

$ port info gmp
gmp @6.1.2_1 (devel, math)

Are the old 4.1/4.2 versions required?

> or, failing this, try to use a system-wide installed one by linking with -lgmp

How exactly?

Does anyone in this mailing list succeeded to install ECLiPSE 7.x on macOS? If so, can you share the exact steps?

Thanks,
Paulo

-----------------------------------------------------------------
Paulo Moura
Logtalk developer

Re: [eclipse-clp-users] Errors when trying to build 7.0#37 from sources on macOS 10.13.3 (High Sierra)

[Joachim S. <jsc...@co...>]

On 03/02/18 15:26, Paulo Moura wrote:
> Hi,
> 
> I'm trying to build 7.0#37 from sources on macOS 10.13.3 (High Sierra):
> 
...
> unimplemented functionality in integer_list(9007474139725823, 31, _404977)

This is caused by building without bignum support (GMP or MPIR library):
the compiler can then not process clauses with too many environment variables.


The current configure script will look for gmp/mpir in

$ECLIPSETHIRDPARTY/{gmp4.2,gmp4.1,mpir1.3.1}/$ECLIPSEARCH/{lib,include}

or, failing this, try to use a system-wide installed one by linking with -lgmp


Cheers,
Joachim

[eclipse-clp-users] Errors when trying to build 7.0#37 from sources on macOS 10.13.3 (High Sierra)

[Paulo M. <pm...@lo...>]

Hi,

I'm trying to build 7.0#37 from sources on macOS 10.13.3 (High Sierra):

15:13 $ export ECLIPSEARCH=x86_64_macosx
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37 
15:13 $ export ECLIPSETHIRDPARTY=/opt/local
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37 
15:14 $ CONFIG_SITE=config.$ECLIPSEARCH
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37 
15:14 $ ./configure
checking build system type... x86_64-apple-darwin17.4.0
checking host system type... x86_64-apple-darwin17.4.0
configure: Building for ECLiPSe architecture ----- x86_64_macosx -----
configure: No file 
checking for gcc... gcc
checking whether the C compiler works... yes
checking for C compiler default output file name... a.out
checking for suffix of executables... 
checking whether we are cross compiling... no
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
configure: WARNING: C compiler doesn't understand -fno-schedule-insns
configure: WARNING: C compiler doesn't understand -fno-schedule-insns2
configure: WARNING: C compiler doesn't understand -fira-loop-pressure
configure: WARNING: C compiler doesn't understand -fschedule-fusion
configure: WARNING: C compiler doesn't understand -frename-registers
configure: GCCLIBDIR is /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/lib/clang/9.0.0/lib/darwin17.4.0
configure: OSTYPE is darwin17
...
15:15 $ make -f Makefile.$ECLIPSEARCH
make[2]: Nothing to be done for `objects'.
LD_LIBRARY_PATH=/lib: ../bin/x86_64_macosx/eclipse -e 'get_flag(library_path,P),set_flag(library_path,["."|P]),lib(fcompile), set_flag(variable_names,off), fcompile("graph_algorithms", [outdir:"../lib"])'
make[2]: Nothing to be done for `objects'.
LD_LIBRARY_PATH=/lib: ../bin/x86_64_macosx/eclipse -e 'get_flag(library_path,P),set_flag(library_path,["."|P]),lib(fcompile), set_flag(variable_names,off), fcompile("heap_array", [outdir:"../lib"])'
make[2]: Nothing to be done for `objects'.
LD_LIBRARY_PATH=/lib: ../bin/x86_64_macosx/eclipse -e 'get_flag(library_path,P),set_flag(library_path,["."|P]),lib(fcompile), set_flag(variable_names,off), fcompile("fd_sets", [outdir:"../lib"])'
make[2]: Nothing to be done for `objects'.
LD_LIBRARY_PATH=/lib: ../bin/x86_64_macosx/eclipse -e 'get_flag(library_path,P),set_flag(library_path,["."|P]),lib(fcompile), set_flag(variable_names,off), fcompile("ic_sets", [outdir:"../lib"])'
unimplemented functionality in integer_list(9007474139725823, 31, _404977)
WARNING: Directive failed or aborted in file ic.ecl, line 98:
:- reexport ic_constraints.
unimplemented functionality in integer_list(9007474139725823, 31, _379092)
WARNING: Directive failed or aborted in file ic_search.ecl, line 59:
:- use_module(ic_constraints).
make[2]: Nothing to be done for `objects'.
LD_LIBRARY_PATH=/lib: ../bin/x86_64_macosx/eclipse -e 'get_flag(library_path,P),set_flag(library_path,["."|P]),lib(fcompile), set_flag(variable_names,off), fcompile("ic", [outdir:"../lib"])'
unimplemented functionality in integer_list(9007474139725823, 31, _409385)
WARNING: Directive failed or aborted in file ic.ecl, line 98:
:- reexport ic_constraints.
unimplemented functionality in integer_list(9007474139725823, 31, _380596)
WARNING: Directive failed or aborted in file ic_search.ecl, line 59:
:- use_module(ic_constraints).
make[2]: Nothing to be done for `objects'.
LD_LIBRARY_PATH=/lib: ../bin/x86_64_macosx/eclipse -e 'get_flag(library_path,P),set_flag(library_path,["."|P]),lib(fcompile), set_flag(variable_names,off), fcompile("ic_kernel", [outdir:"../lib"])'
make[2]: Nothing to be done for `objects'.
LD_LIBRARY_PATH=/lib: ../bin/x86_64_macosx/eclipse -e 'get_flag(library_path,P),set_flag(library_path,["."|P]),lib(fcompile), set_flag(variable_names,off), fcompile("ic_constraints", [outdir:"../lib"])'
unimplemented functionality in integer_list(9007474139725823, 31, _139956)
abort
make[1]: *** [../lib/ic_constraints.eco] Error 2
rm ../doc/examples/tutorial/.stamp ../doc/examples/ech/.stamp
make: *** [make_icparc_solvers] Error 2


