open-axiom-help — Discussions about uses of OpenAxiom

[open-axiom-help] C++ API

[Mario C. <mca...@en...>]

Good Day Open Axiom Users!

My name is Mario. I am an engineer / software developer and I have recently discovered OpenAxiom. I would like to remove the GUI component from OpenAxiom and write C++ calls to the code undernearth so that I might embed certain CAS functions in my application. Right now, I am mostly interested in the ability to write a C++ call to the part of the code that solves systems of equations.

If there is any information anyone can give me to help me get started in this direction, I would greatly appreciate it. Any work that I do I gladly would offer up to the community as part of the project. Essentially, I would like to turn OpenAxiom into a C++ library in the same manner that SymPy is the library for Python users.


Sincerely,
Mario Capuozzo

[open-axiom-help] don't start

[Blatrix <bl...@la...>]

Hello,

Excuse me for my language but I'm french.

I tried to compile open-axiom under slackware 14.2, slackware64 14.2 and
slackware64 14.1.
The compilation is made with clisp; with sbcl I have an error :
package-locks.
 The compilation is complete with no error but when I start open-axiom I
don't have cursor, I can't enter command. In ps ax I can see the other
programs run (session, clef, spadclient, viewman) but I don't have the gui.

If I lauch gui with argument -, gui start and I can enter command but I
have no graphics possible......

What's wrong?

Thank you for your help,

Yann

Re: [open-axiom-help] Problem compiling open-axiom on Ubuntu

[Gabriel D. R. <gd...@in...>]

Thanks for the report, Doug.

OpenAxiom has not been ported to QT5.  I will look into that.

-- Gaby

On Tue, Sep 30, 2014 at 12:25 PM, T.D. Telford <do...@do...> wrote:

> My first email was evidently too large (with config.log attached), so have
> deleted config.log
>
>   ----- Forwarded Message -----
>  *From:* T.D. Telford <do...@do...>
> *To:* "ope...@li..." <
> ope...@li...>
> *Sent:* Tuesday, September 30, 2014 1:02 PM
> *Subject:* Problem compiling open-axiom on Ubuntu
>
> The code compiled for about 20+ miniutes (4 processors @ 3 gHz), then got
> message
>
> ...
> ...
> echo timestamp > stamp
> 24 finished
> make[2]: Leaving directory `/home/doug/oa-build/src/graph'
> cd gui && make
> make[2]: Entering directory `/home/doug/oa-build/src/gui'
> g++ -c -m64 -pipe -std=c++11 -O2 -Wall -W -D_REENTRANT -fPIE -DQT_NO_DEBUG
> -DQT_GUI_LIB -DQT_CORE_LIB
> -I/usr/lib/x86_64-linux-gnu/qt5/mkspecs/linux-g++-64 -I.
> -I../../x86_64-unknown-linux-gnu/include -I../../../oa-trunk.svn/src/gui
> -I../../../oa-trunk.svn/src/include -I../../config -I/usr/include/qt5
> -I/usr/include/qt5/QtGui -I/usr/include/qt5/QtCore -I. -o server.o
> ../../../oa-trunk.svn/src/gui/server.cc
> ../../../oa-trunk.svn/src/gui/server.cc:
>
> In member function ‘void OpenAxiom::Server::input(const QString&)’:
> ../../../oa-trunk.svn/src/gui/server.cc:56:15: error: ‘const class
> QString’ has no member named ‘toAscii’
>        write(s.toAscii());
>                ^
> make[2]: *** [server.o] Error 1
> make[2]: Leaving directory `/home/doug/oa-build/src/gui'
> make[1]: *** [all-gui] Error 2
> make[1]: Leaving directory `/home/doug/oa-build/src'
> make: *** [all-local] Error 2
>
>
>
> I installed libgd2-xpm-dev and qt-sdk.
> I am using sbcl 1.1.14 and g++ 4.8.2 on the latest version of Ubuntu,
> 14.04.
>
> The config.log is not attached.
>
> Any suggestions?
>
> Regards,
> Doug
>
>
>
>
>
> ------------------------------------------------------------------------------
> Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer
> Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports
> Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper
> Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer
>
> http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk
> _______________________________________________
> open-axiom-help mailing list
> ope...@li...
> https://lists.sourceforge.net/lists/listinfo/open-axiom-help
>
>

[open-axiom-help] Problem compiling open-axiom on Ubuntu

[T.D. T. <do...@do...>]

