Re: [sage-support] Re: Error building 9.8 on Ubuntu 23.04

[Thierry Dumont]

Le 07/04/2023 à 23:53, John H Palmieri a écrit :
Some recent versions of Singular don't seem to work with Sage. You could 
try "make distclean" (to start over) and "./configure 
--with-system-singular=no" to force Sage to build its own Singular. Then 
"make".


Yes, I had the same problem on Debian Testing. BUT: remove all singular 
from your system befor (re)building Sage. Otherwise, you will encounter 
a problem again, as the building mechanism will first look at somme 
singular files (probably the dev one).


t.d.

--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-support/8158b189-6d52-0263-3122-239c8aa005fa%40math.univ-lyon1.fr.

Re: [sage-support] Re: Matrix_generic_dense.det() is quite slow compared to Numpy

[Thierry Dumont]

When doing numérical computations, you must use RDF floats, except for 
very special purpose. RDF is your processor's  float. The default Real 
Field, based on MPFR (which by default as the same precision as RDF) is 
slow as it is based on MPFR, a very good library, but MPFR is a software 
based library.
When you make elementary linear algebra computations (factorization 
eigenvalues, ...) with RDF floats, Sage uses the Lapack library based on 
OpenBlas (Basic linear algebra subroutine), which is optimized for your 
computer. These libraries use all your processor capacities (like the 
vectorization with avx(2) instructions).
It is extremely difficult to compute faster than Lapack + OpenBlas (it 
could be a bit faster with Intel own Blas, on Intels's processors...just 
a bit).

To see this just compute an LU factorization:

sage: A=random_matrix(RDF,1000,1000)
sage: %time Q=A.LU()
The cost for an LU factorization is n^3 for a matrix of size n.
On my old processor (3 Ghz, avx), it takes 53.2 ms, so factorization is 
computed with a speed  of about 18 Gigafloats, which is impressive.


 You must use RealField with a large precision (say RealField(prec=500))
if you want to solve very ill conditioned problems (example:  Hilbert 
matrix Aij= 1./(1.+i+j) with size n >20), but then you have to "pay" 
with a slow computation.


TINA (there is no alternative :-( ), sadly.

yours,
t.d.

Le 09/01/2023 à 06:19, John H Palmieri a écrit :
I am not at all an expert in numerical linear algebra, but (a) certainly 
parts of Sage can be improved and (b) contributions are welcome. Others 
can speak more knowledgeably about this particular issue.


On Sunday, January 8, 2023 at 11:45:00 AM UTC-8 ivana...@gmail.com wrote:

Is there any space to improve the computation of MPFR which is used
by RealField?

John H Palmieri 在 2023年1月6日 星期五凌晨3:01:37 [UTC+8] 的信中寫道：

One way to speed it up might be to work with RDF or QQ or RLF.
The documentation for the generic determinant method says, "Note
that for matrices over most rings, more sophisticated algorithms
can be used."

sage: %time ones_matrix(RDF, 600, 600).determinant()
CPU times: user 78.6 ms, sys: 7.6 ms, total: 86.2 ms
Wall time: 77.9 ms
0.0
sage: %time ones_matrix(QQ, 600, 600).determinant()
CPU times: user 280 ms, sys: 36.3 ms, total: 317 ms
Wall time: 325 ms
0
sage: %time ones_matrix(RLF, 600, 600).determinant()
CPU times: user 32.7 s, sys: 183 ms, total: 32.9 s
Wall time: 40.3 s
0


On Thursday, January 5, 2023 at 1:26:12 AM UTC-8
ivana...@gmail.com wrote:

I am doing a numerical analysis which needs to calculate
many determinants of "huge" dense matrices. The sizes of the
matrix are less than 100*100.

I found the computation of the determinant is quite slow
with Matrix_generic_dense. I followed the method on What is
the time complexity of numpy.linalg.det?


 and made minor modifications to the script to benchmark for Sagemath.
1. np.arange(1,10001, 100) => np.arange(1, 292, 10) because
the speed is too slow, I have to reduce the testing range to
make the script finish within a reasonable time.
2. np.ones((size, size))  => ones_matrix(RR, size, size)
3. timeit('np.linalg.det(A)', globals={'np':np, 'A':A},
number=1) => timeit('A.det(algorithm="df")', globals={'A':
A}, number=1)

Here is the full script I used.

from timeit import timeit

import matplotlib.pyplot as plt
import numpy as np
from sage.rings.real_mpfr import RR
from sklearn.linear_model import LinearRegression
from sklearn.preprocessing import PolynomialFeatures

sizes = np.arange(1, 292, 10)
times = []

for size in sizes:
     A = ones_matrix(RR, size, size)
     time = timeit('A.det(algorithm="df")', globals={'A':
A}, number=1)
     times.append(time)
     print(size, time)

sizes = sizes.reshape(-1, 1)
times = np.array(times).reshape(-1, 1)

quad_sizes = PolynomialFeatures(degree=2).fit_transform(sizes)
quad_times = LinearRegression().fit(quad_sizes,
times).predict(quad_sizes)
cubic_sizes = PolynomialFeatures(degree=3).fit_transform(sizes)
cubic_times = LinearRegression().fit(cubic_sizes,
times).predict(cubic_sizes)
quartic_sizes =
PolynomialFeatures(degree=4).fit_transform(sizes)
quartic_times = LinearRegression().fit(quartic_sizes,
  
 

Re: [sage-support] Compiling Sage on Ubuntu 20.05

[Thierry Dumont]

You should start by installing "build-essential" and gfortran. To do 
this you have to do:


sudo apt update
sudo apt install build-essential gfortran

sudo is mandatory (rivileged (root) access).

build-essential contains gcc and g++ compilers as well as some libraries 
and tools, but not the fortran compiler (gfortran).


You will certainly have to install other tools and libraries. Follow 
sage installation procedure, you will get the list of what too install.


Yours,
t.d.

Le 27/02/2022 à 12:22, Saksham a écrit :

Hi all,

I have been trying to install sage 9.5 from source on Ubuntu.

After running the following command i get an error.

CC=gcc-7 CXX=g++ FC=gfortran-7 ./configure

The error that i get is.

configure: error:

     Given --with-system-gcc=force, but no system package could be used.
     That's an error.  Please install the indicated package to continue.
     (To override this error, use ./configure --without-system-gcc)

Does anyone have a solution for this ? (using -without-system-gcc does 
fix this error but why could system package not be used)


Thanks,
Saksham.

--
You received this message because you are subscribed to the Google 
Groups "sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send 
an email to sage-support+unsubscr...@googlegroups.com 
.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-support/54776375-dc6c-46a5-91d8-ace3d5681ca0n%40googlegroups.com 
.


--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-support/a3d08d4e-a3a3-2787-8379-96e3cb4bceb0%40math.univ-lyon1.fr.

Re: [sage-support] much memory is used...

[Thierry Dumont]

Le 08/10/2018 à 12:56, Dima Pasechnik a écrit :
> On Mon, Oct 8, 2018 at 11:34 AM Thierry Dumont
>  wrote:
>>
>> I do:
>>
>> N= 10^8
>> theta= RDF(2*pi/N)
>> phi=[i*theta for i in range(0,N)]
>>
>> not something sophisticated...!
>>
>> Also: it is not a good idea do do that: better use numpy. BUT
>>
>> looking at "top" during the computation of
>> phi=[i*theta for i in range(0,N)],
>> the memory used jumps to 3Gb and then grows, grows... until it reaches
>> the limits of my 8GB machine.
>>
>> python lists are not known to be very efficient, ok. But this is a bit
>> too much, no ?
> 
> If you limit the memory to 4GB you can do the same out of memory stunt
> using plain python:
> 
> $ ulimit -Sv 400
> $ ./sage --python
> Python 2.7.15 (default, Sep 19 2018, 15:29:25)
> [GCC 7.3.1 20180130 (Red Hat 7.3.1-2)] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
>>>> N=10**8
>>>> theta=6.283185307179586e-08
>>>> phi=[i*theta for i in range(0,N)]
> Traceback (most recent call last):
>   File "", line 1, in 
> MemoryError
> 
> 
> Perhaps RDF takes more memory, I don't know - anyway, it's not
> Sage-specific IMHO.
> 
>>
RDF are machine double floats; that is to say C "double", which take 8
bytes.
So my list should use less than 1gb, if it phi was a C array.
Am I wrong ?


>> --
>> You received this message because you are subscribed to the Google Groups 
>> "sage-support" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to sage-support+unsubscr...@googlegroups.com.
>> To post to this group, send email to sage-support@googlegroups.com.
>> Visit this group at https://groups.google.com/group/sage-support.
>> For more options, visit https://groups.google.com/d/optout.
> 

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] much memory is used...

[Thierry Dumont]

I do:

N= 10^8
theta= RDF(2*pi/N)
phi=[i*theta for i in range(0,N)]

not something sophisticated...!

Also: it is not a good idea do do that: better use numpy. BUT

looking at "top" during the computation of
phi=[i*theta for i in range(0,N)],
the memory used jumps to 3Gb and then grows, grows... until it reaches
the limits of my 8GB machine.

python lists are not known to be very efficient, ok. But this is a bit
too much, no ?
yours
t.d.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] legendre_P and chebyshev_T : strange (for me).

[Thierry Dumont]

Let's do this:
sage: R = PolynomialRing(AA,"x")
sage: x = R.gen()
sage: n=3

Then:
sage: chebyshev_T(n,x).parent()
Univariate Polynomial Ring in x over Algebraic Real Field

Ok. But:

sage: legendre_P(n,x).parent()
Symbolic Ring

why that ?

Yours,

t.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] problems installing sage on ubuntu 18.04

[Thierry Dumont]

Le 27/07/2018 à 01:11, rudolf ordoyne a écrit :
> my guess is that i'm unable to install sage since the most recent
> release of sage is for 16.04.  after unpacking the .tar, running "sudo
> ./sage" gives "sudo: ./sage: command not found"
> 
> has anyone else been using ubuntu 18.04?
> 
Yes, without anay problem (unpack the tar, ./sage).
> -- 
> You received this message because you are subscribed to the Google
> Groups "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sage-support+unsubscr...@googlegroups.com
> .
> To post to this group, send email to sage-support@googlegroups.com
> .
> Visit this group at https://groups.google.com/group/sage-support.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: editor for sage commands

[Thierry Dumont]

Many people like me like emacs but in the "new generation" of developers
many (some?)  have switched to VisualStudio, which is now available on
Linux.
What about Sage and VisualStudio ? There is a python environment which
seems good:
https://www.visualstudio.com/vs/features/python/?rr=https%3A%2F%2Fwww.google.com%2F
(but I don't like vs!).
t.

Le 23/05/2018 à 16:11, Emmanuel Charpentier a écrit :
> Dear Francesco,
> 
> Le dimanche 20 mai 2018 18:57:54 UTC+2, Francesco a écrit :
> 
> I tried to use emacs with sage-shell-mode, but I have some
> difficulty to configure all things.
> 
> 
> What is (are) the problem(s) ?
>  
> 
> Furthermore I am a new user of emacs..
> 
> 
> Aha ! This may explain that : emacs itself as a learning curve (not
> unlike the Matterhorn ,
> according to some...). You may need to invest a bit in learning it.
> (This investment usually has a huge ROI...).
> 
> I strongly recommend that to give yourself a few hours to read the emacs
> tutorial (accessible from the "Help" menu of emacs) ;  keep in mind that
> some things, such as the use of control characters as movement keys,
> predate modern keyboards, are somewhat obsolete... Also, skim at least
> the table of contents of the emacs manual
> ,
> the introduction
> 
> to emacs lisp and the emacs lisp manual
>  ;
> n particular, try to familiarize yourself with the "Info mode", which is
> used to access the online help.
> 
> This will help you understanding *why* and *how* you may customize emacs
> to use sage-shell-mode. Learning to customize this customization will
> need a bit more of understanding emacs lisp. But at this point, things
> should ave become easy...
>  
> 
> Is there a clear guide of the operations to configure
> sage-shell-mode of emacs?
> 
> 
> Look at the sage-shell-mode github page
> . The README.md displayed
> there has a relatively extensive set of instructions for installation,
> customization and use, which require but the very basics of emacs use
> essentially the ability to edit your .emacs file)...
> 
> HTH,
> 
> --
> Emmanuel Charpentier
> 
> Il giorno sabato 19 maggio 2018 18:58:29 UTC+2, Francesco ha scritto:
> 
> Can you tell me a good editor for sage shell commands ? I try
> Texmacs, but it crashes.
> I have kubuntu 18.04 64bit and the version of sage in the
> repositories, the 8.1.
> 
> -- 
> You received this message because you are subscribed to the Google
> Groups "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sage-support+unsubscr...@googlegroups.com
> .
> To post to this group, send email to sage-support@googlegroups.com
> .
> Visit this group at https://groups.google.com/group/sage-support.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] OpenBlas multithreaded or not ?

[Thierry Dumont]

Hi,

I am doing some performances measurements  with Sage/scipy.
It's matrix x matrix multiplication (not a very new benchmark).

It appears that, on the same computer Sage version 8.0 uses Openblas
with multithreading, but not Sage version 8.1.beta3:



-Example (4 core, broadwell, 3.4ghz. Peak performance 217 Gflops !):
size of matrices: 640

  -> Sage version 8.0 :
  Gflops=  138.359946456  Portion peak perf =  0.6358  (top says %cpu=400)

  -> Sage version 8.1.beta3:
  Gflops=  48.5945679306  Portion peak perf =  0.2233 (%cpu=100)

Both of Sage versions where compiled from source.

My questions:
Will OpenBlas be fixed to 1 thread in the next versions of sage?

t.d.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
# Broadwell architecture, 4 core, 3.4 Ghz
# Each cpu as 2 vector ports (AVX), each one doing 4 vector operations
#y=a*x + b (that is: 8 flops)  by clock cycle.
ncore = 4
freq  = 3.4*10^9
vector_ports = 2
perf  = 8*vector_ports*ncore*freq
print "Peak perf, Gflops:",perf
for n in [10*2^i for i in [1..10]]:
c= random_matrix(RDF,n)
d= random_matrix(RDF,n)
t= timeit("c*d",seconds=True)
flops= 2*n^3/t
print n,t,"Gflops= ",flops/10^9," Portion peak perf = ",flops/perf
#

<>

Re: [sage-support] Compilation of 7.6.beta1 on Ubuntu 16-10

[Thierry Dumont]

Le 31/01/2017 à 12:30, Jeroen Demeyer a écrit :
> On 2017-01-31 12:24, Thierry Dumont wrote:
>> I just git clone sage, then checkout origin/develop and make.
>> Then after some time, Python 3 is installed and compiled.
> 
> Are you really sure about that? Did you set the SAGE_PYTHON3 environment
> variable?
> 
> Could you please send your logs/install.log for a single serial build
> such that we can see what might be wrong?
> 
Apologize...yes, the variable SAGE_PYTHON3 was set (from a preceeding
experience).
Thanks
t.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Compilation of 7.6.beta1 on Ubuntu 16-10

[Thierry Dumont]

Le 31/01/2017 à 11:49, Jeroen Demeyer a écrit :
> On 2017-01-31 09:42, Thierry Dumont wrote:
>> When trying to compile 7.6.beta1, which installed Python3, I got a
>> cython problem (cmp...):
> 
> That's not news... Nobody ever said that Sage works with Python 3.
> 
Yes, I know, but but why is Python 3 installed ?