Trying to execute the generated binary:


15:20 $ cd bin
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37/bin 
15:20 $ ll
total 0
drwxr-sr-x  5 pmoura  staff  160 Feb  3 15:18 x86_64_macosx
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37/bin 
15:20 $ cd x86_64_macosx/
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37/bin/x86_64_macosx 
15:20 $ ll
total 24
-rwxr-xr-x  1 pmoura  staff  404 Feb  3 15:18 eclipse
-rwxr-xr-x  1 pmoura  staff  631 Feb  3 15:18 tkeclipse
-rwxr-xr-x  1 pmoura  staff  590 Feb  3 15:18 tktools
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37/bin/x86_64_macosx 
15:20 $ ./eclipse 
ECLiPSe Constraint Logic Programming System [kernel threads]
Kernel and basic libraries copyright Cisco Systems, Inc.
and subject to the Cisco-style Mozilla Public Licence 1.1
(see legal/cmpl.txt or http://eclipseclp.org/licence)
Source available at www.sourceforge.org/projects/eclipse-clp
For other libraries see their individual copyright notices
Version 7.0 #37 (x86_64_macosx), Sat Feb  3 15:17 2018
[eclipse 67]: ^D
bye
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37/bin/x86_64_macosx 
15:20 $ export PATH=.:$PATH
✔ ~/Documents/Prolog/eclipse/Eclipse_7.0_37/bin/x86_64_macosx 
15:20 $ eclipselgt
unimplemented functionality in integer_list(3691512192, 31, _235256)
ERROR: /opt/local/share/logtalk-3.14.1-rc1/integration/logtalk_eclipse.pl:27:
  Query exited (abort): :- compile($LOGTALKHOME/integration/logtalk_comp_eclipse.pl)
Error(s) occurred while compiling /opt/local/share/logtalk-3.14.1-rc1/integration/logtalk_eclipse.pl
Aborting execution ...
abort


No idea how to fix the "unimplemented functionality in integer_list" errors that I get during build or when trying to use the binary. Any hints?