My first email was evidently too large (with config.log attached), so have deleted config.log


----- Forwarded Message -----
From: T.D. Telford <do...@do...>
To: "ope...@li..." <ope...@li...> 
Sent: Tuesday, September 30, 2014 1:02 PM
Subject: Problem compiling open-axiom on Ubuntu
 


The code compiled for about 20+ miniutes (4 processors @ 3 gHz), then got message

...
...
echo timestamp > stamp
24 finished
make[2]: Leaving directory `/home/doug/oa-build/src/graph'
cd gui && make 
make[2]: Entering directory `/home/doug/oa-build/src/gui'
g++
 -c -m64 -pipe -std=c++11 -O2 -Wall -W -D_REENTRANT -fPIE -DQT_NO_DEBUG 
-DQT_GUI_LIB -DQT_CORE_LIB 
-I/usr/lib/x86_64-linux-gnu/qt5/mkspecs/linux-g++-64 -I. 
-I../../x86_64-unknown-linux-gnu/include -I../../../oa-trunk.svn/src/gui
 -I../../../oa-trunk.svn/src/include -I../../config -I/usr/include/qt5 
-I/usr/include/qt5/QtGui -I/usr/include/qt5/QtCore -I. -o
 server.o ../../../oa-trunk.svn/src/gui/server.cc
../../../oa-trunk.svn/src/gui/server.cc: 

In member function ‘void OpenAxiom::Server::input(const QString&)’:
../../../oa-trunk.svn/src/gui/server.cc:56:15: error: ‘const class QString’ has no member named ‘toAscii’
       write(s.toAscii());
               ^
make[2]: *** [server.o] Error 1
make[2]: Leaving directory `/home/doug/oa-build/src/gui'
make[1]: *** [all-gui] Error 2
make[1]: Leaving directory `/home/doug/oa-build/src'
make: *** [all-local] Error 2




I installed libgd2-xpm-dev and qt-sdk.
I am using sbcl 1.1.14 and g++ 4.8.2 on the latest version of Ubuntu, 14.04.

The config.log is not attached.

Any suggestions?

Regards,
Doug

Re: [open-axiom-help] New OpenAxiom in Debian

[Gabriel D. R. <gd...@in...>]

On Thu, Jan 2, 2014 at 1:40 PM, Игорь Пашев <pas...@gm...> wrote:
> It's also available for stable Debian (now it is "wheezy") through
> backports: http://packages.debian.org/wheezy-backports/open-axiom

Thanks!

-- Gaby

Re: [open-axiom-help] New OpenAxiom in Debian

[Игорь П. <pas...@gm...>]

It's also available for stable Debian (now it is "wheezy") through
backports: http://packages.debian.org/wheezy-backports/open-axiom

Re: [open-axiom-help] New OpenAxiom in Debian

[Gabriel D. R. <gd...@cs...>]

Игорь Пашев <pas...@gm...> writes:

| Hi!
| 
| I've prepared new package for Debian:
| 
| http://packages.debian.org/sid/open-axiom


Thank you, Igor!

-- Gaby

[open-axiom-help] New OpenAxiom in Debian

[Игорь П. <pas...@gm...>]

Hi!

I've prepared new package for Debian:

http://packages.debian.org/sid/open-axiom

Re: [open-axiom-help] Question on references between lists (beginner)

[Gabriel D. R. <gd...@cs...>]

Ralf Hemmecke <ra...@he...> writes:

| But the following seems to be a bit inconsistent.
| 
| (1) -> List Integer has Finite
| 
|    (1)  false
| 
| (2) -> List Integer has finiteAggregate
| 
|    (2)  true


  * Finite is a property of a domain, e.g. it says the domain has a finite 
    number of values.  IntegerMod 12 has Finite.

  * finiteAggregate is a property that indicates whether a given aggregate
    has a finite number of nodes.  Note that the domain itself may have
    a potentially infinite number of values (e.g. all data structures of
    type List Integer), but each value of that domain (e.g. a specific
    instance of List Integer) is guaranteed to have a finite number of nodes.

There is no obvious relationship between the two.

By the way, in OpenAxiom-1.5.0 the attribute finiteAggregate has been
replaced with the category  FiniteAggregate which exports, among other
things:

     #: % -> NonNegativeInteger

-- Gaby

Re: [open-axiom-help] HyperDoc crashes

[<Lis...@gm...>]

Hello,

> Clicking first on "draw(sin(..." (which works!) and then on "Arrow down"
> will crash open-axiom completely :(

Browse -> Operations -> yes
yields
(1) -> sh: 1: cannot create /usr/lib/open-axiom//algebra/target.txt.5688: Permission denied

Of course this can not work because my "normal user" is not allowed to write below /usr/lib/...