I just git clone sage, then checkout origin/develop and make.
Then after some time, Python 3 is installed and compiled.
I started from scratch !

t.d.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] Compilation of 7.6.beta1 on Ubuntu 16-10

[Thierry Dumont]

When trying to compile 7.6.beta1, which installed Python3, I got a
cython problem (cmp...):

[sagelib-7.6.beta1] [353/465] Cythonizing sage/rings/morphism.pyx
[sagelib-7.6.beta1]
[sagelib-7.6.beta1] Error compiling Cython file:
[sagelib-7.6.beta1]

[sagelib-7.6.beta1] ...
[sagelib-7.6.beta1]
[sagelib-7.6.beta1] sage: Zmod(8).lift() == 1
[sagelib-7.6.beta1] False
[sagelib-7.6.beta1] """
[sagelib-7.6.beta1] if not isinstance(other, RingMap_lift):
[sagelib-7.6.beta1] return cmp(type(self), type(other))
[sagelib-7.6.beta1]  ^
[sagelib-7.6.beta1]

[sagelib-7.6.beta1]
[sagelib-7.6.beta1] sage/rings/morphism.pyx:525:22: undeclared name not
builtin: cmp
[sagelib-7.6.beta1] Traceback (most recent call last):
[sagelib-7.6.beta1]   File
"/usr/local/sage-develop/sage/local/lib/python3.5/site-packages/Cython/Build/Dependencies.py",
line 1074, in cythonize_one_helper
[sagelib-7.6.beta1] return cythonize_one(*m)
[sagelib-7.6.beta1]   File
"/usr/local/sage-develop/sage/local/lib/python3.5/site-packages/Cython/Build/Dependencies.py",
line 1056, in cythonize_one
[sagelib-7.6.beta1] raise CompileError(None, pyx_file)
[sagelib-7.6.beta1] Cython.Compiler.Errors.CompileError:
sage/rings/morphism.pyx
[sagelib-7.6.beta1] [354/465] Cythonizing
sage/rings/noncommutative_ideals.pyx
[sagelib-7.6.beta1]


t.d.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: Error building 7.5.1

[Thierry Dumont]

Le 17/01/2017 à 00:09, Volker Braun a écrit :
> You can force a particular architecture with 
> 
> OPENBLAS_CONFIGURE="TARGET=PRESCOTT" make -p openblas
>  
> 
Yes! it works, thanks.

Otherwise there are solutions when the vitual machine is run by KVM/libvirt:

1) if you add

  

in the definition of the vm, it will appear like the underlying machine,
2)
Define the type of cpu:

SandyBridge


yours
t.
> 
> On Monday, January 16, 2017 at 11:54:56 AM UTC+1, tdumont wrote:
> 
> 
> I tried to install 7.5.1 from source on an Ubuntu 14.04 (yes, I have no
> choice).
> This fails when compiling openblas-0.2.19.p0, when building
> libopenblas_banias-r0.2.19.a.
> 
> ar -ru ../libopenblas_banias-r0.2.19.a samax_k.o samin_k.o
> .
> .
> 
> ar: sgemm_kernel.o: No such file or directory.
> 
> In the root of sage, I then do:
> 
> find . -name "sgemm_kernel*"
> 
> and find:
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/power/sgemm_kernel_16x8_power8.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm64/sgemm_kernel_4x4.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm64/sgemm_kernel_8x8.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm64/sgemm_kernel_16x4.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/mips64/sgemm_kernel_loongson3b_4x4.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/mips64/sgemm_kernel_loongson3a_4x4.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/mips64/sgemm_kernel_8x4_ps.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/mips/sgemm_kernel_8x8_msa.c
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm/sgemm_kernel_4x4_vfpv3.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm/sgemm_kernel_4x2_vfp.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_16x4_haswell.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_8x8_sandy.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_8x4_bulldozer.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_16x2_piledriver.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_16x2_bulldozer.S
> 
> 
> ./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_16x4_sandy.S
> 
> 
> which are certainly different versions of this sgemm_kernel, for
> different architecture.
> 
> So, what happens ?
> Should it be that the architecture of the machine is not detected ?
> I am
> on a virtual machine.
> 
> Any idea ?
> 
> Yours
> 
> t.d.
> 
> -- 
> You received this message because you are subscribed to the Google
> Groups "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sage-support+unsubscr...@googlegroups.com
> .
> To post to this group, send email to sage-support@googlegroups.com
> .
> Visit this group at https://groups.google.com/group/sage-support.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] Error building 7.5.1

[Thierry Dumont]

I tried to install 7.5.1 from source on an Ubuntu 14.04 (yes, I have no
choice).
This fails when compiling openblas-0.2.19.p0, when building
libopenblas_banias-r0.2.19.a.

ar -ru ../libopenblas_banias-r0.2.19.a samax_k.o samin_k.o
.
.

ar: sgemm_kernel.o: No such file or directory.

In the root of sage, I then do:

find . -name "sgemm_kernel*"

and find:

./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/power/sgemm_kernel_16x8_power8.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm64/sgemm_kernel_4x4.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm64/sgemm_kernel_8x8.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm64/sgemm_kernel_16x4.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/mips64/sgemm_kernel_loongson3b_4x4.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/mips64/sgemm_kernel_loongson3a_4x4.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/mips64/sgemm_kernel_8x4_ps.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/mips/sgemm_kernel_8x8_msa.c
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm/sgemm_kernel_4x4_vfpv3.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/arm/sgemm_kernel_4x2_vfp.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_16x4_haswell.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_8x8_sandy.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_8x4_bulldozer.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_16x2_piledriver.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_16x2_bulldozer.S
./local/var/tmp/sage/build/openblas-0.2.19.p0/src/kernel/x86_64/sgemm_kernel_16x4_sandy.S

which are certainly different versions of this sgemm_kernel, for
different architecture.

So, what happens ?
Should it be that the architecture of the machine is not detected ? I am
on a virtual machine.

Any idea ?

Yours

t.d.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: Question about upgrading.

[Thierry Dumont]

Le 13/01/2017 à 11:09, Dima Pasechnik a écrit :
> IMHO you should be able to finish build by (repeatedly) running make.
> (It would in the end give you working Sage, although make will report an
> error in the end)
> 
> Just in case, coxeter3 installs in 7.5 just fine for me (on a gentoo
> linux box)
> 
True; sage -i -f coxeter3 solved the problem.
Thanks!
t.
> On Friday, January 13, 2017 at 8:40:10 AM UTC, tdumont wrote:
> 
> Today, I have tried to upgrade from 7.4 to 7.5 (git pull; make ...)
> 
> I have a problem with the optional package coxeter3 (it does not
> compile).
> 
> Before trying to solve this problem with coxeter, I would like to know
> if there is a canonical way to remove coxeter3 from sage tree to finish
> the build correctly. coxeter3 appears there:
> 
> ./upstream/coxeter3-1.1.tar.gz
> ./local/var/tmp/sage/build/coxeter3-1.1
> ./local/lib/python2.7/site-packages/sage/libs/coxeter3
> ./logs/pkgs/coxeter3-1.1.log
> ./build/pkgs/coxeter3
> ./src/sage/libs/coxeter3
> ./src/build/cythonized/sage/libs/coxeter3
> ./src/build/temp.linux-x86_64-2.7/sage/libs/coxeter3
> ./src/build/lib.linux-x86_64-2.7/sage/libs/coxeter3
> 
> Is is ok to remove all this before doing make again or is there
> something cleaner ?
> 
> t.
> 
> -- 
> You received this message because you are subscribed to the Google
> Groups "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sage-support+unsubscr...@googlegroups.com
> .
> To post to this group, send email to sage-support@googlegroups.com
> .
> Visit this group at https://groups.google.com/group/sage-support.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] Question about upgrading.

[Thierry Dumont]

Today, I have tried to upgrade from 7.4 to 7.5 (git pull; make ...)

I have a problem with the optional package coxeter3 (it does not compile).

Before trying to solve this problem with coxeter, I would like to know
if there is a canonical way to remove coxeter3 from sage tree to finish
the build correctly. coxeter3 appears there:

./upstream/coxeter3-1.1.tar.gz
./local/var/tmp/sage/build/coxeter3-1.1
./local/lib/python2.7/site-packages/sage/libs/coxeter3
./logs/pkgs/coxeter3-1.1.log
./build/pkgs/coxeter3
./src/sage/libs/coxeter3
./src/build/cythonized/sage/libs/coxeter3
./src/build/temp.linux-x86_64-2.7/sage/libs/coxeter3
./src/build/lib.linux-x86_64-2.7/sage/libs/coxeter3

Is is ok to remove all this before doing make again or is there
something cleaner ?

t.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: Stange behavior with QQbar[z]

[Thierry Dumont]

Le 08/11/2016 à 11:05, Vincent Delecroix a écrit :
> On 8 November 2016 at 10:17, Thierry Dumont <tdum...@math.univ-lyon1.fr> 
> wrote:
>> Le 08/11/2016 à 08:43, Vincent Delecroix a écrit :
>>> Concerning representation of algebraic numbers, it is printed as an
>>> exact rational if and only if it is stored as an exact rational. It
>>> will be if the method exactify has been called on the underlying
>>> representation of the number. Here is a simple example that shows the
>>> difference
>>>
>>
>> Ha, yes...
>>
>> But I am not sure to understand.
>>
>> sage: y=QQbar(cos(pi/18))
>> sage: y.radical_expression()
>> 1/4*(4*(1/128*I*sqrt(3) + 1/128)^(1/3) + 1)/(1/128*I*sqrt(3) + 1/128)^(1/6)
>>
>> ok! good!
>>
>> sage: y
>> 0.9848077530122081? + 0.?e-18*I
>>
>> ok.
>> sage: y.imag()
>> 0.?e-18
>> sage: y.imag() == 0
>> True
>> I accept this as 0 is "in" 0.?e-18
>>
>> Now:
>>
>> sage: y.exactify()
>> sage: y
>> 0.9848077530122081? + 0.?e-18*I
>>
>> raaahhh ! grrr !
> 
> This is *not* a rational!! 

sure...

>We might want to special case the
> representation of real numbers of QQbar. I opened
> https://trac.sagemath.org/ticket/21838 for that purpose.
> 
> Vincent
> 
In my code, all values are real in QQbar. So at least temporary, I wrote

realify=lambda x: x.real()

which is not as beautiful as could be dreamed, but it works.
Thanks!

t.d.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: Stange behavior with QQbar[z]

[Thierry Dumont]

Le 08/11/2016 à 08:43, Vincent Delecroix a écrit :
> Concerning representation of algebraic numbers, it is printed as an
> exact rational if and only if it is stored as an exact rational. It
> will be if the method exactify has been called on the underlying
> representation of the number. Here is a simple example that shows the
> difference
> 

Ha, yes...

But I am not sure to understand.

sage: y=QQbar(cos(pi/18))
sage: y.radical_expression()
1/4*(4*(1/128*I*sqrt(3) + 1/128)^(1/3) + 1)/(1/128*I*sqrt(3) + 1/128)^(1/6)

ok! good!

sage: y
0.9848077530122081? + 0.?e-18*I

ok.
sage: y.imag()
0.?e-18
sage: y.imag() == 0
True
I accept this as 0 is "in" 0.?e-18

Now:

sage: y.exactify()
sage: y
0.9848077530122081? + 0.?e-18*I

raaahhh ! grrr !

t.


> sage: a = QQbar(2).sqrt() + QQbar(3).sqrt()
> sage: b = a**2 - 2*QQbar(6).sqrt()
> 
> sage: b
> 5.000?
> sage: type(b._descr)# b is an formal sum
> 
> 
> sage: b == 5   # calls exactify
> True
> sage: b# now prints as 5...
> 5
> sage: type(b._descr)   # because it *is* an exact rational
> 
> 
> 
> Vincent
> 
> On 7 November 2016 at 17:28, Nils Bruin  wrote:
>> On Monday, November 7, 2016 at 5:59:14 AM UTC-8, tdumont wrote:
>>>
>>>
>>> okay. Now let us construct a polynomial from l:
>>> sage: print sum([l[i]*z^i for i in range(0,len(l))])
>>> (-0.4022786138875136? + 0.?e-18*I)*z + 1
>>>   ^
>>>Why? an imaginary part?
>>>--
>>
>> There are polynomials for which this doesn't happen. It would be better to
>> include enough information to reproduce your observations.
>>
>> The representation you see is a float approximation to the algebraic number.
>> You can approximate real numbers quite well with complex numbers with
>> non-zero imaginary part.
>>
>> sage: P.=QQbar[]
>> sage: u=[u for u in (x^3+x+1).roots(multiplicities=false) if u.imag()==0][0]
>> sage: u
>> -0.6823278038280193?
>> sage: v=(u*QQbar.zeta(8))^4; v
>> -0.2167565719512513? + 0.?e-18*I
>> sage: v.imag() == 0
>> True
>> sage: v._exact_value()
>> a^4 where a^12 + 2*a^8 + 5*a^4 + 1 = 0 and a in 0.4824786170789168? -
>> 0.4824786170789168?*I
>> sage: v.simplify()
>> sage: v
>> -0.2167565719512513?
>> sage: v._exact_value()
>> a^2 - a where a^3 + a - 1 = 0 and a in 0.6823278038280193?
>> sage: v.minpoly()
>> x^3 + 2*x^2 + 5*x + 1
>> sage: v.minpoly().roots(QQbar)
>> [(-0.2167565719512513?, 1),
>>  (-0.891621714024375? - 1.954093392512700?*I, 1),
>>  (-0.891621714024375? + 1.954093392512700?*I, 1)]
>>
>> (I have seen cases where "simplify" didn't do the trick, but recomputing the
>> algebraic number by taking the roots of the minimal polynomial did)
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "sage-support" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to sage-support+unsubscr...@googlegroups.com.
>> To post to this group, send email to sage-support@googlegroups.com.
>> Visit this group at https://groups.google.com/group/sage-support.
>> For more options, visit https://groups.google.com/d/optout.
> 

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] Stange behavior with QQbar[z]

[Thierry Dumont]

Something strange (at least for me).

I am computing with polynomials in QQbar[z]. So I do:

sage: P.=QQbar[]

And I define 2 polynomials N and D. Coefficients of N and D are all
real, algebraic numbers (say like  QQbar(sqrt(3))) and I consider the
rational fraction R=N/D

sage: R.parent()
Fraction Field of Univariate Polynomial Ring in z over Algebraic Field

Now, I want to look at R on the imaginary axis. I do (may be this is not
a good idea):

sage: PP.=QQbar[]
sage: RIaxe=R(z=QQbar(I)*x)
sage: RIaxe.parent()
Fraction Field of Univariate Polynomial Ring in x over Algebraic Field

it seems ok.

I take the numerator:
sage: pp=RIaxe.numerator()
sage: print pp
0.05638275005921478?*I*x^3 - 0.4022786138875136?*x^2 - 1/2*I*x + 1

okay.

sage: pp.parent()
Univariate Polynomial Ring in x over Algebraic Field

okay.

-
Now starts my problem; I want to extract from pp the polynomial which
has real coefficients:
sage: l=[s for s in pp.coefficients() if s.imag()==0]
sage: print l
[1, -0.4022786138875136?]

okay. Now let us construct a polynomial from l:
sage: print sum([l[i]*z^i for i in range(0,len(l))])
(-0.4022786138875136? + 0.?e-18*I)*z + 1
  ^
   Why? an imaginary part?
   --

Now if I use the radical_expression method:

sage: l=[QQbar(pp.coefficients()[i].radical_expression())*x^i for i in
range(0,pp.degree()+1) if pp.coefficients()[i].imag()==0]
sage: -0.4022786138875136?*x^2 + 1

which is ok, (but... a bit slow).

Why ?

Yours,
t.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Sage always crashes

[Thierry Dumont]

Le 01/09/2016 à 21:29, Pierre-Yves Bienvenu a écrit :
> Hi, I'm trying to use Sage 7.3 on Ubuntu. I've downloaded the archive,
> used archive manager to extract it and got a directory SageMath. Inside
> I double click the  sage executable. The terminal opens and after a lot
> of lines starting with "patching" it says "SageMath version 7.3, release
> date..." etc but immediately after "Ooops sage crashed". No way to find
> the allegedly created report so I don't know exactly what's wrong. Any
> help would be greatly appreciated (I'm an ignorant I must say).
> 
> -- 
>
I think you shoul not process like this.
Open a shell(terminal), go in the uncompresed directory and launch
./sage (do not click).
At least it will be cleaner. When "clicking", I am not sure the
environment is the same.

Yours
t.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: huge virtual memory size when launching 7.3

[Thierry Dumont]

Le 31/08/2016 à 00:19, Volker Braun a écrit :
> On Tuesday, August 30, 2016 at 11:03:39 PM UTC+2, leif wrote:
> 
> 48 GB claimed by *every* Sage process here
> 
> 
> Its a tiny fraction of the 128TB address space. We are not talking about
> used RAM here.
> 
> which is *total* (not free!) physical RAM + total (not free) swap
> space,
> in bytes.
> 
> 
> Really, you need to take enough address space so that any computation
> that you can possibly do isn't artificially limited by the address space.
> 
> -- 

Testing :
x=sage.misc.memory_info.MemoryInfo().virtual_memory_limit()/1024.^3
on very different machines:

1) My own computer, 8GB RAM + swap: x:  23.1133804321289  GB

2) a machine with, 96 GB Ram + swap: x: 153.79 GB

3) My raspberry pi3 :-) 1GB, No swap : x = 0.939544677734375 GB

On all these machines, sage starts gently, even if slowly (as expected)
on machine 2 (uses nfs), and 3.