Cheers,
Paulo

-----------------------------------------------------------------
Paulo Moura
Logtalk developer

Re: [eclipse-clp-users] Loading an .asm file

[Paulo M. <pm...@lo...>]

Hi Joachim,

> On 3 Feb 2018, at 11:20, Joachim Schimpf <jsc...@co...> wrote:
> 
> On 02/02/18 14:39, Paulo Moura wrote:
>> Hi,
>> First of all, congratulations on the release of ECLiPSe 7.0!
>> I wrote a Bash shell script that generates a logtalk.asm file with the Logtalk compiler and runtime:
>> https://github.com/LogtalkDotOrg/logtalk3/tree/master/scripts/embedding/eclipse
>> ECliPSe documentation state that .asm files can be loaded by ECLiPSe but any attempts result in loading errors. What's the correct way of loading an .asm file?
> 
> 
> An .asm file can be loaded using any of ensure_loaded/1, compile/1,2
> or [file].  Or you can load it using the -f option when starting eclipse.
> 
> The only potential problem I can see in your setup is that maybe
> you are creating the .asm file with ISO-Syntax enabled, but trying
> to load it without?  This could result in quoting issues.
> 
> 
> Apart from that, if your intention is to pre-compile your application
> to speed up loading, you should use .eco format rather than .asm.
> This stores the result after the assembler has been run, and loads
> even more quickly.  There should be no drawbacks compared to .asm.
> 
> To create an .eco file, use compile(my_file, [output:eco]).
> This too can be loaded using any of the alternatives above.

Changed the script to create a .eco file instead. Thanks for the feedback.

>> Thanks,
>> Paulo
>> P.S. Any ETA for ECLiPSe 7.0 macOS binaries?
> 
> As soon as a macOS user takes an interest in creating them, I suppose!

I tried it but didn't get very far. Will discuss in a separate thread.

Cheers,
Paulo

-----------------------------------------------------------------
Paulo Moura
Logtalk developer

Re: [eclipse-clp-users] Loading an .asm file

[Joachim S. <jsc...@co...>]

On 02/02/18 14:39, Paulo Moura wrote:
> Hi,
> 
> First of all, congratulations on the release of ECLiPSe 7.0!
> 
> I wrote a Bash shell script that generates a logtalk.asm file with the Logtalk compiler and runtime:
> 
> https://github.com/LogtalkDotOrg/logtalk3/tree/master/scripts/embedding/eclipse
> 
> ECliPSe documentation state that .asm files can be loaded by ECLiPSe but any attempts result in loading errors. What's the correct way of loading an .asm file?


An .asm file can be loaded using any of ensure_loaded/1, compile/1,2
or [file].  Or you can load it using the -f option when starting eclipse.

The only potential problem I can see in your setup is that maybe
you are creating the .asm file with ISO-Syntax enabled, but trying
to load it without?  This could result in quoting issues.


Apart from that, if your intention is to pre-compile your application
to speed up loading, you should use .eco format rather than .asm.
This stores the result after the assembler has been run, and loads
even more quickly.  There should be no drawbacks compared to .asm.

To create an .eco file, use compile(my_file, [output:eco]).
This too can be loaded using any of the alternatives above.


> 
> Thanks,
> Paulo
> 
> P.S. Any ETA for ECLiPSe 7.0 macOS binaries?

As soon as a macOS user takes an interest in creating them, I suppose!


Cheers,
Joachim

[eclipse-clp-users] Loading an .asm file

[Paulo M. <pm...@lo...>]

Hi,

First of all, congratulations on the release of ECLiPSe 7.0!

I wrote a Bash shell script that generates a logtalk.asm file with the Logtalk compiler and runtime:

https://github.com/LogtalkDotOrg/logtalk3/tree/master/scripts/embedding/eclipse

ECliPSe documentation state that .asm files can be loaded by ECLiPSe but any attempts result in loading errors. What's the correct way of loading an .asm file?

Thanks,
Paulo

P.S. Any ETA for ECLiPSe 7.0 macOS binaries?

-----------------------------------------------------------------
Paulo Moura
Logtalk developer

[eclipse-clp-users] ECLiPSe 7.0

[Joachim S. <jsc...@co...>]

Dear ECLiPSe Users,

Happy New Year!

You may have noticed that ECLiPSe 7.0 has been available for
a while from our web site http://eclipseclp.org and mirrors.

This release contains significant changes "under the hood"
to support the new multi-engine/multi-thread functionality,
which enables a number of new programming techniques such as
multithreaded servers and working with multiple independent
search trees.

Beyond that, there are other new features and libraries too
numerous to list here, so please refer to the release notes at
http://eclipseclp.org/relnotes/rel70.html

If you are upgrading from an earlier version, please check the
section on potential incompatibilities at the end of the notes.