In case of custom-build rev2989 it works ($HOME/lib/...) but that's no so a good idea IMHO
because these "search results" target.txt.PID will never get deleted from there.
Better locate them under /tmp (as fricas does) if any.

Another crash is
Browse -> Attributes -> Complex
(1) -> top level syntax error
 
   >> System error:
   The variable G725 is unbound.

(1) ->

Thank you very much.

Regards,
Edi

Re: [open-axiom-help] HyperDoc crashes

[<Lis...@gm...>]

Hello,

> rev2989:
> BasicCommands -> Calculus -> Differentiate -> Continue

This works at least in open-axiom 1.4.1+svn~2626-2 (debian) but then
Examples -> Graphics -> Assorted Examples -> "Arrow down" on first example "draw(sin(x*y)..."
crashed (same for rev 2989).

Clicking first on "draw(sin(..." (which works!) and then on "Arrow down"
will crash open-axiom completely :(

Regards,
Edi

[open-axiom-help] HyperDoc crashes

[<Lis...@gm...>]

Hello,

rev2989:
BasicCommands -> Calculus -> Differentiate -> Continue
(1) ->  
   >> System error:
   The variable G724 is unbound.

(1) ->

and the process "open-axiom/x86_64-unknown-linux-gnu/1.5.0-2013-05-19/lib/hypertex" is gone.

$ hardening-check hypertex 
hypertex:
 Position Independent Executable: yes
 Stack protected: yes
 Fortify Source functions: yes (some protected functions found)
 Read-only relocations: yes
 Immediate binding: yes


sendto(12, "\\begin{page}{G724}{Differentiate"..., 913, 0, NULL, 0) = 913
sendto(12, "c\0\0\0", 4, 0, NULL, 0)    = 4
sendto(14, "\3\0\0\0", 4, 0, NULL, 0)   = 4
select(1024, [0 6 12 14], NULL, NULL, NULL) = 1 (in [14])
recvfrom(14, "\2\0\0\0", 4, 0, NULL, NULL) = 4
recvfrom(14, "\0\0\0\0", 4, 0, NULL, NULL) = 4
select(1024, [0 6 12 14], NULL, NULL, NULL) = 1 (in [12])
recvfrom(12, "\6\0\0\0", 4, 0, NULL, NULL) = 4
recvfrom(12, "\311\0\0\0", 4, 0, NULL, NULL) = 4
recvfrom(12, "(|htDoneButton| '|bcDifferentiat"..., 201, 0, NULL, NULL) = 201
--- SIGSEGV (Segmentation fault) @ 0 (0) ---
mprotect(0x10001e8000, 32768, PROT_READ|PROT_WRITE|PROT_EXEC) = 0
rt_sigreturn(0)                         = 34494
--- SIGILL (Illegal instruction) @ 0 (0) ---
rt_sigprocmask(SIG_BLOCK, NULL, [HUP INT QUIT USR2 PIPE ALRM TERM CHLD TSTP URG XCPU XFSZ VTALRM PROF WINCH IO], 8) = 0
rt_sigprocmask(SIG_BLOCK, [HUP INT QUIT USR2 PIPE ALRM TERM CHLD TSTP URG XCPU XFSZ VTALRM PROF WINCH IO], NULL, 8) = 0

Thank you very much.

Regards,
Edi

Re: [open-axiom-help] rev2971: build fails with ecl

[Gabriel D. R. <gd...@in...>]

On Sat, Jun 1, 2013 at 11:05 AM,  <Lis...@gm...> wrote:
> Hello,
>
>> That failure is mysterious; I can't reproduce it at revision 2971.
>
> You're using ecl 13.5.1.
>
> So I have installed ecl from debian experimental which is version 12.12.1.
> (stable/testing/unstable have 11.1.1)
> and it seems rev2975 it's working _with_ hardening-wrapper.
> I stopped after 15 min but that's way beyond the former failure.

I just finished an ECL-based build.  It took an hour and 15min on
my machine.  That is essentially 5 times slower than SBCL.