On machine 2, sage stating process (as mesured with top) uses about 39GB
(no more...) out of 153.79 available.

t.




-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] huge virtual memory size when launching 7.3

[Thierry Dumont]

Le 30/08/2016 à 19:06, William Stein a écrit :
> On Tue, Aug 30, 2016 at 9:56 AM, Thierry Dumont
> <tdum...@math.univ-lyon1.fr> wrote:
>> I have two computers, and sage installed on both:
>>
>> 1) Ubuntu 12.04 , sage-7.3 on an nfs mount,y
>>
>> 2) Ubuntu 16.04,  sage-7.3 on a local system, on a ssd.
>>
>> With the first one, sage starts slowly (as could be expected!), and I
>> have time to look at sage starting with "top". And during the starting
>> phase, before the prompt, sage uses more than 16 GB of virtual memory
>> (VIRT: 16.380g); the same quantity is used when stopping sage (in the
>> console interface).
>> It seems quite large, no ?
>>
>> With the second machine, it's more difficult to see what happens with
>> top, but it does not seems to use more than 7gb.
>>
>> Why such a large amount of memory ? (ok, it's  virtual, but it's large,
>> no ?).
> 
> I think this is a hack that is used by PARI/GP, due to their
> "interesting" primitive (but fast!) memory design.
> 
>  -- William
> 
>>
>> t.d.
>>

Mhhh, yes.And they certainly do something with what they allocate; I
tried this (c++):

---
#define DO_SOMETHING
int main()
{
  //allocate a 16 gb array.
  const long int n=2*10;
  long int *x=new long int[n];
#ifdef DO_SOMETHING
  for(long int i=0;i<n;i++)
x[i]=0;
#endif
}
--

With  DO_SOMETHING defined, time ./a.out:
real0m7.352s
user0m2.611s
sys 0m4.648s

and  DO_SOMETHING undefined (as just allocating has quite no cost):
real0m0.004s
user0m0.004s
sys 0m0.000s

t.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] huge virtual memory size when launching 7.3

[Thierry Dumont]

I have two computers, and sage installed on both:

1) Ubuntu 12.04 , sage-7.3 on an nfs mount,y

2) Ubuntu 16.04,  sage-7.3 on a local system, on a ssd.

With the first one, sage starts slowly (as could be expected!), and I
have time to look at sage starting with "top". And during the starting
phase, before the prompt, sage uses more than 16 GB of virtual memory
(VIRT: 16.380g); the same quantity is used when stopping sage (in the
console interface).
It seems quite large, no ?

With the second machine, it's more difficult to see what happens with
top, but it does not seems to use more than 7gb.

Why such a large amount of memory ? (ok, it's  virtual, but it's large,
no ?).

t.d.




-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] jupyter lab

[Thierry Dumont]

Le 20/08/2016 à 08:09, HG a écrit :
> I am testing jupyter lab, all my kernels are recognized except one : sage ?
> Is there a way to config it manually to be used in jupyterlab ?
> Regards
> Henri
> 
THis is the first time I heard about jupyter lab ! (things evoluate very
fast...). But, in my "classical" jupyter(jupyter hub) installations, I
need to add "kernels" for the different language/applications I want to
run. They go in /usr/local/share/jupyter/kernels.
Exemple:
/usr/local/share/jupyter/kernels/sage
which contains a kernel.json (which says how to run sage):
{"display_name": "Sage 7.3, "argv": ["/usr/local/sage/sage", "-python",
"-m", "sage.repl.ipython_kernel", "-f", "{connection_file}"]}

Yours
t.
> -- 
> You received this message because you are subscribed to the Google
> Groups "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sage-support+unsubscr...@googlegroups.com
> .
> To post to this group, send email to sage-support@googlegroups.com
> .
> Visit this group at https://groups.google.com/group/sage-support.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: Problem installing database_gap-4.7.9 (sage-7.1)

[Thierry Dumont]

Le 21/04/2016 19:35, Volker Braun a écrit :

Linking with gsl is a known problem for the binary, should be fixed in 7.2



Actually, using a compiled buymyself 7.2 version, I have no problem.

No other solution than compiling?

t.d.


On Thursday, April 21, 2016 at 6:12:35 PM UTC+2, tdumont wrote:

Hello,

On an Ubuntu 14-04 machine, I download a sage-7.1 binary version,
install... ok.

The installation of database_gap-4.7.9 fails:

sage -i database_gap-4.7.9
...
Installed trans database.
Installed tomlib database.
Traceback (most recent call last):
File "/home/sage/sage-7.1/src/bin/sage-eval", line 4, in 
  from sage.all import *
File
"/home/sage/sage-7.1/local/lib/python2.7/site-packages/sage/all.py",
line 85, in 
  from sage.misc.all   import * # takes a while
File
"/home/sage/sage-7.1/local/lib/python2.7/site-packages/sage/misc/all.py",
line
91, in 
  from functional import (additive_order,
File

"/home/sage/sage-7.1/local/lib/python2.7/site-packages/sage/misc/functional.py",

line 30, in 
  from sage.rings.complex_double import CDF
ImportError:

/home/sage/sage-7.1/local/lib/python2.7/site-packages/sage/rings/complex_double.so:

undefined symbol: gsl_complex_sin


any idea?

t.d.

--
You received this message because you are subscribed to the Google
Groups "sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to sage-support+unsubscr...@googlegroups.com
.
To post to this group, send email to sage-support@googlegroups.com
.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.


--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Benchmark linear system solve (sage x python(jupyter))

[Thierry Dumont]

Le 30/03/2016 22:33, jmarcellopere...@ufpi.edu.br a écrit :


start = time.time()

x = np.linalg.solve(A,B)

end = time.time()

print(end - start)



I have tried this juste now; sage version is 7.0

I get :

1) first version:

CPU times: user 6.88 s, sys: 76 ms, total: 6.95 s
Wall time: 2.03 s

2) second version:

CPU times: user 6.86 s, sys: 104 ms, total: 6.96 s
Wall time: 2.304  s


Note from the first execution that we probably use the multitheraded 
version of the blas (my computer has 4 core).


Yours,

t.d.

--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] Fwd: Re: Strange behavior with RealField(n)

[Thierry Dumont]

 Message transféré 
Sujet : Re: Strange behavior with RealField(n)
Date : Mon, 22 Feb 2016 10:17:30 +0100
De : paul zimmermann <paul.zimmerm...@inria.fr>
Pour : Thierry Dumont <tdum...@math.univ-lyon1.fr>
Copie à : nbr...@sfu.ca

[please forward to sage-support, I am not allowed to post there]

   Thierry,

I can reproduce the problem as follows (in Sage 6.8):

sage: alarm(10)
sage: R=RealField(10^9)
sage: pii=4*atan(R(1))
sage: R=RealField(10^2)
sage: pii=4*atan(R(1))
sage: pii
NaN

My guess is that the interrupted calculation leaves MPFR in a strange stage
(exponent range, flags, cached values, ...)

Best regards,
Paul

> From: Thierry Dumont <tdum...@math.univ-lyon1.fr>
> Date: Sun, 21 Feb 2016 19:10:04 +0100
> User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101
>  Thunderbird/38.5.1
> 
> 
> [1:text/plain Hide]
> 
> I have students who want to compute decimals of pi...so, what can we do 
> with RealField(n) ?
> I make the following script (pi.sage):
> 
> 
> for p in [2..10]:
>  R=RealField(10^p)
>  pii=4*atan(R(1))
>  print p,R,pii
> 
> 
> Then, using sage 7.0 or 7.1.beta4:
> 
> attach("pi.sage")
> 
> This produces a lot of seemingly correct output, but, as it takes a too 
> long time to finish :-), I interrupt the computation (Ctrl-c).
> 
> So, lets try again; replace 10 by 5 in the for statement (I do not leave 
> sage). I get NaNs:
> 
> 
> 2 Real Field with 100 bits of precision NaN
> 3 Real Field with 1000 bits of precision NaN
> 4 Real Field with 1 bits of precision NaN
> 5 Real Field with 10 bits of precision NaN
> .
> 
> Strange.
> 
> Yours
> t.d.
> 
> 
> 
> [2:text/x-vcard Show Save:tdumont.vcf (390B)]
> 


-- 
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Strange behavior with RealField(n)

[Thierry Dumont]

Le 21/02/2016 20:40, John Cremona a écrit :

Try RealField(500).pi() and similar.

Yes, it works... but my small piece of code should also give correct 
results...

thanks.
t.


On 21 Feb 2016 18:10, "Thierry Dumont" <tdum...@math.univ-lyon1.fr
<mailto:tdum...@math.univ-lyon1.fr>> wrote:

I have students who want to compute decimals of pi...so, what can we
do with RealField(n) ?
I make the following script (pi.sage):


for p in [2..10]:
 R=RealField(10^p)
 pii=4*atan(R(1))
 print p,R,pii


Then, using sage 7.0 or 7.1.beta4:

attach("pi.sage")

This produces a lot of seemingly correct output, but, as it takes a
too long time to finish :-), I interrupt the computation (Ctrl-c).

So, lets try again; replace 10 by 5 in the for statement (I do not
leave sage). I get NaNs:


2 Real Field with 100 bits of precision NaN
3 Real Field with 1000 bits of precision NaN
4 Real Field with 1 bits of precision NaN
5 Real Field with 10 bits of precision NaN
.

Strange.

Yours
t.d.


--
You received this message because you are subscribed to the Google
Groups "sage-support" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to sage-support+unsubscr...@googlegroups.com
<mailto:sage-support%2bunsubscr...@googlegroups.com>.
To post to this group, send email to sage-support@googlegroups.com
<mailto:sage-support@googlegroups.com>.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google
Groups "sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to sage-support+unsubscr...@googlegroups.com
<mailto:sage-support+unsubscr...@googlegroups.com>.
To post to this group, send email to sage-support@googlegroups.com
<mailto:sage-support@googlegroups.com>.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.


--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] Strange behavior with RealField(n)

[Thierry Dumont]

I have students who want to compute decimals of pi...so, what can we do 
with RealField(n) ?

I make the following script (pi.sage):


for p in [2..10]:
R=RealField(10^p)
pii=4*atan(R(1))
print p,R,pii


Then, using sage 7.0 or 7.1.beta4:

attach("pi.sage")

This produces a lot of seemingly correct output, but, as it takes a too 
long time to finish :-), I interrupt the computation (Ctrl-c).


So, lets try again; replace 10 by 5 in the for statement (I do not leave 
sage). I get NaNs:



2 Real Field with 100 bits of precision NaN
3 Real Field with 1000 bits of precision NaN
4 Real Field with 1 bits of precision NaN
5 Real Field with 10 bits of precision NaN
.

Strange.

Yours
t.d.


--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Install Sage failed both on 64 bit as on 32 bit Linux Mint

[Thierry Dumont]

Le 30/01/2016 22:49, Paul van Gorsel a écrit :

First I ran a 64 bits Linux Mint Debian Edition, unfortunately the
installation of Sage failed. Than I installed a 32 bit version (Linux
Mint 17.3), the same thing happened.

This is my laptop: HP-ProBook-4520s

Linux pgo-HP-ProBook-4520s 3.19.0-32-generic #37~14.04.1-Ubuntu SMP Thu
Oct 22 09:37:25 UTC 2015 i686 i686 i686 GNU/Linux
memory:
  total   used   free sharedbuffers cached
Mem:  2884   2662221141 88   1842
-/+ buffers/cache:731   2153
Swap: 2925 20   2905

It's such a waste of time. Can you advise me which distro to choose, or
even better, what causes these problems?

With regards,
Paul van Gorsel (The Netherlands)



The installation failed... ok, but let us know which message you get, 
what happened...

I am completely sure that Sage can run on Mint.

Yours
t.

--
You received this message because you are subscribed to the Google
Groups "sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to sage-support+unsubscr...@googlegroups.com
.
To post to this group, send email to sage-support@googlegroups.com
.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.


--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: problem with nfs or... what ? (7.0)

[Thierry Dumont]

Le 27/01/2016 16:42, Volker Braun a écrit :

On Tuesday, January 26, 2016 at 12:43:28 PM UTC-5, tdumont wrote:

For binary distributions, we can install it at any place we want, no?
What is the difference?


The binary is built with a long path and comes with a script that
search it with a shorter path.


ok! thanks!
Can I run this script on my compiled version ? how?
yours
t.d.

--
You received this message because you are subscribed to the Google
Groups "sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to sage-support+unsubscr...@googlegroups.com
.
To post to this group, send email to sage-support@googlegroups.com
.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.


--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Re: problem with nfs or... what ? (7.0)

[Thierry Dumont]

Le 25/01/2016 15:37, Volker Braun a écrit :

The directory path has to match, if its different on server and client
then you have to compile it on the client (or mount on the server in the
same path first).



Ok it's working. But is it something new ?

For binary distributions, we can install it at any place we want, no?
What is the difference?

Yours (thanks).
t.


On Monday, January 25, 2016 at 9:20:17 AM UTC-5, tdumont wrote:

Hello,

I have tried to switch from 6.10 to 7.0, but... I have some problems.

1) Let me fist explain the architecture of our client/server
installation:

-We have a nfs server (Debian + Zfs) which export a directory
(/srv/local64) to a large number of clients (all Ubuntu 14.O4-LTS) but
with different generations of Intel CPUs.

So, on the server I compiled sage from the 7.0 source with
export SAGE_FAT_BINARY="yes"
make distclean
make

On the server, I can launch this compiled sage 7.0 without any
problem..

2) On the clients, I have a problem:
./sage

   File

"/usr/local/sage/local/lib/python2.7/site-packages/sage/misc/temporary_file.py",

line 51, in delete_tmpfiles
  from sage.misc.misc import SAGE_TMP
File
"/usr/local/sage/local/lib/python2.7/site-packages/sage/misc/misc.py",
line 57, in 
  import sage.misc.prandom as random
File
"/usr/local/sage/local/lib/python2.7/site-packages/sage/misc/prandom.py",
line
58, in 
  from sage.misc.randstate import current_randstate
ImportError: libgmp.so.16: cannot open shared object file: No such file
or directory

Permissions seems ok... So, no idea for me.


This is new: I have kept a tree with a 6.10 version and everything
is ok.

???
Yours

t.d.

--
You received this message because you are subscribed to the Google
Groups "sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to sage-support+unsubscr...@googlegroups.com
.
To post to this group, send email to sage-support@googlegroups.com
.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.


--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] problem with nfs or... what ? (7.0)

[Thierry Dumont]

Hello,

I have tried to switch from 6.10 to 7.0, but... I have some problems.

1) Let me fist explain the architecture of our client/server installation:

-We have a nfs server (Debian + Zfs) which export a directory 
(/srv/local64) to a large number of clients (all Ubuntu 14.O4-LTS) but 
with different generations of Intel CPUs.


So, on the server I compiled sage from the 7.0 source with
export SAGE_FAT_BINARY="yes"
make distclean
make

On the server, I can launch this compiled sage 7.0 without any problem..

2) On the clients, I have a problem:
./sage

 File 
"/usr/local/sage/local/lib/python2.7/site-packages/sage/misc/temporary_file.py", 
line 51, in delete_tmpfiles

from sage.misc.misc import SAGE_TMP
  File 
"/usr/local/sage/local/lib/python2.7/site-packages/sage/misc/misc.py", 
line 57, in 

import sage.misc.prandom as random
  File 
"/usr/local/sage/local/lib/python2.7/site-packages/sage/misc/prandom.py", line 
58, in 

from sage.misc.randstate import current_randstate
ImportError: libgmp.so.16: cannot open shared object file: No such file 
or directory


Permissions seems ok... So, no idea for me.


This is new: I have kept a tree with a 6.10 version and everything is ok.

???
Yours

t.d.

--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] Ipython notebook : mathjax availability depends on the way one starts sage...