Those who are building ECLiPSe from the sources will want to
know that we have recently migrated our source repository from
CVS to GIT version control.  For the time being, the reference
repository remains hosted at Sourceforge and can be found at
https://sourceforge.net/p/eclipse-clp/git


Best regards,
Joachim Schimpf

[eclipse-clp-users] minimax problems

[<ms...@an...>]

I'm trying to solve a minimax problem - given an objective function of
some variables X_i and Y_i, find values of the X_i to minimize the
maximum of the objective given those X_i and any Y_i.  I expect to be able
to get reasonable results on this in a fairly straightforward way with
ic and branch_and_bound, simply nesting the max inside the min.  But is
there anything specific I should be watching out for?  Non-obvious
techniques specific to this problem that make it work better?  Previous
work on minimax optimization in ECLiPSe, or anything like that?
-- 
Matthew Skala
ms...@an...                 People before principles.
http://ansuz.sooke.bc.ca/

[eclipse-clp-users] Does Ciao have plan to support unicode atom like SWI-Prolog?

[Arthur W. <rth...@gm...>]

Re: [eclipse-clp-users] How to get comments inside of a clause?

[Joachim S. <jsc...@co...>]

On 06/10/17 14:51, Arthur Wang wrote:
> Hi there,
> 
> In code formatting of  VSC-Prolog, I can handle comments between clauses but 
> it's buggy to keep comments inside of a clause. I use regex in node.js to find 
> comments inside of a clause because I have not found any predicates to get 
> information about inside comments. The problem of regex is that it's hardly 
> possible to distinguish '%' in printf with that '%' of line comment.

In principle, regular expressions should be able to tell what is inside and what 
is outside of a string.


 > Are there any mechanisms to get inside comments in ECLiPSe like SWI-Prolog?

No, there is no support for reading such comments.


-- Joachim

[eclipse-clp-users] How to get comments inside of a clause?

[Arthur W. <rth...@gm...>]

Hi there,

In code formatting of  VSC-Prolog, I can handle comments between clauses 
but it's buggy to keep comments inside of a clause. I use regex in 
node.js to find comments inside of a clause because I have not found any 
predicates to get information about inside comments. The problem of 
regex is that it's hardly possible to distinguish '%' in printf with 
that '%' of line comment. Are there any mechanisms to get inside 
comments in ECLiPSe like SWI-Prolog?

Any helps are greatly appreciated!

Arthur Wang

Re: [eclipse-clp-users] vsc-prolog (VS Code extension) supports eclipseclp now

[Joachim S. <jsc...@co...>]

Hi Arthur,

On 17/09/17 16:53, Arthur Wang wrote:
> Hi all,
> 
> Several weeks ago, I published an extension of VS Code to support prolog, 
> initially for swi-prolog. Now I have extended some features for eclipseclp. You 
> are welcome to have a try if you use VS Code and eclipseclp. Visit the 
> repository for details: https://github.com/arthwang/vsc-prolog__

Thank you very much for this contribution, I have added a link at
http://eclipseclp.org/software/index.html


> 
> By the way, I came across some difficulties when tried to implement advanced 
> features such as "find all references", "refactor predicates" and "debugger". I 
> have not found any equivalents of swi-prolog such as prolog_codewalk, 
> prolog_trace_interception in eclipseclp documents.

This is of course highly non-standard functionality, but I would
expect that ECLiPSe provides what you need, in some form or other.

For cross-referencing, have a look at library(xref), which reads a
source file and generates a call graph.  It is built on top of
the generic library(source_processor), which is also underlies
the ECLiPSe compiler and other source processing tools.

See:
http://eclipseclp.org/doc/bips/lib/xref/index.html
http://eclipseclp.org/doc/bips/lib/source_processor/index.html


It is also possible to intercept the tracer, although this is
less well documented.  Here is an example:

% Begin file my_tracer.ecl
:- import sepia_kernel.
my_trace_line_handler(_, trace_line{port:Port,frame:tf{goal:Goal}}) :-
         printf("My trace: port=%w, goal=%w%n", [Port,Goal]).
:- set_event_handler(252, my_trace_line_handler/2).
% End file my_tracer.ecl


Compile the above file and see the customized tracer output:

[eclipse 1]: [my_tracer].
my_tracer.ecl compiled 392 bytes in 0.00 seconds
Yes (0.08s cpu)

[eclipse 2]: trace((X is 3+4, writeln(x=X))).
My trace: port=call, goal=X is 3 + 4
My trace: port=exit, goal=7 is 3 + 4
My trace: port=call, goal=writeln(x = 7)
x = 7
My trace: port=exit, goal=writeln(x = 7)