-- Gaby

Re: [open-axiom-help] rev2971: build fails with ecl

[Gabriel D. R. <gd...@in...>]

On Sat, Jun 1, 2013 at 11:05 AM,  <Lis...@gm...> wrote:
> Hello,
>
>> That failure is mysterious; I can't reproduce it at revision 2971.
>
> You're using ecl 13.5.1.
>
> So I have installed ecl from debian experimental which is version 12.12.1.
> (stable/testing/unstable have 11.1.1)
> and it seems rev2975 it's working _with_ hardening-wrapper.
> I stopped after 15 min but that's way beyond the former failure.

So, ECL has indeed improved; that is good to hear.
I suspect we need a note in the INSTALL file then.

Re: [open-axiom-help] rev2971: build fails with ecl

[<Lis...@gm...>]

Hello,

> That failure is mysterious; I can't reproduce it at revision 2971.

You're using ecl 13.5.1.

So I have installed ecl from debian experimental which is version 12.12.1.
(stable/testing/unstable have 11.1.1)
and it seems rev2975 it's working _with_ hardening-wrapper.
I stopped after 15 min but that's way beyond the former failure.

Regards,
Edi

Re: [open-axiom-help] rev2971: build fails with ecl

[Gabriel D. R. <gd...@in...>]

When building the algebra, OA optimizes at level 3 (see --optimize=3)
which requires slightly more time as we go over function definitions
several times before emitting codes.

-- Gaby

Re: [open-axiom-help] rev2971: build fails with ecl

[Gabriel D. R. <gd...@in...>]

On Sat, Jun 1, 2013 at 7:20 AM,  <Lis...@gm...> wrote:
> Hello,
>
>> Remembering former problems I removed the debian package "hardening-wrapper" but
>> the second try was not much better [2].
>
> 2973 was much better, it compilied at least 295 min then I stopped ;)

That failure is mysterious; I can't reproduce it at revision 2971.

> Considering it was still in "src/algebra" and the binaries "view2d, ..." will take at least 5 min
> it's a 5 hours build (maybe 6) for me.
> (axiom enters after 5 hours the tests and if Tim's "1:2.5" still holds complete build would take 12+ h
> whereas fricas is just 90 mins.)
>
> Does anybody have (relative) times for other lisps or is sbcl (2.5 h for OA) the fastest compiler anyway?

SBCL-based build takes between 18min and 20min on my linux machines.
CLisp-based build takes less than 2 hours.
CCL takes about 40 min.

Sometimes ago, I thought ECL has comparative buildtime as SBCL.

-- Gaby

and I wrote Juanjo about it, but it turns out that I confused with SCBL.

Re: [open-axiom-help] rev2971: build fails with ecl

[<Lis...@gm...>]

Hello,

> Remembering former problems I removed the debian package "hardening-wrapper" but
> the second try was not much better [2].

2973 was much better, it compilied at least 295 min then I stopped ;)
Considering it was still in "src/algebra" and the binaries "view2d, ..." will take at least 5 min
it's a 5 hours build (maybe 6) for me.
(axiom enters after 5 hours the tests and if Tim's "1:2.5" still holds complete build would take 12+ h
whereas fricas is just 90 mins.)

Does anybody have (relative) times for other lisps or is sbcl (2.5 h for OA) the fastest compiler anyway?

Thank you very much.

Regards,
Edi

[open-axiom-help] rev2971: build fails with ecl

[<Lis...@gm...>]

Hello,

reading the changelog for 2971 I thought why not try out ecl? :)

But the build went not far [1] :(
Remembering former problems I removed the debian package "hardening-wrapper" but
the second try was not much better [2].

Thank you very much.

Regards,
Edi

$ dpkg -l | grep ecl
ii  ecl                                   11.1.1+dfsg1-2                     amd64        Embeddable Common-Lisp: has an interpreter and can compile to C