[Thierry Dumont]

This seems related to the problem I had when running sage under jupyter 
hub, and also in jupyter (the later: launching jupyter, then sage 
6.0beta1):
-some parts of the documentation (the documentation in the sage tree) 
cannot be accessed (FAQs for example).


But now  (since 6.0beta1?), jsmol is ok, and for me mathjax too.

This all looks like a symlink problem...


Le 23/10/2015 18:19, Emmanuel Charpentier a écrit :

Le vendredi 23 octobre 2015 13:48:59 UTC+2, Jeroen Demeyer a écrit :

Now I am *totally* confused. Can you please just reboot this thread,
i.e. pretend that I haven't read anything of this thread and explain
from scratch what your setup is and what your problem is.


Okay.

I have a Sagemath (currently 6.10beta1) installation in /usr/local/sage.
For the record, /usr/local/bin/sage is symlinked to /usr/local/sage/sage.

I tried two things :

1) using Sagemath's version of the IPython notebook (lainched with "sage
-n ipython" or "sage -n jupyter") : it turns out that the resulting notebook
- *does* have access to mathjax when working on sheets stored in the
directory where the command has been given (the sheet opens and displays
math correctly) ;
- does *not* have access to mathjax when one navigates in the notebook
to a different directory in order to work on sheets stored in that
directory :
 + an error dialog box appears telling mathjax is inaccessible ;
 + the *relative* path to matjax is in the title of the dialog box ;
 + the typeset output does not appear (nothing appears in the cells
where a typeset result is expected).

2) I also tried to install the Sagemath kernel in a systemwide
installation of Ijupyter(obtained by "pi³ install jupyter"). This
installation has been previously tested by installing the IRKernel
interface to a systemwide R. As suggested by William, I symlinked
$SAGE_LOCAL/share/jupyter/kernels/* and
$SAGE_LOCAL/share/jupyter/nbextensions/* to the corresponding
subdirectories of /usr/local/share/jupyter/ (Debian's idea of the
systemwide configuration of Jupyter).
This gave me a notebook interface where I could operate on Python,
Sagemath and R sheets ; the Sage sheets could use mathjax wherever they
were launched.

I hope to have been clearer...

HTH,

--
Emmanuel Charpentier

--
You received this message because you are subscribed to the Google
Groups "sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to sage-support+unsubscr...@googlegroups.com
.
To post to this group, send email to sage-support@googlegroups.com
.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.


--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

Re: [sage-support] 6.9rc3 : Ipython notebook seems seriously out of whack.

[Thierry Dumont]

Le 10/10/2015 22:32, Jeroen Demeyer a écrit :

On 2015-10-10 21:24, Emmanuel Charpentier wrote:

In [2]:
%display simple

solve(x^2+1==0,x,to_poly_solve="force")
Out[2]:

[]


Huh ?

Works for me...

In any case, it would be very unlikely that a Sage command actually
gives different output in the command line and in the Jupyter notebook.



>jupyter notebook
sage 6.9 rc3
This works for me too.
t;d.

--
You received this message because you are subscribed to the Google Groups 
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
<>

[sage-support] playing with sage and jupyterhub: some (small) problems.

[Thierry Dumont]

Hi,

With a colleague well known here (N. T. from Orsay) we started to play
with Jupyterhub and Sage during a rainy social day of the congress of
the French Society for Applied Maths (SMAI).

Actually, building a Jupyterhub server is something easy, and running
Sage in it also. But there remain for me two problems:

- the first one in certainly the less important: a part of the
documentation cannot be accessed when running the sage kernel (this is
the inside part, stored in sage's home)  (404 not found...).

- the second one is, may be more important: I was not able to run Jmol;
the canonic example is:

var(u v)
f = (lambda u,v: cos(u), lambda u,v: sin(u)+cos(v), lambda u,v: sin(v))

which only produces an empty cell as result.

I guess these problems can be related to the configuration of jupyter,
which is in /usr/local/share/jupyter/{hub,kernels}

--- /usr/local/share/jupyter/kernels contains a directory sage:

--- in this directory  I have  the file kernel.json:
{display_name: Sage, argv: [/opt/sage/sage, -python, -m,
sage.repl.ipython_kernel, -f, {connection_file}]}
(which is correct, but is it sufficient ?)

 in the same directory /usr/local/share/jupyter/kernels/sage, I
have doc which is a symbolic link to /opt/sage/src/doc/output/html/en
(/opt/sage is sage's home).


---/usr/local/share/jupyter/hub contains nbextensions/jsmol which was
copied from my $HOME/.ipython on an other machine... There is certainly
something to change, but what ?

Yours.

t.d.

-- 
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
attachment: tdumont.vcf

Re: [sage-support] Re: Algebraic Field or Symbolic Ring ?

[Thierry Dumont]

Le 01/07/2015 08:23, Ralf Stephan a écrit :
 On Tuesday, June 30, 2015 at 1:24:23 PM UTC+2, tdumont wrote:
 
 Why is xxc in Symbolic Ring and not in Algebraic Field?
 mxx and xxc are essentialy the same thing, no ?
 
 When you input sqrt(2) it is not clear which object you want,
 Sage traditionally decides in favor of a symbolic. If you specifically
 need the number field element you must say so with
 
 sage: K.sqrt2 = QuadraticField(2)
 sage: sqrt2^5
 4*sqrt2
 
 That said, I think the sqrt function of Rational (which is called here)
 could be enhanced with an option to return a number field element
 for convenience.
 
 Regards,  
 
Thanks,
But this is not what I do!
I do
xx=QQ(2)
xxc=xx.sqrt()
xxc.parent()
Symbolic Ring

It should not be the same thing as xxc=sqrt(2)...

Yours
t.


 -- 
 You received this message because you are subscribed to the Google
 Groups sage-support group.
 To unsubscribe from this group and stop receiving emails from it, send
 an email to sage-support+unsubscr...@googlegroups.com
 mailto:sage-support+unsubscr...@googlegroups.com.
 To post to this group, send email to sage-support@googlegroups.com
 mailto:sage-support@googlegroups.com.
 Visit this group at http://groups.google.com/group/sage-support.
 For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
attachment: tdumont.vcf

[sage-support] Algebraic Field or Symbolic Ring ?

[Thierry Dumont]

I have found something a bit strange for me:

sage: xx=QQ(2)
sage: xx.parent()
Rational Field

ok...

sage: xxc=xx.sqrt()
sage: xxc.parent()
Symbolic Ring


Why not, but now:

sage: M1=matrix(QQ,1,1,[2])

sage: M1.parent()
Full MatrixSpace of 1 by 1 dense matrices over Rational Field


sage: M1C=M1.cholesky()
sage: M1C
[1.414213562373095?]
sage: M1C.parent()
Full MatrixSpace of 1 by 1 dense matrices over Algebraic Field

So, we have:

xxc: Symbolic Ring
M1C: Full MatrixSpace of 1 by 1 dense matrices over Algebraic Field

and also:
sage: mxx=M1C.row(0)[0]
sage: mxx.parent()
Algebraic Field
sage:  mxx==xxc
True

Why is xxc in Symbolic Ring and not in Algebraic Field?
mxx and xxc are essentialy the same thing, no ?

A bit strange for me...

Yours
 t.d.

-- 
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
attachment: tdumont.vcf

Re: [sage-support] Infinite loop in matrix.change_ring

[Thierry Dumont]

Le 08/05/2015 10:23, Vincent Delecroix a écrit :

Bonjour Thierry,

Could you post your matrices Prec, M, mfeEast? I am not able to
reproduce the problem with change_ring...

Vincent



Hello Vincent.
Yes,
I join a part of my code, which reproduces the problem...

Yours
t.

--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
lprec=[(2250247185651397249701766144/940135272187668275673825470054308951, 
0, -7044387942611760194091358546315/940135272187668275673825470054308951, 0), 
(0, 628091429911656478463873925/115769812361316527145683860922368, 0, 
-12849448754121499186194957/7235613272582282946605241307648), 
(-7044387942611760194091358546315/940135272187668275673825470054308951, 0, 
4500494371302794499403532288000/940135272187668275673825470054308951, 0), (0, 
-12849448754121499186194957/7235613272582282946605241307648, 0, 
52574617459123014035/226112914768196342081413790864)]

mlist=[(-40960, 1423743620867035193/22298143521175, 
-483387868093604552/7539677263923, 34549818091021825/797033922567),
 (-1423743620867035193/22298143521175, 98304, 
-311185622448809473/3269359628156, 3297567640129996/58560241971),
 (-483387868093604552/7539677263923, 311185622448809473/3269359628156, 
-3752999689502719/45812984491, 8587337044602103/354376193361),
 (-34549818091021825/797033922567, 3297567640129996/58560241971, 
-8587337044602103/354376193361, 
-4621642813466987640258501351/80595054643327775067341)]


Prec=matrix(QQ,lprec)
M=matrix(QQ,mlist)
print Prec:,Prec.parent()
print M:,M.parent()

LC=Prec.cholesky()
print LC: ,LC.parent()
print LC: ,LC.list()


LCt=LC.transpose()
print LCt: ,LCt.parent()

#
Z=LC*M*LCt
print LC*M*LCt: ,Z.parent()
#
print Now, we play...
M1=Z.change_ring(RDF)

attachment: tdumont.vcf

[sage-support] Infinite loop in matrix.change_ring

[Thierry Dumont]

Hello,

I have 2 matrices: Prec and M.

sage: Prec.parent
Full MatrixSpace of 4 by 4 dense matrices over Rational Field
sage: M.parent()
Full MatrixSpace of 4 by 4 dense matrices over Rational Field
sage: Prec.is_symetric()
True

ok. I also know that Prec is positive. Then, I do:

sage: LC=Prec.cholesky()
sage: LC.parent()
Full MatrixSpace of 4 by 4 dense matrices over Algebraic Field

This is ok.


sage: print LC.list()
[0.004892377274102731?, 0, 0, 0, 0, 0.002329237476562281?, 0, 0, 
-0.001531556341905637?, 0, 1/640, 0, 0, -0.0007624219319910414?, 0, 
0.001320553523013307?]

Everyting seems ok.


sage: LCt=LC.transpose()
sage: LCt.parent()
Full MatrixSpace of 4 by 4 dense matrices over Algebraic Field
ok! (of course).

sage: Z=LC*mfeEast*LCt

sage: Z.parent()
Full MatrixSpace of 4 by 4 dense matrices over Algebraic Field

This is perfect.

---
NOW, my problem:

M1=Z.change_ring(RDF)

enters an infinite loop
--

For information:

sage: version()
'SageMath Version 6.6, Release Date: 2015-04-14'

sage: print Z.list()
[-92170124724281231347784341258240/940135272187668275673825470054308951, 
0.7276068751089991?, -0.1831867383804206?, 0.04189076099436858?, 
-0.7276068751089991?, 
7537097158939877741566487100/14132057173012271380088361929, 
-0.1186331808041865?, -0.001369231431906186?, -0.1831867383804206?, 
0.1186331808041865?, 0.010771471259174368?, -0.07650322439253288?, 
-0.04189076099436858?, -0.001369231431906186?, 0.07650322439253288?, 
-0.1562464847599110?]



Yours!

t.

--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
attachment: tdumont.vcf

Re: [sage-support] Infinite loop in matrix.change_ring

[Thierry Dumont]

Le 08/05/2015 11:00, Vincent Delecroix a écrit :

On 08/05/15 10:52, Thierry Dumont wrote:

Le 08/05/2015 10:23, Vincent Delecroix a écrit :

Bonjour Thierry,

Could you post your matrices Prec, M, mfeEast? I am not able to
reproduce the problem with change_ring...

Vincent



Hello Vincent.
Yes,
I join a part of my code, which reproduces the problem...


Hello,

It is not an infinite loop. The problem is that suddenly, the numbers
change their parent from AA (=Algebraic Real Field) to QQbar (=Algebraic
Field). I have to look more carefully in your code where this does happen.

If you have an algebraic number that is not known to be real, then the
conversion to RDF checks that the imaginary part is zero. It is
currently very slow but there is some work in this direction [1]. I am
pretty sure that if you are very patient it should end up with
something, but it might takes several hours!

Workaround:

sage: ZZ = matrix(AA, [z.real() for z in Z.list()])
sage: ZZ.change_ring(RDF)

Vincent

  [1] http://trac.sagemath.org/ticket/18333



aaahh, yes !

Thanks!

t.

--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
attachment: tdumont.vcf

[sage-support] knowing which {packae, library} is used.

[Thierry Dumont]

I apologize for this question: I am sure I have seen the answer 
somewhere, but I cannot find int..


I run some sage code: for example I solve a linear system with 
coefficients in QQ, then an other one with coefficients in RDF.

I want to list which library (package) is used in both cases.

This is for pedagogical purpose.

I apologize again.

t.d.

--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
attachment: tdumont.vcf

Re: [sage-support] Re: Computational geometry in the plane: in Sage?

[Thierry Dumont]

Le 22/04/2014 18:33, leif a écrit :

Volker Braun wrote:

On Tuesday, April 22, 2014 4:43:43 PM UTC+1, tdumont wrote:

I am in 6.1.1; as far as as know, I have built sage from source,
using
the system Atlas blasand I have just upgraded from Ubuntu
13.10 to
14.04. Could the problem be here?


most likely, you switched out the system ATLAS library and now it
crashes in the ATLAS library.


Well, presumably, but not much of an issue.  Simply reinstalling the
ATLAS spkg (with SAGE_ATLAS_LIB set appropriately) and rebuildung every
package that depends on ATLAS /should/^TM be sufficient:

$ env SAGE_ATLAS_LIB=/path/to/system/atlas/libs ./sage -f atlas



Yes, this is enough... now it works.
Yours
t.




$ env SAGE_UPGRADING=yes make build

(Note sure whether some Cython extension modules need to get
rebuilt/relinked as well, and if so, whether the dependencies are
complete in that the above would automatically rebuild them.)


-leif



--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
attachment: tdumont.vcf

[sage-support] Big problem with ppa (6.1).

[Thierry Dumont]

Dear all,

I have juste installed 6.1 with the Ubuntu ppa on my lab's machines...:

apt-get nstall sagemath-upstream-binary sagemath-optional

(we are in LTS, Ubuntu 12.04);

The directories:

sloane/
jones/

are installed in /   !!! (yes: i say /)

idem if I do:

sage -i sage-mode

a directory emacs/ is created in / with sage-mode and related files in it.

This is a least not suitable 

Yours
t.



-- 
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.
attachment: tdumont.vcf

Re: [sage-support] Re: ppa 6.0

[Thierry Dumont]

Le 10/02/2014 17:07, slelievre a écrit :

Jan Groenewald prepared a PPA for Sage 6.1.1, see his post on sage-release:
   https://groups.google.com/d/msg/sage-release/qhQ2BPM34Z0/kmneDmNsuroJ


Great, it works fine.
Thanks!

May be it would be good to make mirrors of the ppa?
Nawadays I download sage about 50 times from the ppa!

t.

tdumont wrote:

Will there be a ppa version of 6.0 ?

--
You received this message because you are subscribed to the Google
Groups sage-support group.
To unsubscribe from this group and stop receiving emails from it, send
an email to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.


--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.
attachment: tdumont.vcf

[sage-support] LDAP and START_TLS in the notebook (sage 6.0) ?

[Thierry Dumont]

Hello,

may be you remember I made a patch to (old) versions of the notebook to
use ldap.

But, I am switching to sage 6.0, and I want to use ldap again.

So, I installed (system wide)

libldap2-dev libsasl2-dev
and
openssl

Then in sage:
./sage -sh
(sage-sh) easy_install python-ldap

Ok.
Then starting the notebook (with secure=true) and login as admin, I can
configure Ldap.

But...  I cannot use ldaps:// (only ldap:// which is not secure!)

Is it possible to use TLS?

Yours,
t.

-- 
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.
attachment: tdumont.vcf

[sage-support] sage 6.0: cannot build graphviz

[Thierry Dumont]

 I cannot install graphviz:

sage -i graphviz...



make[3]: *** Pas de règle pour fabriquer la cible « 
../../plugin/pango/libgvplugin_pango.la », nécessaire pour « 
dot_builtins ». Arrêt.
make[3]: quittant le répertoire « 
/usr/local/sage-6.0/local/var/tmp/sage/build/graphviz-2.16.1.p0/src/cmd/dot 
»


I am in Ubuntu 13.10.
sage 6.0 was compiled from source (not an update).