> 
> Any suggestions and helps are greatly appreciated.
> 
> With best regards,
> 
> Arthur Wang


Cheers,
Joachim

[eclipse-clp-users] vsc-prolog (VS Code extension) supports eclipseclp now

[Arthur W. <rth...@gm...>]

Hi all,

Several weeks ago, I published an extension of VS Code to support 
prolog, initially for swi-prolog. Now I have extended some features for 
eclipseclp. You are welcome to have a try if you use VS Code and 
eclipseclp. Visit the repository for details: 
https://github.com/arthwang/vsc-prolog__

By the way, I came across some difficulties when tried to implement 
advanced features such as "find all references", "refactor predicates" 
and "debugger". I have not found any equivalents of swi-prolog such as 
prolog_codewalk, prolog_trace_interception in eclipseclp documents.

Any suggestions and helps are greatly appreciated.

With best regards,

Arthur Wang

Re: [eclipse-clp-users] Issue with exported structures

[Philippe de R. <ph...@fr...>]

I found a way:

:- compile('persons2').

:- import p2m.

main :-
	top(
		[
			person{name:philippe,age:53,ba:bank_account{bal:1000}}
		],
		PMBL),
		writeln(PMBL).


> Le 5 sept. 2017 à 19:55, Philippe de Rochambeau <ph...@fr...> a écrit :
> 
> 
> Hi,
> 
> let’s say, I have another .pl file with the following contents:
> 
> --------
> 
> :- compile('persons2').
> 
> :- import p2m.
> 
> top(
> 	[
> 		person{name:philippe,age:53,ba:bank_account{bal:1000}}
> 	],
> 	PMBL).
> 
> —————
> 
> How do I make ECLIPSe display PMBL’s value?
> 
> Best regards,
> 
> Philippe
> 
> 
>> Le 5 sept. 2017 à 11:47, Kish Shen <kis...@gm... <mailto:kis...@gm...>> a écrit :
>> 
>> Hi Philippe,
>> 
>> The module qualification :/2 is for predicates only. To use the other exported items, you need to import the module:
>> 
>> [eclipse 2]: import p2m.
>> 
>> Yes (0.00s cpu)
>> 
>> you can then use exported items from p2m unqualified:
>> 
>> [eclipse 6]: top([person{name:philippe,age:53,ba:bank_account{bal:1000}}],PMBL).
>> 
>> PMBL = [1000]
>> Yes (0.00s cpu)
>> 
>> Cheers,
>> 
>> Kish
>> 
>> 
>> 
>> 
>> 
>> On Tue, Sep 5, 2017 at 7:21 AM, Philippe de Rochambeau <ph...@fr... <mailto:ph...@fr...>> wrote:
>> Hello,
>> 
>> I’m having an issue with an exported structure:
>> 
>> When I run
>> 
>> 
>> p2m:top(
>> 	[
>> 		p2m:person{name:philippe,age:50,p2m:ba:bank_account{bal:1000}}
>> 	],PMBL).	
>> 
>> 
>> I get the below warnings.
>> 
>> Any help would be much appreciated.
>> 
>> Philippe
>> 
>> 
>> 
>> —————
>> 
>> WARNING: Unrecognized structure name in module eclipse in
>>     bank_account{bal : 1000}
>> WARNING: Unrecognized structure name in module eclipse in
>>     person{name : philippe, age : 53, p2m : ba : bank_account{bal : 1000}}
>> 
>> PMBL = []
>> Yes (0.00s cpu)
>> [eclipse 15]: p2m:top(
>> 	[
>> 		p2m:person{name:philippe,age:53,p2m:ba:bank_account{bal:1000}}
>> 	],PMBL).			
>> WARNING: Unrecognized structure name in module eclipse in
>>     bank_account{bal : 1000}
>> WARNING: Unrecognized structure name in module eclipse in
>>     person{name : philippe, age : 53, p2m : ba : bank_account{bal : 1000}}
>> 
>> PMBL = []
>> Yes (0.00s cpu)
>> 
>> 
>> -------------
>> 
>> ——————————— code ——————————————
>> 
>> 
>> /*
>> 	Find people whose balance is above a certain limit.
>> */
>> 
>> :- module(p2m).
>> :- export top/2.
>> :- export filterBalance/3.
>> :- export top/2.
>> :- lib(ic).
>> :- export struct(bank_account(bal)).
>> :- export struct(person(name,age,ba:bank_account)).
>> 
>> top(FamilyList, PersonsMatchingBalanceLimit) :-
>> 		BalanceLimit = 0,
>> 		filterBalance(BalanceLimit, FamilyList, Ys), PersonsMatchingBalanceLimit = Ys.
>> 		
>> filterBalance(BalanceLimit, Family, Ys) :-
>> 	(
>> 		% foreach person in family
>> 		foreach(Person,Family),
>> 		fromto(Ys,Ys2,Ys1,[]),
>> 		param(BalanceLimit)
>> 	do
>> 		(
>> 			person{ba:Bal} = Person,
>> 			arg(1, Bal, Bal2),
>> 			Bal2 #>= BalanceLimit -> Ys2=[Bal2|Ys1] ;
>> 			Ys2=Ys1
>> 		)	
>> 	).
>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org <http://slashdot.org/>! http://sdm.link/slashdot <http://sdm.link/slashdot>
>> _______________________________________________
>> ECLiPSe-CLP-Users mailing list
>> ECL...@li... <mailto:ECL...@li...>
>> https://lists.sourceforge.net/lists/listinfo/eclipse-clp-users <https://lists.sourceforge.net/lists/listinfo/eclipse-clp-users>
>> 
>> 
> 