[1]
;;;   Invoking external command:
;;;   gcc -o core.fas -L/usr/lib/ /tmp/eclinitX9E7Tg.o core.o -Wl,--rpath,/usr/lib/ -shared -lecl -lgmp -lgc -lpthread -ldl -lm mkdir -p ../../x86_64-unknown-linux-gnu/lisp/
cp core.o ../../x86_64-unknown-linux-gnu/lisp/core.o
rm -f ../../x86_64-unknown-linux-gnu/lisp/linkset
echo '("core.o")' > ../../x86_64-unknown-linux-gnu/lisp/linkset
ecl \
                -norc -eval '(load "core")' \
                -norc -eval '(|AxiomCore|::|link| "base-lisp" (quote ("core.o")) "|AxiomCore|::|topLevel|")'
;;; Loading #P"/home/edi/oa-build-rev2971/src/lisp/core.fas"
;;; Note:
;;;   Invoking external command:
;;;   gcc -I. -I/usr/include/ -D_GNU_SOURCE -D_FILE_OFFSET_BITS=64 -g -O2 -fPIC -D_THREAD_SAFE -Dlinux -O2 -w -c /tmp/eclinitKtjvq3.c -o /tmp/eclinitKtjvq3.o 
;;; Note:
;;;   Invoking external command:
;;;   g++ -o base-lisp -L/usr/lib/ /tmp/eclinitKtjvq3.o core.o -lutil -lm -Wl,--rpath,/usr/lib/ -lecl -lgmp -lgc -lpthread -ldl -lm -m64 mkdir -p ../../build/x86_64-unknown-linux-gnu/bin
cp -p base-lisp ../../build/x86_64-unknown-linux-gnu/bin/lisp && cp -p base-lisp lisp
echo timestamp > stamp
make[2]: Leaving directory `/home/edi/oa-build-rev2971/src/lisp'
cd boot && make  all-boot
make[2]: Entering directory `/home/edi/oa-build-rev2971/src/boot'
mkdir -p strap
echo timestamp > strap/.started
../../src/driver/open-axiom --execpath=../lisp/lisp --output=strap/utility.o --compile --load-directory=strap ../../../data/oa-2971.svn/src/boot/strap/utility.clisp
make[2]: *** [strap/utility.o] Speicherzugriffsfehler
make[2]: Leaving directory `/home/edi/oa-build-rev2971/src/boot'
make[1]: *** [all-boot] Fehler 2
make[1]: Leaving directory `/home/edi/oa-build-rev2971/src'
make: *** [all-local] Fehler 2

real    1m49.390s
user    1m33.750s
sys     0m8.241s
edi@host:~/oa-build-rev2971$

[2]
libtool: compile:  g++ -c -m64 -D_GNU_SOURCE -std=c++11 -g -O2 -O2 -Wall -std=c++11 -g -O2 -O2 -Wall -DOPENAXIOM_ROOT_DIRECTORY=\"/home/edi/lib/open-axiom/x86_64-unknown-linux-gnu/1.5.0-2013-05-19\" -I../../../data/oa-2971.svn/src/include -I../../config ../../../data/oa-2971.svn/src/driver/main.cc -o main.o >/dev/null 2>&1
../../libtool --tag=CXX --mode=link g++ -static -m64 -all-static -o open-axiom main.lo -L./../utils -L./../lib -lOpenAxiom -lopen-axiom-core -lutil   -lm -lm
libtool: link: g++ -m64 -static -o open-axiom .libs/main.o  -L./../utils -L./../lib -lOpenAxiom -lopen-axiom-core -lutil -lm 
rm -f stamp
echo timestamp > stamp
make[2]: Leaving directory `/home/edi/oa-build-rev2971n/src/driver'
cd lisp && make  all-lisp
make[2]: Entering directory `/home/edi/oa-build-rev2971n/src/lisp'
oa_enable_lisp_profiling= nil
sed -e 's|@open_axiom_installdir[@]|/home/edi/lib/open-axiom/x86_64-unknown-linux-gnu/1.5.0-2013-05-19|g' -e 's|@oa_optimize_options[@]|speed|g' -e 's|@CXX[@]|g++|g' -e 's|@LDFLAGS[@]|-m64|g' -e 's|@oa_editor[@]|/usr/bin/vi|g' -e 's/@oa_keep_files[@]//g' -e 's|@host[@]|x86_64-unknown-linux-gnu|g' -e 's|@build[@]|x86_64-unknown-linux-gnu|g' -e 's|@target[@]|x86_64-unknown-linux-gnu|g' -e 's|@SHREXT[@]|.so|g' -e 's|@LIBEXT[@]|a|g' -e 's|@oa_c_runtime_extra[@]|"-lutil" "-lm"|g' -e 's|@oa_standard_linking[@]|yes|g' -e 's|@oa_enable_profiling[@]|nil|g' -e 's|@oa_delay_ffi[@]|no|g' -e 's|@void_type[@]|:void|g' -e 's|@char_type[@]|:char|g' -e 's|@int_type[@]|:int|g' -e 's|@float_type[@]|:float|g' -e 's|@double_type[@]|:double|g' -e 's|@string_type[@]|:cstring|g' ../../../data/oa-2971.svn/src/lisp/core.lisp.in > core.lisp.tmp
../../../data/oa-2971.svn/config/move-if-change core.lisp.tmp core.lisp
ecl  \
                -norc -eval '(require (quote cmp))' \
                -norc -eval '(compile-file "core.lisp" :system-p t)' \
                -norc -eval '(c::build-fasl "core.fas" :lisp-files (quote ("core.o")))' \
                -norc -eval '(quit)'
;;; Loading #P"/usr/lib/ecl-11.1.1/cmp.fas"
An error occurred during initialization:
12 is not of type LIST..
make[2]: *** [core.fas] Fehler 1
make[2]: Leaving directory `/home/edi/oa-build-rev2971n/src/lisp'
make[1]: *** [all-lisp] Fehler 2
make[1]: Leaving directory `/home/edi/oa-build-rev2971n/src'
make: *** [all-local] Fehler 2