Yours
t.


--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.
attachment: tdumont.vcf

[sage-support] ppa 6.0

[Thierry Dumont]

Dear colleagues,

Will there be a ppa version of 6.0 ?

Since I need to install sage on about 20 machines, the *.deb is nice

But I have time

t.d.

--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.
attachment: tdumont.vcf

[sage-support] Polynomial question.

[Thierry Dumont]

Hello, all,

I am going to manipulate Legendre (P) polynomials.
So I do something like this:

sage: P.x=QQ[]

sage: #generate de first Lagrange polynomial
sage: s=[legendre_P(i,x) for i in [0..2]]
sage: print s
[1, x, 3/2*x^2 - 1/2]

sage: #now, look at the parents
sage: s[0].parent()
Integer Ring
sage: s[1].parent()
Univariate Polynomial Ring in x over Rational Field

Ok, this is correct, and seems nice; but I want to evaluate these
polynomials for  different values of x, and you cannot evaluate a member
of Integer Ring at say, x=1/21... So I need to compute the parent of
polynomials (which possibly are scalars)  in all my functions, make test
and so on

I have found a work-around:
sage:  s=[P(legendre_P(i,x)) for i in [0..2]]

Then, I get:

sage: s[0].parent()
Univariate Polynomial Ring in x over Rational Field

Ok, but is there some other way to do this?

Yours
t.

-- 
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.
attachment: tdumont.vcf

Re: [sage-support] n-dimensional numerical integration

[Thierry Dumont]

Le 29/06/2013 14:21, Martin Albrecht a écrit :

Hi all,

do we have n-dimensional numerical integration in Sage? All I can find is 3D,
but what about higher dimensions?

Cheers,
Martin

MMhhh... I do not know any (free) library doing n-dimensional numerical 
integration with n large. And so, there is certainly no such method in Sage.
If you can by use a n-cube as integration domain, tensor product of 
Gaussian formulas will do the job; otherwise in a more general domain, 
and for n3 people generally use a Monte Carlo method.

Yours,
t.d.

--
name: Martin Albrecht
_pgp: http://pgp.mit.edu:11371/pks/lookup?op=getsearch=0x6532AFB4
_otr: 47F43D1A 5D68C36F 468BAEBA 640E8856 D7951CCF
_www: http://martinralbrecht.wordpress.com/
_jab: martinralbre...@jabber.ccc.de



--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.


attachment: tdumont.vcf

[sage-support] Problem installing 5.10 withy ppa on Ubuntu 13.04

[Thierry Dumont]

Hi,

Install fails with:

Error installing package scitools++


Before this I get:


Removing 
/usr/lib/sagemath/local/lib/python2.7/site-packages/SciTools_-1.0-py2.7.egg-info
Writing 
/usr/lib/sagemath/local/lib/python2.7/site-packages/SciTools_-1.0-py2.7.egg-info
cp: impossible de créer le fichier standard 
«/usr/lib/sagemath/local/lib/python2.5/site-packages/scitools/»: Aucun 
fichier ou dossier de ce type


Yes my computer do speak French.
Ths last line means: impossible to create 
/usr/lib/sagemath/local/lib/python2.5/site-packages/scitools


and actually in /usr/lib/sagemath/local/lib there is
lrwxrwxrwx  1 root root9 nov.  12  2012 python - python2.7
drwxr-xr-x 30 root root61440 juin  22 12:42 python2.7

Actually, doing:
ln -s python2.7 python2.5
solves the problem... but...


Yours
t.

--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.


attachment: tdumont.vcf

Re: [sage-support] Problem installing 5.10 withy ppa on Ubuntu 13.04

[Thierry Dumont]

Le 22/06/2013 15:52, Jan Groenewald a écrit :

Hi

This error is not particular to the PPA. It will happen with a binary
from sagemath.org http://sagemath.org or a fresh from-source install
as well.
It seems to be not during apt-get install, instead during a sage -i
scitools++.spkg. The scitools spkg should be fixed, as far as I can see.

Regards,
Jan



Yes,
I apologize... it is a scitools problem...
yours
t.




On 22 June 2013 12:54, Thierry Dumont tdum...@math.univ-lyon1.fr
mailto:tdum...@math.univ-lyon1.fr wrote:

Hi,

Install fails with:
**__**__
Error installing package scitools++
**__**__

Before this I get:


Removing

/usr/lib/sagemath/local/lib/__python2.7/site-packages/__SciTools_-1.0-py2.7.egg-info
Writing

/usr/lib/sagemath/local/lib/__python2.7/site-packages/__SciTools_-1.0-py2.7.egg-info
cp: impossible de créer le fichier standard
«/usr/lib/sagemath/local/lib/__python2.5/site-packages/__scitools/»:
Aucun fichier ou dossier de ce type

Yes my computer do speak French.
Ths last line means: impossible to create
/usr/lib/sagemath/local/lib/__python2.5/site-packages/__scitools

and actually in /usr/lib/sagemath/local/lib there is
lrwxrwxrwx  1 root root9 nov.  12  2012 python - python2.7
drwxr-xr-x 30 root root61440 juin  22 12:42 python2.7

Actually, doing:
ln -s python2.7 python2.5
solves the problem... but...


Yours
t.

--
You received this message because you are subscribed to the Google
Groups sage-support group.
To unsubscribe from this group and stop receiving emails from it,
send an email to sage-support+unsubscribe@__googlegroups.com
mailto:sage-support%2bunsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com
mailto:sage-support@googlegroups.com.
Visit this group at http://groups.google.com/__group/sage-support
http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/__groups/opt_out
https://groups.google.com/groups/opt_out.





--
   .~.
   /V\ Jan Groenewald
  /( )\ www.aims.ac.za http://www.aims.ac.za
  ^^-^^

--
You received this message because you are subscribed to the Google
Groups sage-support group.
To unsubscribe from this group and stop receiving emails from it, send
an email to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.




--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/groups/opt_out.


attachment: tdumont.vcf

Re: [sage-support] Does SAGE support bifurcation analysis?

[Thierry Dumont]

Le 22/02/2013 15:18, Piero a écrit :

Is there any support of bifurcation analysis in SAGE? Alternatively, are there 
binding tools to AUTO or XPPAUT?
Many thanks


Not yet, as far as I know.
This would be great to integrate this in Sage!
t.d.

--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


attachment: tdumont.vcf

Re: [sage-support] LDAP (or AD or...) again

[Thierry Dumont]

Le 22/11/2012 08:53, Jori Mantysalo a écrit :

Getting back to this old question... Version 5.4. has been released. How
to set up ldap? Or some kind of external authentication, like CAS or AD?

Has anyone made this? I'm stuck with old errors (this time Failed to
load application: No module named flask_version) and don't quite know
where to start debugging.



I am also intersted by this old question. I used to patch the old 
notebook to get an ldap access and identification). I recognize that 
what I did is neither clean nor generic, but it works since some years 
in my University server, and in a national server for French Mathematicians.


I do not understand where to patch the new Notebook, and thus I am stuck 
to 4.* versions. If somebody can give some hints, It would be nice.


Yours
t.d.

--
You received this message because you are subscribed to the Google Groups 
sage-support group.
To post to this group, send email to sage-support@googlegroups.com.
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support?hl=en.


attachment: tdumont.vcf

[sage-support] Graphviz do not build (Sage 5.0/Ubuntu 10.04)

[Thierry Dumont]

Trying:

sage -i graphviz...

I get:



mv -f .deps/no_demand_loading.Tpo .deps/no_demand_loading.Po
make[3]: *** Pas de règle pour fabriquer la cible «
../../plugin/pango/libgvplugin_pango.la », nécessaire pour «
dot_builtins ». Arrêt.
make[3]: quittant le répertoire «
/usr/local/sage-5.0/spkg/build/graphviz-2.16.1.p0/src/cmd/dot »
make[2]: *** [all-recursive] Erreur 1
make[2]: quittant le répertoire «
/usr/local/sage-5.0/spkg/build/graphviz-2.16.1.p0/src/cmd »
make[1]: *** [all-recursive] Erreur 1
make[1]: quittant le répertoire «
/usr/local/sage-5.0/spkg/build/graphviz-2.16.1.p0/src »
make: *** [all] Erreur 2
Error building Graphviz


It seems that a pango lib is missing...

Yoors
t.d.

-- 
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

Re: [sage-support] Re: Java crash with plot3d

[Thierry Dumont]

Le 08/03/2012 14:44, Jori Mantysalo a écrit :

On Mon, 5 Mar 2012, P Purkayastha wrote:


It works here on vanilla sage-4.8, sage-4.8+new flask notebook+new
jmol,
and sage-5.0beta2+new flask notebook+new jmol with both opera-11.61 and
firefox-10.0.2 on Gentoo 64 bit. java is the 64bit variety from sun,
version 1.6.0.31.


I just tested this on virtual machine. It worked when I installed Fedora
16. But after yum update it crashed Firefox. At least FF version was
updated from 7.x to 10.x with yum, but this might depend on java
version, not on FF version.

Next question: How can I save java-applet from Sage to another page, so
that I can send error report to FF mailing list?



On Ubuntu (11.04, 10.04 and maybe others), the policy of Oracle resulted 
(as far as I know) in the withdrawal of all packages sun*java from the 
distribution.
Once, our computers, after a weekly update, could no more play Jmol, 
because they had regressed to open-jdk and other packages.


I believe that we must suppress any Java packages from sage Sage 
must not depend on this Money Oriented Language (see: 
http://www.stlport.org/resources/StepanovUSA.html).

But it will not be so easy :-(

t.d.

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] Bug when running in the notebook, not in command line.

[Thierry Dumont]

Hello,

We use sage 4.8

The following line:

W=WeylGroup(F4)

works in command line, BUT
do not work when evaluated in the Notebook:

Error message is:
Traceback (most recent call last):
  File stdin, line 1, in module
  File _sage_input_4.py, line 10, in module
exec compile(u'open(___code___.py,w).write(# -*- coding: utf-8
-*-\\n +
_support_.preparse_worksheet_cell(base64.b64decode(Um9vdFN5c3RlbShDYXJ0YW5UeXBlKCdGNCcpKS5hbWJpZW50X3NwYWNlKCk=),globals())+\\n);
execfile(os.path.abspath(___code___.py))
  File , line 1, in module

  File /tmp/tmpnRD5eR/___code___.py, line 2, in module
exec compile(uRootSystem(CartanType('F4')).ambient_space() + '\n',
'', 'single')
  File , line 1, in module

  File
/usr/local/sage-4.7.1/local/lib/python2.6/site-packages/sage/misc/classcall_metaclass.py,
line 276, in __call__
return cls.__classcall__(cls, *args, **options)
  File
/usr/local/sage-4.7.1/local/lib/python2.6/site-packages/sage/combinat/root_system/root_system.py,
line 226, in __classcall__
return super(RootSystem, cls).__classcall__(cls,
CartanType(cartan_type), as_dual_of)
  File
/usr/local/sage-4.7.1/local/lib/python2.6/site-packages/sage/misc/cachefunc.py,
line 178, in __call__
w = self.f(*args, **kwds)
  File
/usr/local/sage-4.7.1/local/lib/python2.6/site-packages/sage/structure/unique_representation.py,
line 449, in __classcall__
instance = type.__call__(cls, *args, **options)
  File
/usr/local/sage-4.7.1/local/lib/python2.6/site-packages/sage/combinat/root_system/root_system.py,
line 243, in __init__
self.dual = RootSystem(self._cartan_type.dual(), as_dual_of=self);
  File
/usr/local/sage-4.7.1/local/lib/python2.6/site-packages/sage/combinat/root_system/cartan_type.py,
line 689, in dual
import type_dual
  File
/usr/local/sage-4.7.1/local/lib/python2.6/site-packages/sage/combinat/root_system/type_dual.py,
line 214, in module
class
CartanType_affine(sage.combinat.root_system.cartan_type.CartanType_affine):
AttributeError: 'module' object has no attribute 'cartan_type'

Strange...
Yours
t.d.

-- 
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] Apologize for: trouble upgrading from 4.7.1 to 4.7.2...

[Thierry Dumont]

I apologize for my preceding e-mail (04/11); I recall that I got the 
message:



sage -upgrade
...
Do you want to continue [y/N]? y
There are uncommitted changes in the Sage root repository. Aborting.


Actually, this was the first time I got this message (I have made at 
least 5 upgrades in the same directory, at each new release).


Fixing this is certainly well known for those who are ok with mercurial:
1) go in SAGE_ROOT
2)sage -hg commit -u root
(replace root by the owner).


BUT: *why* did I got this message ? hg shows the ./sage script as 
modified in SAGE_ROOT. But okay, I modified it a long time ago.

(this is not very important!)

Yours
t.d.


--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] trouble upgrading from 4.7.1 to 4.7.2...

[Thierry Dumont]

Hello,

sage -upgrade
...
Do you want to continue [y/N]? y
There are uncommitted changes in the Sage root repository. Aborting.


1) I did not change anything in the Sage repository since the 
installation of 4.7.1, except the sage script in SAGE_ROOT.

I did not get this message at the preceeding upgrades... So what?

2) issuing sage -hg diff  in SAGE_ROOT gives:

--- a/sage  Thu Aug 11 09:50:09 2011 +
+++ b/sage  Fri Nov 04 07:23:15 2011 +0100
@@ -1,7 +1,7 @@
 #!/usr/bin/env bash

 # Set SAGE_ROOT to the location of the sage install.
-SAGE_ROOT=.
+SAGE_ROOT=/usr/local/sage-4.7.1/

 CUR=`pwd`   # save the current directory, so can change back after 
startup


which seems rather normal...


So, what can I do?

Yours

t.

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] Picture visualization in the notebook ?

[Thierry Dumont]

Dear sage-support,


Is there a possibility ti visualize a Picture in the notebook ?

For example: we read a .png image, and we would like to modify it, treat
it and so on (we know how to to this with sage); at the end visualize
the result in the notebook (we do not know how to do this...).

yours
t.d.

-- 
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] which system is called ?

[Thierry Dumont]

For a small introduction to Sage, I want to show what is really called, 
for a set of problems, depending on, say, the field we use:


Example:
from sage.misc.citation import get_systems
A=Matrix(RDF,[[1,3,2],[1,4,2],[0,5,2]])
B=vector(RDF,[1,2,3])
get_systems ('A.solve_left(B)')
['numpy', 'scipy']
okay!
but:
get_systems ('A\B')
[]

Why ? we do not use the LU factorization of numpy in the second case?


Other example:

A=Matrix(QQ,[[1,3,2],[1,4,2],[0,5,2]])
B=vector(QQ,[1,2,3])
get_systems ('A.solve_left(B)')
[]
It seems that linbox would be the good tool.


Is it because get_systems do not give the good answer? In that case, how 
to make a trace of what is really called?


Yours
t.d.



--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] libgmp bug?

[Thierry Dumont]

Dear Colleagues,
Using 4.7, we got the following bug,... Sage-4.7 was obtained by 
upgrading from source the last preceeding version.

Here is the gdb run.
Yours
t.d.


germoni@damysos:~$ sage -gdb
--
| Sage Version 4.7, Release Date: 2011-05-23 |
| Type notebook() for the GUI, and license() for information.|
--
/usr/local/sage/local/bin/sage-ipython
GNU gdb (GDB) 7.1-ubuntu
Copyright (C) 2010 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
http://gnu.org/licenses/gpl.html
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type show copying
and show warranty for details.
This GDB was configured as x86_64-linux-gnu.
For bug reporting instructions, please see:
http://www.gnu.org/software/gdb/bugs/...
Reading symbols from /usr/local/sage/local/bin/python...done.
[Thread debugging using libthread_db enabled]
Python 2.6.4 (r264, May 28 2011, 10:37:01)
[GCC 4.4.3] on linux2
Type help, copyright, credits or license for more information.
sage: var(x y)
(x, y)
sage:
parametric_plot3d((x,y,1-(x^2+y^2)/6),(x,-1,1),(y,-1,1),opacity=.6).show(viewer=tachyon,aspect_ratio=1)

Program received signal SIGILL, Illegal instruction.
0x72265dd3 in mpn_popcount ()
   from /usr/local/sage/local/lib/libgmp.so.3
(gdb)

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] sage -upgrade problem...