Re: [eclipse-clp-users] Issue with exported structures

[Philippe de R. <ph...@fr...>]

Hi,

let’s say, I have another .pl file with the following contents:

--------

:- compile('persons2').

:- import p2m.

top(
	[
		person{name:philippe,age:53,ba:bank_account{bal:1000}}
	],
	PMBL).

—————

How do I make ECLIPSe display PMBL’s value?

Best regards,

Philippe


> Le 5 sept. 2017 à 11:47, Kish Shen <kis...@gm...> a écrit :
> 
> Hi Philippe,
> 
> The module qualification :/2 is for predicates only. To use the other exported items, you need to import the module:
> 
> [eclipse 2]: import p2m.
> 
> Yes (0.00s cpu)
> 
> you can then use exported items from p2m unqualified:
> 
> [eclipse 6]: top([person{name:philippe,age:53,ba:bank_account{bal:1000}}],PMBL).
> 
> PMBL = [1000]
> Yes (0.00s cpu)
> 
> Cheers,
> 
> Kish
> 
> 
> 
> 
> 
> On Tue, Sep 5, 2017 at 7:21 AM, Philippe de Rochambeau <ph...@fr... <mailto:ph...@fr...>> wrote:
> Hello,
> 
> I’m having an issue with an exported structure:
> 
> When I run
> 
> 
> p2m:top(
> 	[
> 		p2m:person{name:philippe,age:50,p2m:ba:bank_account{bal:1000}}
> 	],PMBL).	
> 
> 
> I get the below warnings.
> 
> Any help would be much appreciated.
> 
> Philippe
> 
> 
> 
> —————
> 
> WARNING: Unrecognized structure name in module eclipse in
>     bank_account{bal : 1000}
> WARNING: Unrecognized structure name in module eclipse in
>     person{name : philippe, age : 53, p2m : ba : bank_account{bal : 1000}}
> 
> PMBL = []
> Yes (0.00s cpu)
> [eclipse 15]: p2m:top(
> 	[
> 		p2m:person{name:philippe,age:53,p2m:ba:bank_account{bal:1000}}
> 	],PMBL).			
> WARNING: Unrecognized structure name in module eclipse in
>     bank_account{bal : 1000}
> WARNING: Unrecognized structure name in module eclipse in
>     person{name : philippe, age : 53, p2m : ba : bank_account{bal : 1000}}
> 
> PMBL = []
> Yes (0.00s cpu)
> 
> 
> -------------
> 
> ——————————— code ——————————————
> 
> 
> /*
> 	Find people whose balance is above a certain limit.
> */
> 
> :- module(p2m).
> :- export top/2.
> :- export filterBalance/3.
> :- export top/2.
> :- lib(ic).
> :- export struct(bank_account(bal)).
> :- export struct(person(name,age,ba:bank_account)).
> 
> top(FamilyList, PersonsMatchingBalanceLimit) :-
> 		BalanceLimit = 0,
> 		filterBalance(BalanceLimit, FamilyList, Ys), PersonsMatchingBalanceLimit = Ys.
> 		
> filterBalance(BalanceLimit, Family, Ys) :-
> 	(
> 		% foreach person in family
> 		foreach(Person,Family),
> 		fromto(Ys,Ys2,Ys1,[]),
> 		param(BalanceLimit)
> 	do
> 		(
> 			person{ba:Bal} = Person,
> 			arg(1, Bal, Bal2),
> 			Bal2 #>= BalanceLimit -> Ys2=[Bal2|Ys1] ;
> 			Ys2=Ys1
> 		)	
> 	).
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot <http://sdm.link/slashdot>
> _______________________________________________
> ECLiPSe-CLP-Users mailing list
> ECL...@li... <mailto:ECL...@li...>
> https://lists.sourceforge.net/lists/listinfo/eclipse-clp-users <https://lists.sourceforge.net/lists/listinfo/eclipse-clp-users>
> 
> 