real    1m28.598s
user    1m19.261s
sys     0m6.644s
edi@host:~/oa-build-rev2971n$

Re: [open-axiom-help] Backspace character

[<Lis...@gm...>]

Hello,

thanks to Waldek who showed me the light!

$ /usr/bin/rlwrap /home/edi/bin/open-axiom --no-server

Regards,
Edi

Re: [open-axiom-help] tutchap2.input: strange complexSolve behaviour in rev 2905

[<Lis...@gm...>]

Hello,

> The "Command-Cache-Problem" is really strange, since e.g.
> complexSolve(x^5=3,0.00001)
> works (cmd executed 2nd time same result)
> whereas
> complexSolve(x^3=3,0.00001)
> does not.

Oh, you fixed that cache-problem.
:-)

Thank you!

Regards,
Edi

[open-axiom-help] An ol' asin bug

[<Lis...@gm...>]

Hello,

digging in [1] I was wondering if
(1) -> asin(complex(1.0,0.0))
 
   >> Error detected within library code:
   catdef: division by zero
is still present (true in axiom & open-axiom).

In fricas I get
(1) -> asin(complex(1.0,0.0))

   (1)  1.5707963267_948966192
                                                         Type: Complex(Float)

but I couldn't figured out the difference since the definition of asin
asin x == atan(x/sqrt(1-x**2))
in "src/algebra/trigcat.spad.pamphlet" seems to be equal
so I guess it's the handling of "complex(float)" which leads to different results. (?)

Thank you very much.

Regards,
Edi

[1] https://savannah.nongnu.org/bugs/?10115

(1) -> asin(1.0)
   (1)  1.5707963267 948966192
                                                                  Type: Float
(2) -> asin(1)               
        %pi
   (2)  ---
         2
                                                     Type: Expression Integer
(3) -> asin(complex(1,0))
        %pi
   (3)  ---
         2
                                             Type: Expression Complex Integer
(4) ->

Re: [open-axiom-help] "complete 0" is 0 ?

[Gabriel D. R. <gd...@in...>]

On Sun, May 19, 2013 at 11:58 AM,  <Lis...@gm...> wrote:

> Is this related to the new results (revision 2905) from the Schur function?
>
> (1) -> SFunction [1,1]
>           1         1   3      2
>    (1)  - - (2 1) - - (1 ) + (1 )
>           2         2
>                                    Type: SymmetricPolynomial Fraction Integer
> (2) ->
>
> I would have expected as in fricas:
>
> (1) -> SFunction [1,1]
>
>           1       1   2
>    (1)  - - (2) + - (1 )
>           2       2
>                                  Type: SymmetricPolynomial(Fraction(Integer))

Fixed on trunk at revision 2937.

-- Gaby

Re: [open-axiom-help] rev2933: too much honour ;-)

[Gabriel D. R. <gd...@cs...>]

Lis...@gm... writes:

[...]

| createPrimitivePoly(2)$FFPOLY(gf2)
|  
|  
|    >> Error detected within library code:
|    (1 . failed) cannot be coerce to mode SparseUnivariatePolynomial
| #<(SIMPLE-VECTOR 42) {1004611E1F}>

These are all fixed.

-- Gaby

Re: [open-axiom-help] Scherk's Minimal Surface in Open-Axiom: Bug?

[Gabriel D. R. <gd...@in...>]

On Sat, May 25, 2013 at 1:14 PM,  <Lis...@gm...> wrote:

> Could you add "--verbose" to the "--help"-output-message?

Done at revision 2935.

-- Gaby