[Thierry Dumont]

It seems that there is a problem with content.wuala.com.

sage -upgrade



Automatically selected server content.wuala.com 
(http://content.wuala.com/contents/phatsphere/edoras/sage-mirror/).
Downloading packages from 
'http://content.wuala.com/contents/phatsphere/edoras/sage-mirror//spkg'.

Reading package lists
404 Error: Package server 
'http://content.wuala.com/contents/phatsphere/edoras/sage-mirror//spkg' 
not found.

Abort.

I am in 4.6.2

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] sage upgrade on MacOs, Japanese envirinment.

[Thierry Dumont]

One of my colleagues uses a Macintosh with a Japanese environment.
He cannot upgrade sage (4.6.1).Here is a transcription of what happens.
Any idea?


gcc version 4.2.1 (Apple Inc. build 5664)

abort: unknown encoding: x-mac-japanese, please check your locale settings
pulling from /Applications/sage/spkg/build/extcode-4.6.2
searching for changes
adding changesets
transaction abort!
rollback completed
abort: unknown encoding: x-mac-japanese, please check your locale settings
abort: merging with a working directory ancestor has no effect
abort: unknown encoding: x-mac-japanese, please check your locale settings
abort: unknown encoding: x-mac-japanese, please check your locale settings

real0m0.445s
user0m0.250s
sys 0m0.181s
sage: An error occurred while installing extcode-4.6.2
Please email sage-devel http://groups.google.com/group/sage-devel
explaining the problem and send the relevant part of
of /Applications/sage/install.log.  Describe your computer, operating
system, etc.
If you want to try to fix the problem yourself, *don't* just cd to
/Applications/sage/spkg/build/extcode-4.6.2 and type 'make check' or
whatever is appropriate.
Instead, the following commands setup all environment variables
correctly and load a subshell for you to debug the error:
(cd '/Applications/sage/spkg/build/extcode-4.6.2' 
'/Applications/sage/sage' -sh)
When you are done debugging, you can type exit to leave the
subshell.
make: *** [installed/extcode-4.6.2] Error 1


  Je ne sais pas quoi faire...

Amities,
Kenji




 - http://math.univ-lyon1.fr/intranet/spip.php?article453



 ** Elections au CNU. **


 Objet : Elections C.N.U. 2011.

 2011 étant l'année de renouvellement des sections CNU, des élections sont
 organisées. Les enseignants-chercheurs sont électeurs de droit et
 figureront sur les listes d'électeurs. Les directeurs de recherche et les
 chargés de recherche relevant du décret n° 83-1260 du 30-12-1983 désirant
 être électeurs devront répondre aux conditions citées par la circulaire du
 23-12-2010 en son chapitre I paragraphe 1, et demander leur inscription
 sur
 les listes électorales suivant l'annexe IV de la circulaire au plus tard
 le
 3 février 2011.

 La situation des électeurs s'appréciant au 31/12/2010, nous nous
 permettons d'adresser ce message aux (...)



 ___
 Math mailing list
 m...@math.univ-lyon1.fr
 https://math.univ-lyon1.fr/cgi-bin/mailman/listinfo/math



-- 
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

Re: [sage-support] Re: Having trouble getting *ultra* high precision in this infinite sum calculation...

[Thierry Dumont]

Why not compute in QQ? Sage  is done for this...:

sage: s1=sum(10^(-q^2/1) for q in range(-1,1))
(wait some time...)
sage: s2=sum(10^(-k^2/1) for k in range(-2,2))
(drink a large cup of coffee)..

then:
sage: t=s1-s2

sage: float(t)
0.0
ok, let's try more precise:

sage: R=RealField(1000)
sage: R(t)
-1.02019. e-1000

e-1000! ah, ah!

sage: R=RealField(100)
sage: R(t)
-1.0201972272249067475972377296e-1

ok.
sage: R=RealField(53)
sage: R(t)
-1.02019722722491e-1
ok!

BUT:
sage: RDF(t)
0.0

RDF and RealField(53) have the same mantissa, but not the same exposant 
size!


Yours
t.d.

Le 13/12/2010 09:47, Simon King a écrit :

Hi Chris,

disclaimer: I am no expert in numerics.

On 13 Dez., 07:34, Chris Seberinocseber...@gmail.com  wrote:

Why isn't the error improving as I increase the number of terms that
are summed?  Am I doing something wrong in Sage?  (Yes it is possible
that this infinite sum converges unimaginably slowly so I wanted to
check first I wasn't doing something dumb.)


I think you use a precision that is too small

The summands are of course very small. So, comparing the two results
in a field with 53 digits precision may be pointless:
   sage: sum(10^(-k^2/1.0) for k in range(-2,2)) ==
sum(10^(-k^2/1.0) for k in range(-1,1))
   True

Note that the denominator 1.0 in your exponent belongs to RR,
which by default has 53 digits precision. Let us raise it to 6000:
   sage: d = RealField(6000)(1)
   sage: d.precision()
   6000

Unfortunately, the sums are now taking a very long time to compute.
But we are only interested in their difference.
So, it suffices to do (still taking about one minute):
   sage: 2*sum(10^(-k^2/d) for k in range(1,2))
   2.02019722722490674759723772962542922944721452394745083...e-1

So there is a small progress in the summation! But I have no idea
whether at the end of the day the small progress will be enough to
cover the big difference -1.27897692436818e-13 to your theoretical
result.

Cheers,
Simon



--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

[sage-support] Lot of segfaults...

[Thierry Dumont]

Hello,

On our Sage server, we have a lot a students doing simple computer algebra.
Our version of Sage is 4.5.3 on Debian Lenny.

We have a lot of segfaults in maxima:


2329777.996283] maxima[26150]: segfault at 7fff13abfa30 ip 7fa21cac8921 
sp 7fff13abfa30 error 6 in libgc.so.1.0.3[7fa21caaf000+26000]
[2329778.540890] maxima[26124]: segfault at 7fffc8305ff8 ip 7f9d89ea792e 
sp 7fffc8306000 error 6 in libc-2.7.so[7f9d89e38000+14a000]
[2329780.210588] maxima[26098]: segfault at 7fffda913ff8 ip 7fbee7e481d2 
sp 7fffda914000 error 6 in libecl.so.10.2.1[7fbee7d3d000+1a3000]
[2329781.020317] maxima[26340]: segfault at 72ae3fb0 ip 7f2774f7c921 
sp 72ae3fb0 error 6 in libgc.so.1.0.3[7f2774f63000+26000]
[2340141.918631] maxima[30437]: segfault at 7fff46e0bfd8 ip 7f7f88168f28 
sp 7fff46e0bf90 error 6 in libecl.so.10.2.1[7f7f88089000+1a3000]


hundreds of lines like this one...

Yours
t.d.

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

Re: [sage-support] Re: Arbitrary precision in Cython NumPy?

[Thierry Dumont]

Le 18/09/2010 16:31, KvS a écrit :

Hi again,

I hope you don't mind me bumping this thread one more time. I started
experimenting with trying a few things for fast arbitrary precision
computations using Cython. Above it was suggested to use MPFR
directly, so without the RealNumber wrapper, as the fastest way. Here
is a bit of code that does the same thing (computing x**2+i*x for
input x and i in range(10**7)) using RealNumber, using doubles and
using MPFR directly:



What surprises/disappoints me a bit is that both RealNumber and MPFR
directly are a factor 100 slower than using doubles, even though I'm
using RealField(53). Is this something I just have to live with, i.e.
computations with doubles are somehow just more optimized (?), or did
I do something wrong/is there something I can do to improve the speed
of the RealNumber/MPFR directly variations?



Hello,

There is no way to work around this:

-RDF (Double Precision numbers) use the processor arithmetic
-RealField(x) use the MPFR library.

In the first case, Sage (python, cython) passes the operations to the 
processor which performs each operation in 1 (or may be 2) clock cycles.
In the second case, MFR, will call a (may be small) routine to perform 
the task. Even if MPFR is perfectly optimized, this is much more costly.
Use MPFR when you need high precision, whatever is costs, or a special 
rounding method.


Even worst (or better): the processor implements pipelining, parallelism 
and so on.


Something I find very spectacular:

def prod(a,b):
for i in range(1,1):
a*b
c=random_matrix(RDF,1000)
d=random_matrix(RDF,1000)

prod(c,d)

try it; count the operations (10^9 multiplications). On my computer 
(3ghz), prod(c,d) takes 0.18 second: that is to say 5 Gigaflops! (or 10 
is you take account of the additions): more than 1 operation by clock 
cycle. This because we use here: 1) the processor fpu, 2) the blas 
Atlas, which allows the processor to perform at full speed (and 
minimizes cache misses).


Now, change RDF to RR. It takes 421 seconds to perform prod(c,d)  on my 
computer: that is to say *2000* times more than with RDF numbers, 
because 1) we do not use the fpu of the processor, 2) there is no 
optimization of the matrix product (and a lot of cache misses also!).


For number crunching use the processor's fpu (RDF) directly!

Yours
t.d.




Many thanks, Kees



--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] How to install a customized version of sagenb?

[Thierry Dumont]

I want to patch sagenb (to add ldap identification). I did it some time 
ago by modifying notebook.py, avatars.py in in a cloned verion of Sage.


Now that we have a separate package sagenb, what is the best way to 
proceed ? I can download sagenb, untar it, make my modifications and then???


Or is there an other way to proceed?

I just need to change notebook.py, avatars.py and to add on file nldap.py.

Yours

t.d.

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

Re: [sage-support] Notebook Login and registration of users : Legacy DB or LDAP

[Thierry Dumont]

Le 20/06/2010 16:14, Patrick ABOU BAKAR a écrit :

I know that SAGE contains its own installation of pretty much of all
the modules it depends on..

My assumption is that there is a table that holds the username and
password of the users as they sign up..

What SGBD is used for it? (SQLite, PostreSQL, MySQL)

I search but couldn't find any collaborative work on how to use a
legacy database of users (username, password) to authenticate users
when working on Notebook... Or maybe the usage of LDAP or anything of
the sort..

Any directions or answers would be appreciated.

Thanks for your help!



Hello,

I have a patch for the notebook which allow to use LDAP (Open Ldap, 
Active Directory,...). We use it in our University server ( 
http://sage-math.univ-lyon1.fr ). If you are intersted, I can help you 
installing it.

Yours
t.d.

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

Re: [sage-support] Re: XPPAUT and Sage

[Thierry Dumont]

As a marginal remark, PyDSTools use Radau code from Hairer and Wanner, 
which a much more robust method than all what is included in GSL (and 
Scipy). The tools for integrating ODEs available in Sage are far from 
being the best available.


t.d.

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

Re: [sage-support] Re: Using LiE

[Thierry Dumont]

Le 04/06/2010 11:17, John Cremona a écrit :

Brilliant suggestion.  I installed libncurses5-dev and then installing
lie worked.  (And then I found your message, afterwards).

Here is an awkward fact:  when I run our system-wide install (e.g. on
the server) the command optional_packages() fails since it is trying
to write a file into /usr/local/sage/sage-4.4.2/tmp/list which fails
as the permissions are wrong.  Is this a bug, as it appears (surely
users should be able to find out whether some optional package is
installed without needing to write any files!)?  Or something wrong in
our installation?

John

For system wide install you *must* do this as priviledged user (sudo or 
su depending on your system).

Otherwise you will not have write access to /usr/local

Thierry



--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

Re: [sage-support] Re: Using LiE

[Thierry Dumont]

To compile Lie you must (or your systeme administrator) add some tools 
on your machine:

-bison
-the ncurses lib.
On an Ubuntu machine, I installed (apt-get): libncurses5, libncursesw5 , 
ncurses-base ncurses-bin and bison. It will take 5 minutes to your 
administrator. Then it will be ok.


t.d.


Le 03/06/2010 21:17, William Stein a écrit :

On Thu, Jun 3, 2010 at 12:07 PM, John Cremonajohn.crem...@gmail.com  wrote:

Just to chip in as the administrator of the machine Bruce is using
(which belongs to me and Bill Hart), and of its sage server:   I have
tried to install the lie spkg (which worked before, i.e. with 4.4.1)
but for some reason it failed.  I probably have permissions wrong
somewhere.  Sorry, Bruce.

Here's the error: after compiling lie it says

gcc  -o Lie.exe lexer.o parser.o non-ANSI.o bigint.o binmat.o
creatop.o gettype.o getvalue.o init.o learn.o main.o mem.o node.o
onoff.o output.o poly.o sym.o print.o getl.o date.o static/*.o box/*.o
-lreadline -lncurses
/usr/bin/ld: cannot find -lncurses
collect2: ld returned 1 exit status
make[1]: *** [Lie.exe] Error 1
make[1]: Leaving directory `/usr/local/sage/sage-4.4.2/local/lib/lie'
make: *** [all] Error 2
Error building Lie.

real0m7.818s
user0m5.896s
sys 0m1.316s
sage: An error occurred while installing lie-2.2.2.p3

but I don't know where libncurses has gone.


Have you installed ncurses?   You have to install it system-wide I
think, e.g, some package with a name like :

libncurses5-dev - Developer's libraries and docs for ncurses



--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] RDF Sparse matrix.

[Thierry Dumont]

I have questions about RDF (and CDF) sparse matrices. How are they 
implemented?


-for dense matrices, sage uses Scipy matrices and this is transparent.

-but, how are sparse matrices (RDF,CDF) implemented?
   1) Are they  Scipy matrices ?
   2) if yes: there are different data structures for sparse matrices 
 in scipy:

 a) an intermediate version which uses a list representation,
not good for number crunching,
 b) csc and csr format, which are extremely common in numerical 
linear algebra (SuperLU uses them, and iterative methods too).


   How hare matrix(RDF,...sparse=True) stored?

If it is possible for sage to build automatically csc or csr matrices, 
then using sparse solvers is trivial. Otherwise I think it is necessary 
to build Scipy matrices (lil matrices converted to csr or csc format).


Thank you in advance.
t.d.


--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

Re: [sage-support] RDF Sparse matrix.

[Thierry Dumont]

Le 30/04/2010 18:54, Robert Bradshaw a écrit :

O

sage: m = matrix(RDF, 5, sparse=True)
sage: type(m)
type 'sage.matrix.matrix_generic_sparse.Matrix_generic_sparse'

So it's our completely generic sparse implementation, stored as a
dictionary of non-zero entries.

http://hg.sagemath.org/sage-main/file/e2ccb846f296/sage/matrix/matrix_generic_sparse.pyx#l1



If it is possible for sage to build automatically csc or csr matrices,
then using sparse solvers is trivial. Otherwise I think it is
necessary to build Scipy matrices (lil matrices converted to csr or
csc format).


We don't have support for that, but it would probably be a nice thing to
have.




Ok... may be it is not too complicated to implement: I currently 
transform C++ maps (i,j)- value to csr format in my C++ codes; it is a 
program of 10 lines. I could look at this (when I'll have time to do that).


What would be the best (if possible):
 - a conversion by hand:
Q=matrix(RDF,10,sparse=True)

A=Q.csr()   or A=csr(Q)
 -or try to make it automatically?
  That is to say, when calling a solve method (which will possibly call 
SuperLu, which needs csr storage) on Q, create the csr storage 
representation of Q?


Yours

t.




- Robert



--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

Re: [sage-support] Importing worksheet permission denied error

[Thierry Dumont]

Le 16/04/2010 17:35, William Stein a écrit :

On Friday, April 16, 2010, kcrismankcris...@gmail.com  wrote:

Has anyone ever seen this when importing worksheets?

Permission denied: '/home/notebook/sage_notebook/worksheets/kcrisman/2/
cells/37'






In this case, I saved something from my laptop on 4.3.5 to an older
server with 4.1, and this is what happened.




mhhh, didn't you change your UID or GID? This happens also if you used 
sage as sudo (and then root rewrites, and changes ownership).



I do not understand that sentence above.  Can you rewet it?

Otherwise the server

seems to work normally.  Importing this worksheet to sagenb.org was
also normal.  There is nothing particularly weird I am doing on the
worksheet itself.  Any ideas on what might cause this?

Thanks,
- kcrisman

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org





--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] Question about numerical linear algebra.

[Thierry Dumont]

I want to use RDF linear algebra.