[eclipse-clp-users] Fwd: Issue with exported structures

[Kish S. <kis...@gm...>]

---------- Forwarded message ----------
From: Kish Shen <kis...@gm...>
Date: Tue, Sep 5, 2017 at 10:47 AM
Subject: Re: [eclipse-clp-users] Issue with exported structures
To: Philippe de Rochambeau <ph...@fr...>


Hi Philippe,

The module qualification :/2 is for predicates only. To use the other
exported items, you need to import the module:

[eclipse 2]: import p2m.

Yes (0.00s cpu)

you can then use exported items from p2m unqualified:

[eclipse 6]: top([person{name:philippe,age:53,ba:bank_account{bal:1000}}]
,PMBL).

PMBL = [1000]
Yes (0.00s cpu)

Cheers,

Kish





On Tue, Sep 5, 2017 at 7:21 AM, Philippe de Rochambeau <ph...@fr...>
wrote:

> Hello,
>
> I’m having an issue with an exported structure:
>
> When I run
>
>
> p2m:top(
> [
> p2m:person{name:philippe,age:50,p2m:ba:bank_account{bal:1000}}
> ],PMBL).
>
>
> I get the below warnings.
>
> Any help would be much appreciated.
>
> Philippe
>
>
>
> —————
>
> WARNING: Unrecognized structure name in module eclipse in
>     bank_account{bal : 1000}
> WARNING: Unrecognized structure name in module eclipse in
>     person{name : philippe, age : 53, p2m : ba : bank_account{bal : 1000}}
>
> PMBL = []
> Yes (0.00s cpu)
> [eclipse 15]: p2m:top(
> [
> p2m:person{name:philippe,age:53,p2m:ba:bank_account{bal:1000}}
> ],PMBL).
> WARNING: Unrecognized structure name in module eclipse in
>     bank_account{bal : 1000}
> WARNING: Unrecognized structure name in module eclipse in
>     person{name : philippe, age : 53, p2m : ba : bank_account{bal : 1000}}
>
> PMBL = []
> Yes (0.00s cpu)
>
>
> -------------
>
> ——————————— code ——————————————
>
>
> /*
> Find people whose balance is above a certain limit.
> */
>
> :- module(p2m).
> :- export top/2.
> :- export filterBalance/3.
> :- export top/2.
> :- lib(ic).
> :- export struct(bank_account(bal)).
> :- export struct(person(name,age,ba:bank_account)).
>
> top(FamilyList, PersonsMatchingBalanceLimit) :-
> BalanceLimit = 0,
> filterBalance(BalanceLimit, FamilyList, Ys), PersonsMatchingBalanceLimit =
> Ys.
> filterBalance(BalanceLimit, Family, Ys) :-
> (
> % foreach person in family
> foreach(Person,Family),
> fromto(Ys,Ys2,Ys1,[]),
> param(BalanceLimit)
> do
> (
> person{ba:Bal} = Person,
> arg(1, Bal, Bal2),
> Bal2 #>= BalanceLimit -> Ys2=[Bal2|Ys1] ;
> Ys2=Ys1
> )
> ).
>
>
>
>
> ------------------------------------------------------------
> ------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> ECLiPSe-CLP-Users mailing list
> ECL...@li...
> https://lists.sourceforge.net/lists/listinfo/eclipse-clp-users
>
>