As far as i understand, operations are implemented using numpy/scipy.
But many things differ from a direct call to scipy modules; for example:

(computing the Singular Value Decomposition of a matrix):

sage: A=matrix(RDF,[[1,3,2],[1,2,3],[0,5,2],[1, 1, 1]])
sage: U,Sig,V=A.SVD()

but, a direct call to scipy (even in sage) is:

from scipy import *
from scipy.linalg import *
A=mat('[1 3 2; 1 2 3; 0 5 2; 1, 1, 1]')
M,N=A.shape
U,s,Vh=svd(A)

 -Qusetion: is there some dictionary, some documentation about how 
the scipy functions are mapped to sage? and what can be directly used?
One can use A.LU(), A.QR()... witeh A a matrix(RDF,...), all this 
without importing explicitly anything.


-Remark:
 In the preceding examples, if you compare as given by U,Sig,V=A.SVD()
and U given by U,s,Vh=svd(A) they are transposed :-(

Yours, sincerely

t.




--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

To unsubscribe, reply using remove me as the subject.
attachment: tdumont.vcf

Re: [sage-support] Re: Inverses of Large Sparse Matrices

[Thierry Dumont]

Alec Mihailovs a écrit :

On Apr 9, 8:59 pm, William Stein wst...@gmail.com wrote:


A 5000x5000 matrix just isn't really that big, IMHO...


That's true - should work in just few seconds - I meant REALLY big
matrices - actually, sometimes such a thing should work faster even
for not that big matrices - in case if the virtual memory located on
disk is used - it is faster to write on disk directly.

Alec

But, you really want to *inverse* the matrix, or you just want to solve 
a linear system?
And how is your matrix? what is the sparsity pattern ? What are the 
coefficients (rational, floating...)...?


Yours
t.d.

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

To unsubscribe, reply using remove me as the subject.
attachment: tdumont.vcf

Re: [sage-support] Re: ndsolve??

[Thierry Dumont]

Rajeev a écrit :

Hi,
sage has gsl as one of the included packages, which is very good for
numerical solution of differential equations. have a look at examples
on the wikipage -
http://wiki.sagemath.org/interact/diffeq
'Vector Field with Runga-Kutta-Fehlberg' by Schilly is one of my
favorites. i hope it will help.
Best wishes,
Rajeev


Ok, but... this is a bit sad, but we do not have the best methods for 
integrating odes. This would be nice to have (in scipy ?) the most 
modern methods, and this would not be very difficult to implement them, 
since these method share a rather common interface with odepack (Gear's 
method). Real problems are stiff, that is to say (roughly speaking!) 
that in dU/dt=F(U), the jacobian of F has eigenvalues whose real part 
are distributed in a large negative interval (say... (-10^7, -1) for the 
classical example of the Oregonator).


Having a rather long experience in ODE solving, my conclusion is that 
the most universal methods are those of the Radau family (described in 
the book of Hairer and Wanner Solving ordinary differential equations 
(part 2). These methods are the most robust of all, but they are also 
the fastest ones, for difficult problems.


I wonder whether this would be interesting to have symplectic 
integrators, for the integration of Hamiltonian systems.


t.d.

--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

To unsubscribe from this group, send email to sage-support+unsubscribegooglegroups.com or 
reply to this email with the words REMOVE ME as the subject.
attachment: tdumont.vcf

Re: [sage-support] Easiest method to solve a PDE

[Thierry Dumont]

William Stein a écrit :
 On Fri, Dec 18, 2009 at 11:53 PM, Thierry Dumont
 tdum...@math.univ-lyon1.fr wrote:
 Carlos Córdoba a écrit :
 Hi,

 I know this is not a general mathematical forum, but I hope you can help me.
 I have this PDE:

 \frac{dB}{dt} = F(x,y,z)B(x,y,z) - G(x,y,z)\nabla B(x,y,z)

 and I don't know how to solve it numerically. What would be the easiest
 method to do it? It can be in python, but preferably in C++.

 Thanks for your help,
 Carlos

 Hi,

 1) You will not find in Sage what you are looking for; this is not the
 place and as a specialist of numerics for PDEs, I think it will remain
 like this.
 
 No it won't :-)
 
 2) You problem is not so easy: it is a first order equation of
 hyperbolic type and this is not easy to solve. For example the problem
 du/dt+du/dx=0, for which we have an exact solution is not easy to solve
 numerically (at least one must process with care to avoid
 instabilities). Your problem is a bit more complicated. Solving the heat
 equation for example is much more easy.
 3) for numerics on these first order problems, have a look at clawpack:
 it solves much more complicated problems (first order non linear
 systems- like gas dynamics for example), but it will solve easily your
 problem. You will have to know some bases on the numerical analysis of
 hyperbolic pde.
  http://www.amath.washington.edu/~claw/
 
 For the record, I recently worked with the authors to get them to
 change the license of clawpack to be GPL-compatible and know the
 people who work on that project (who are at the same university as
 me).
 
 William
 
Ok! great. My position is only that I do not think that Sage will be the 
good place to solve true life large PDE systems (industrial problems, 3d 
problems), but integrating clawpack would be great for teaching. This is 
*only* my *personal* vision of this.

A question about numerics in Sage:
what about pydstool ?
http://www.cam.cornell.edu/~rclewley/cgi-bin/moin.cgi/
(this is a tool for dynamical systems, bufurcations and so on).
Is anybody working on an integration in Sage ?

t.d.

-- 
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

Re: [sage-support] Easiest method to solve a PDE

[Thierry Dumont]

Carlos Córdoba a écrit :
 Hi,
 
 I know this is not a general mathematical forum, but I hope you can help me.
 I have this PDE:
 
 \frac{dB}{dt} = F(x,y,z)B(x,y,z) - G(x,y,z)\nabla B(x,y,z)
 
 and I don't know how to solve it numerically. What would be the easiest
 method to do it? It can be in python, but preferably in C++.
 
 Thanks for your help,
 Carlos
 
Hi,

1) You will not find in Sage what you are looking for; this is not the 
place and as a specialist of numerics for PDEs, I think it will remain 
like this.
2) You problem is not so easy: it is a first order equation of 
hyperbolic type and this is not easy to solve. For example the problem
du/dt+du/dx=0, for which we have an exact solution is not easy to solve 
numerically (at least one must process with care to avoid 
instabilities). Your problem is a bit more complicated. Solving the heat 
equation for example is much more easy.
3) for numerics on these first order problems, have a look at clawpack: 
it solves much more complicated problems (first order non linear 
systems- like gas dynamics for example), but it will solve easily your 
problem. You will have to know some bases on the numerical analysis of 
hyperbolic pde.
  http://www.amath.washington.edu/~claw/

t.d.

-- 
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
attachment: tdumont.vcf

[sage-support] Re: Need help in starting a sage server

[Thierry Dumont]

William Stein a écrit :
 On Wed, Nov 11, 2009 at 6:22 PM, Kwankyu ekwan...@gmail.com wrote:
 Hi,

 I have a problem starting a sage server after, I think, upgrading to
 Ubuntu 9.10 server edition. It just hangs at the point Generating a
 2048 bit RSA private key... See the following transcript:

 

 Any idea to fix the problem?

 
 I have seen that.  I think it's a bug in certtool.  I don't know how
 to fix the problem though.  You might try installing all
 openssl-related Ubuntu packages...
 
 William
 
 
When you say that it hangs: how long did you wait? This seems funny but
in my case, after *some*  *minutes* the generation finished (may be 5
minutes !!!).
Is not this related to a problem of entropy in the computer ? I
remember that, may be 10 years ago, it whas necessary to make some mouse
movements during the ssh keys generation procedure, as the random number
generator used the computer activity. Here I think that the generation
finished when I launched top on the machine. But I have only one
experience. Try to make some activity on the machine during the key
generation and wait (a long time).
t.d.

attachment: tdumont.vcf

smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Sage / Gap: memory size question.

[Thierry Dumont]

 One of our Sage users want to make computations with Gap, from inside Sage.
So his commands are:
sage
gap.console()

then he can read his Gap program, start to run it, but the program
aborts by lack of memory.
Classically, he uses gap directly, by typing:
 gap -o 4G   (to get 4 Giga Bytes).

Is there any way to pass such memory size option to gap.console() ?

Yours
t.d.


attachment: tdumont.vcf

smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Re: Wich directories must be in LD_LIBRARY_PATH ?

[Thierry Dumont]

William Stein a écrit :
 On Mon, Sep 28, 2009 at 5:53 AM, Thierry Dumont
 tdum...@math.univ-lyon1.fr wrote:
 Trying to solve my problem:

 
 THERE WAS AN ERROR LOADING THE SAGE LIBRARIES.  Try starting Sage from
 the command line to see what the error is
 

 
 Make it so when the remote (worksheet) user types python they get
 the sage version of Python, i.e., make it so python runs a script
 that runs sage -python.
 
 William
 

I apologize, but can you explain me a littlebit more?

Sage is in SAGE_ROOT=/sage2/sage-4.1.modif/

In sage-env I put:
LD_LIBRARY_PATH=/sage2/sage-4.1.modif/local/lib:\
/sage2/sage-4.1.modif/local/lib/gap-4.4.10/lib:\
/sage2/sage-4.1.modif/local/lib/python;export LD_LIBRARY_PATH

Starting the notebook, as user x with server_chose pool=['sa...@localhost']
I open a new worksheet, I choose python, then type: from os import *
and evaluate,
and get the THERE WAS AN ERROR LOADING THE SAGE LIBRARIES.
What is missing?

Yours
t.
attachment: tdumont.vcf

smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Re: strange Permission denied problem

[Thierry Dumont]

ma...@mendelu.cz a écrit :
 
 
 On 26 zář, 15:27, Thierry Dumont tdum...@math.univ-lyon1.fr wrote:
 ... strange only for me I hope.

 
 I have similar problem. As I undesrstand, if the user sage runs
 program sage and user sageuser is in server_pool, then .sage directory
 should be writeable for sage, but not for sageuser. Right? Otherwise
 sageuser cen delete other worksheets. With this settings I have the
 same error message (with Thierry.Dumont replaced by robert.marik, of
 course :) )
 
 Robert
 --~--~-

Ok; you have to put everithing *outside* the .sage directory (in an 
other drectory), and making this directory readable and writable by the 
group (alle users of the pool must be in a same group).
When you laucn the notebook, add directory parameter...
This will solve *this* problem, isn'it ?

The I wait for you, for the library problem.
What is your Linux distrib?

yours
t.


--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support-unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

attachment: tdumont.vcf

[sage-support] Re: strange Permission denied problem

[Thierry Dumont]

Super..
Question: can we have users of the pool on different machines now? (the 
directory beiing shared by nfs), and having the same groups on the 
different machines?

William Stein a écrit :
 On Tue, Sep 29, 2009 at 1:00 PM, ma...@mendelu.cz ma...@mendelu.cz wrote:


 On 26 zář, 15:27, Thierry Dumont tdum...@math.univ-lyon1.fr wrote:
 ... strange only for me I hope.

 I have similar problem. As I undesrstand, if the user sage runs
 program sage and user sageuser is in server_pool, then .sage directory
 should be writeable for sage, but not for sageuser. Right? Otherwise
 
 The sageuser has to be able to write to the worksheet directories.
 This is a lame aspect of the notebook design.  It's so lame, I completely
 rewrote the worksheet process interface during the last week and got
 rid of this problem -- now the users in the server_pool just have to have
 access to a common filesystem, e.g., /tmp/ and everything will work.
 This is
 
http://trac.sagemath.org/sage_trac/ticket/6983
 
 We will be switching to this new notebook code in sage-4.1.2.
 In the meantime you can try the spkg at
 
   http://sage.math.washington.edu/home/wstein/patches/sagenb/
 
 if you want.   It's version 0.1.14 right now.  To use it, you would do
 the following:
 
  (1) install it via
 
 http://sage.math.washington.edu/home/wstein/patches/sagenb/sagenb-0.1.14.spkg
 
  (2) cd to the directory where your sage_notebook is located (e.g.,
 $HOME/.sage) and type
 
 sage: import sagenb.notebook.notebook_object as n
 sage: n.notebook('sage_notebook')
 
 This will upgrade the notebook directory and launch the new server.
 You can pass all the server_pool etc options that you usuallly pass in.
 
 The above will backup your sage_notebook directory, but since this is
 not yet released, you may want to manually make your own extra backup
 with
 
 tar jcvf sage_notebook.tar.bz2 sage_notebook
 
 before doing the above, just in case.
 
 William
 
 
 sageuser cen delete other worksheets. With this settings I have the
 same error message (with Thierry.Dumont replaced by robert.marik, of
 course :) )

 Robert
 
 
 


--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support-unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

attachment: tdumont.vcf

[sage-support] Re: strange Permission denied problem

[Thierry Dumont]

William Stein a écrit :
 2009/9/29 Thierry Dumont tdum...@math.univ-lyon1.fr:
 Super..
 Question: can we have users of the pool on different machines now? (the
 directory beiing shared by nfs), and having the same groups on the
 different machines?
 
 In theory, yes, if you are willing to edit some code in notebook.py to
 set some options.  It is probably better to wait for the next release
 (i.e., a week?) and that will have the capability you want easily
 usable and documented.
 
 You may want to look at the code in sagenb/interfaces/expect.py and
 sagenb/notebook/notebook.py (that makes a WorksheetProcess) to see
 what is done.  The design is such that it will be very easy to fully
 support the functionality you're requesting soon.
 
  -- William
 

Nice, one week will be ok...
Nowadays, I have my 3 machines, and I associate each user to one single 
machine by hashing his login name... This works but it is not very serious.

(Ans I have not solved the library problem, but I'll try after tomorrow).
yours
t.

--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support-unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

attachment: tdumont.vcf

[sage-support] Re: sage notebook issue

[Thierry Dumont]

Koch Peer-Joachim a écrit :
 Hi,
 
 we are running a sage notebook for the members of our institute.
 In the moment the users can create an account login and use
 the notebook.
 One user has forgotten his password. How can I change
 the passsword of the user ?
 The notebook is started with
 nohup /../sage -python ./notebook.py 21 
 
 So how and where to place the change_password statement ?
 
 Is it possible to use pam or ldap for the users  ?
 
 Thanks, Peer
 
 --~--~-~--~~~---~--~~
 To post to this group, send email to sage-support@googlegroups.com
 To unsubscribe from this group, send email to 
 sage-support-unsubscr...@googlegroups.com
 For more options, visit this group at 
 http://groups.google.com/group/sage-support
 URLs: http://www.sagemath.org
 -~--~~~~--~~--~--~---
 
 

I have patches to use ldap (or AD):
http://math.univ-lyon1.fr/~tdumont/sage
yours
t.d.
attachment: tdumont.vcf

smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Wich directories must be in LD_LIBRARY_PATH ?

[Thierry Dumont]

Trying to solve my problem:


THERE WAS AN ERROR LOADING THE SAGE LIBRARIES.  Try starting Sage from
the command line to see what the error is



when launching the notebook with a server_pool, I added manually the path:

$SAGE_ROOT/local/lib
(with $SAGE_ROOT hard coded).

This changed nothing.

Question: what must be in  LD_LIBRARY_PATH (in sage-exe or in other
scripts)?

But my be the problem is not there.

Yours
t.d.
attachment: tdumont.vcf

smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] An other problem...

[Thierry Dumont]

Since my last mail... I changed one or two things:

  I create a directory, exactly like http://wiki.sagemath.org/SageVirtualBox
(this should not be specific to VBox).

I add my 2 users sage and sage1 to a group sageusers.
chmod g+w nbfiles

I create the notebook.py file like this:

from sage.all import *
notebook(directory='/home/sage/nbfiles',server_pool=['sa...@localhost'],open_viewer=False,address='',secure=True,port=8001,timeout=3600,ulimit='-v
 
5',accounts=True)

The only important difference with the wiki seems to be secure=True.

and I do:
sage -python notebook.py

I can connect, and create a worksheet, but when I evaluate it, I get:

Traceback (click to the left for traceback)
...
THERE WAS AN ERROR LOADING THE SAGE LIBRARIES.  Try starting Sage from
the command line to see what the error is

...???

- Comparing to the preceding mail, I just understand that sage do not 
changes the rights on nbfiles, ok.
- But why this new problem ? And how to debug it ?
(I must solve this in teh next days, otherwise my experience using sage 
with a lot a students will stop!).

Yours

t.d.


--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support-unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

attachment: tdumont.vcf

[sage-support] Re: An other problem...

[Thierry Dumont]

I answer to myself:

Considering the script sage-native-execute, I add :

echo LOOKATPATH
echo $LD_LIBRARY_PATH

before

$@

Then launching sage, I get:

* Open your web browser to https://localhost:8001  *
*  *

There is an admin account.  If you do not remember the password,
quit the notebook and type notebook(reset=True).
/sage2/sage-4.1.modif/local/lib/python2.6/site-packages/twisted/persisted/sob.py:12:
 
DeprecationWarning: the md5 module is deprecated; use hashlib instead
   import os, md5, sys
No remote temporary directory (option server_tmpdir) specified, using 
/tmp/ on sa...@localhost
LOOKATPATH

... and nothing. The LD_LIBRARY_PATH is *empty*, and thus also 
SAGE_ORIG_LD_LIBRARY_PATH.

Is it the problem?

Yours
t.d.

Thierry Dumont a écrit :
 Since my last mail... I changed one or two things:
 
   I create a directory, exactly like http://wiki.sagemath.org/SageVirtualBox
 (this should not be specific to VBox).
 
 I add my 2 users sage and sage1 to a group sageusers.
 chmod g+w nbfiles
 
 I create the notebook.py file like this:
 
 from sage.all import *
 notebook(directory='/home/sage/nbfiles',server_pool=['sa...@localhost'],open_viewer=False,address='',secure=True,port=8001,timeout=3600,ulimit='-v
  
 5',accounts=True)
 
 The only important difference with the wiki seems to be secure=True.
 
 and I do:
 sage -python notebook.py
 
 I can connect, and create a worksheet, but when I evaluate it, I get:
 
 Traceback (click to the left for traceback)
 ...
 THERE WAS AN ERROR LOADING THE SAGE LIBRARIES.  Try starting Sage from
 the command line to see what the error is
 
 ...???
 
 - Comparing to the preceding mail, I just understand that sage do not 
 changes the rights on nbfiles, ok.
 - But why this new problem ? And how to debug it ?
 (I must solve this in teh next days, otherwise my experience using sage 
 with a lot a students will stop!).
 
 Yours
 
 t.d.
 
 
  


--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support-unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

attachment: tdumont.vcf

[sage-support] On the problem error loading the sage libraries

[Thierry Dumont]

Some days (and weeks) ago Serge Salamanka posted a message about the
message error loading the sage libraries he got when trying to use a
pool of servers. I cannot find any answer in the sage-support archive.

Is it fixed ? I have exactly the same problem on my machine (Debian
Lenny, AMD 64).

Yours

t;d.
attachment: tdumont.vcf

smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] server_pool problem.

[Thierry Dumont]

Hi,

something I do not understand.

0) I have 2 users: sage and sage1 (on the same machine).

1) Logged in as sage, I can do ssh sa...@localhost some_command
without passwd (authorized_keys2 in ~sage1/.ssh)

2) Now I launch the notebook:
notebook(address='',secure=True,port=8001,timeout=3600,ulimit='-v
5',accounts=True,server_pool=['sa...@localhost'])

3) With my browser I go to https://localhost:8001
Everything is, ok *and* I can log in the notebook.

4) The I try to create a Worsheet, and type
x=2
Evaluate:

Traceback (click to the left for traceback)
...
IOError: [Errno 13] Permission denied:
'/home/sage/.sage/sage_notebook/worksheets/tdumont/0/code/1.py'

So I am trying to write in /home/sage, but as sage1...I understand
this... But what should I do ?

t.d.

attachment: tdumont.vcf

smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] problem with server_pool.

[Thierry Dumont]

I am trying to use Sage (3.4) with a server pool *and* ldap identification.

- As user sage I run the notebook with the following parameters:
notebook(port=8001,secure=True,address='',server_pool=['sa...@localhost'],open_viewer=False,accounts=True)

sage1 is an other Unix user, and I can ssh from sage user to sage1
without password.

- Suppose that this is the first time I log in sage; I give my login
and password, look for them in the ldap directory and, if everything is
ok, I proceed exactly as for a public server, that is to say that Sage
creates a new user (a new Sage user).

I can verify that Sage is running as sage1.

But the new user is created in the .sage/sage-notebook/worksheets of the
*sage* user (not the *sage1*).
I can connect to the server, but then, and for me it is not surprising,
I have no rights and can do nothing:

Example:

2+1
(evaluate)
Traceback (click to the left for traceback)
...
IOError: [Errno 13] Permission denied:
'/home/sage/.sage/sage_notebook/worksheets/Thierry.Dumont/2/code/1.py'
---


On the public sage-server (your server), it seems that sage is running
as worksheet (that is what says: system('whoami')), but the home
directories belong to sage. So there is something I do not understand.

An other (small) problem: I had to unsubscribe from sage-support list
for some time. I subscribed again (by mail), but I do not receive
answers to my mails, as before unsubscribing. But this not important, I
can look at the messages on the web.

Yours
t.d.

-
French universities are on a permanent strike!
Have a look at the International Call:
http://math.univ-lyon1.fr/appel
-


begin:vcard
fn:Thierry Dumont
n:Dumont;Thierry
org;quoted-printable:CNRS - Universit=C3=A9 Lyon 1. / Villeurbanne France.;Institut Camille Jordan
adr:;;43 Bd du 11 Novembre;Villeurbanne Cedex;F;69621;France
email;internet:tdum...@math.univ-lyon1.fr
title;quoted-printable:Ing=C3=A9nieur de Recherche/Research Ingineer
tel;work:(33) 4 72 44 85 23
x-mozilla-html:FALSE
url:http://math.univ-lyon1.fr/~tdumont
version:2.1
end:vcard



smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Security problem...

[Thierry Dumont]

I am always preparing me Sage University wide server..

I think I have a big security problem:

Sage create users (looking for this in my ldap server, but this changes 
nothing from a public server where every one can create an account).
One an account is created, I can access and *destroy* all other 
accounts; it is sufficient for this to log in  on the notebook, then  to 
launch the python interpreter; at is point I can do:

system('ls -l  /scratch/sage/.sage/sage_notebook/worksheets/')

and worse:

system('rm -rf  /scratch/sage/.sage/sage_notebook/worksheets/foo')

Mmmpf...

Is there any way to avoid this ?

Yours
t.d.



--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support-unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

begin:vcard
fn:Thierry  Dumont
n:Dumont;Thierry 
org;quoted-printable;quoted-printable:Universit=C3=A9 Lyon 1  CNRS.;Institut Camille Jordan -- Math=C3=A9matiques / Mathematics.
adr:;;43 Bd du 11 Novembre.;Villeurbanne;;69621;France
email;internet:tdum...@math.univ-lyon1.fr
title;quoted-printable:Ing=C3=A9nieur de Recherche / Research Engineer.
tel;work:04 72 44 85 23.
tel;fax:04 72 44 80 53
x-mozilla-html:FALSE
url:http://math.univ-lyon1.fr/~tdumont
version:2.1
end:vcard

[sage-support] avatars.py (and others): which one ?

[Thierry Dumont]

I need to make changes to avatars.py

I find 3 versions of this script in the sage tree:

./devel/sage-main/build/lib.linux-x86_64-2.5/sage/server/notebook/avatars.py
./devel/sage-main/build/sage/server/notebook/avatars.py
./devel/sage-main/sage/server/notebook/avatars.py

(the same is true for notebook.py and many other scripts).

Which one is used by the notebook?

I cannot find where all this is described in the documentation... can
you give me a hint?

Yours,
t.d.

-
French universities are on a permanent strike!
Have a look at the International Call:
http://math.univ-lyon1.fr/appel
-


begin:vcard
fn:Thierry Dumont
n:Dumont;Thierry
org;quoted-printable:CNRS - Universit=C3=A9 Lyon 1. / Villeurbanne France.;Institut Camille Jordan
adr:;;43 Bd du 11 Novembre;Villeurbanne Cedex;F;69621;France
email;internet:tdum...@math.univ-lyon1.fr
title;quoted-printable:Ing=C3=A9nieur de Recherche/Research Ingineer
tel;work:(33) 4 72 44 85 23
x-mozilla-html:FALSE
url:http://math.univ-lyon1.fr/~tdumont
version:2.1
end:vcard



smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Re: avatars.py (and others): which one ?

[Thierry Dumont]

Timothy Clemans a écrit :
 ./devel/sage-main/sage/server/notebook/avatars.py
 
Ok thank you...

 Usually one clones the main branch sage --clone nameofclone
 sage -br takes your changes live
 

I do not really understand this. I apologize, but is there a link to
some place where it is explained?
Do tou mean:
sage --clone nameofclone= creates a copy
the we make changes in nameofclone
and sage -br  tranfers the changes in the main branch ?

Yours
t.

 On Fri, Mar 27, 2009 at 1:25 AM, Thierry Dumont
 tdum...@math.univ-lyon1.fr wrote:
 I need to make changes to avatars.py

 I find 3 versions of this script in the sage tree:

 ./devel/sage-main/build/lib.linux-x86_64-2.5/sage/server/notebook/avatars.py
 ./devel/sage-main/build/sage/server/notebook/avatars.py
 ./devel/sage-main/sage/server/notebook/avatars.py

 (the same is true for notebook.py and many other scripts).

 Which one is used by the notebook?

 I cannot find where all this is described in the documentation... can
 you give me a hint?

 Yours,
 t.d.

 -
 French universities are on a permanent strike!
 Have a look at the International Call:
 http://math.univ-lyon1.fr/appel
 -



 
 --~--~-~--~~~---~--~~
 To post to this group, send email to sage-support@googlegroups.com
 To unsubscribe from this group, send email to 
 sage-support-unsubscr...@googlegroups.com
 For more options, visit this group at 
 http://groups.google.com/group/sage-support
 URLs: http://www.sagemath.org
 -~--~~~~--~~--~--~---
 


-- 

-
French universities are on a permanent strike!
Have a look at the International Call:
http://math.univ-lyon1.fr/appel
-


begin:vcard
fn:Thierry Dumont
n:Dumont;Thierry
org;quoted-printable:CNRS - Universit=C3=A9 Lyon 1. / Villeurbanne France.;Institut Camille Jordan
adr:;;43 Bd du 11 Novembre;Villeurbanne Cedex;F;69621;France
email;internet:tdum...@math.univ-lyon1.fr
title;quoted-printable:Ing=C3=A9nieur de Recherche/Research Ingineer
tel;work:(33) 4 72 44 85 23
x-mozilla-html:FALSE
url:http://math.univ-lyon1.fr/~tdumont
version:2.1
end:vcard



smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Printing /debugging

[Thierry Dumont]

 As I want to modify Sage (introducing ldap identification), I need to
modify some parts of Sage, mainly avatars.py.

 It is not yet working correctly :-( and for debugging I put some
print instructions in the code; for example in the class FailedLogin
I put a print statement in __init__.

I start Sage like this:  sageparameter_file

And some message are printed:

2009-03-23 09:15:47+0100 [-] Log opened.
2009-03-23 09:15:47+0100 [-] twistd 8.1.0
.
.
2009-03-23 09:16:14+0100 [HTTPChannel,0,134.214.156.6]

*but* nothing else, even when I simulate a failed login, for example.
Nothing from my print statements.

Questions (may be stupid):

1) where do my print actually writes?
2) Log opened: this means that there is a log file, certainly. But where ?

Yours
t.d.


-
French universities are on a permanent strike!
Have a look at the International Call:
http://math.univ-lyon1.fr/appel
-


begin:vcard
fn:Thierry Dumont
n:Dumont;Thierry
org;quoted-printable:CNRS - Universit=C3=A9 Lyon 1. / Villeurbanne France.;Institut Camille Jordan
adr:;;43 Bd du 11 Novembre;Villeurbanne Cedex;F;69621;France
email;internet:tdum...@math.univ-lyon1.fr
title;quoted-printable:Ing=C3=A9nieur de Recherche/Research Ingineer
tel;work:(33) 4 72 44 85 23
x-mozilla-html:FALSE
url:http://math.univ-lyon1.fr/~tdumont
version:2.1
end:vcard



smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Re: Disabled person using SAGE

[Thierry Dumont]

meitnik a écrit :
 Hi all,

 I am legally blind, legally deaf, some limited finger mobility, and
 some learning disabilities too (all from Rubella). I enjoy mathematics
 and programming but due to my limited income Mathematica is just out
 of my reach even for the Home edition. A friend told me about SAGE.
 Cool!

 Andrew

 --~--~-~--~--
Andrew (and the others),

There is a group of people in my University who would like to integrate 
in Sage the possibility to produce output in braille maths.
It seems that it would be possible to use the braille transcriptor NAT
(see: http://natbraille.free.fr or http://liris.cnrs.fr/nat) The main 
author is Bruno Mascret (he will receive this e-mail, too).
I try to translate what B. Mascret wrote to me:

This would give to braille user students at last the possibility to use 
a tool allowing them to be autonomous in their work.

If there are people interested, please contact B. Mascret 
(bmasc...@free.fr) or me.

Yours
Thierry.



--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to 
sage-support-unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

begin:vcard
fn:Thierry  Dumont
n:Dumont;Thierry 
org;quoted-printable;quoted-printable:Universit=C3=A9 Lyon 1  CNRS.;Institut Camille Jordan -- Math=C3=A9matiques / Mathematics.
adr:;;43 Bd du 11 Novembre.;Villeurbanne;;69621;France
email;internet:tdum...@math.univ-lyon1.fr
title;quoted-printable:Ing=C3=A9nieur de Recherche / Research Engineer.
tel;work:04 72 44 85 23.
tel;fax:04 72 44 80 53
x-mozilla-html:FALSE
url:http://math.univ-lyon1.fr/~tdumont
version:2.1
end:vcard

[sage-support] Another twistd server is running, PID 8301

[Thierry Dumont]

Hello,

As (maybe) you remember I starting the installation of my university
wide Sage server.
One idea is to launch 2 Sage servers on each machine (one machine has
32Gb ram and 4x2 cores). So: I create a sage user, in /home/sage I
create two directories.

In the first one I do:
sagenotebook(port=8001,secure=True,address='',open_viewer=False,accounts=True)


ok... cd to the second directory; there I do:
sagenotebook(port=8002,secure=True,address='',open_viewer=False,accounts=True)

and I get:


Another twistd server is running, PID 8301

This could either be a previously started instance of your application or a
different application entirely. To start a new one, either run it in
some other
directory, or use the --pidfile and --logfile parameters to avoid clashes.


same problem if I change the definition of the pidfile and the logfile.


*But*, anyway, is it the good way to do? I remember something about
launching more than one Sage process, but cannot find this in the doc.

Yours
t.d.


-
French universities are on a permanent strike!
Have a look at the International Call:
http://math.univ-lyon1.fr/appel
-


begin:vcard
fn:Thierry Dumont
n:Dumont;Thierry
org;quoted-printable:CNRS - Universit=C3=A9 Lyon 1. / Villeurbanne France.;Institut Camille Jordan
adr:;;43 Bd du 11 Novembre;Villeurbanne Cedex;F;69621;France
email;internet:tdum...@math.univ-lyon1.fr
title;quoted-printable:Ing=C3=A9nieur de Recherche/Research Ingineer
tel;work:(33) 4 72 44 85 23
x-mozilla-html:FALSE
url:http://math.univ-lyon1.fr/~tdumont
version:2.1
end:vcard



smime.p7s
Description: S/MIME Cryptographic Signature

[sage-support] Re: Another twistd server is running, PID 8301

[Thierry Dumont]

William Stein a écrit :
 On Wed, Mar 18, 2009 at 5:46 AM, Thierry Dumont
 tdum...@math.univ-lyon1.fr wrote:
 Hello,

 As (maybe) you remember I starting the installation of my university
 wide Sage server.
 One idea is to launch 2 Sage servers on each machine (one machine has
 32Gb ram and 4x2 cores). So: I create a sage user, in /home/sage I
 create two directories.

 In the first one I do:
 sagenotebook(port=8001,secure=True,address='',open_viewer=False,accounts=True)
 
 You must do
 
 notebook(dir='local_notebook',
 port=8001,secure=True,address='',open_viewer=False,accounts=True)
 

thanks, but... it does not work (or I did not understand); the error
message is:
TypeError: notebook_twisted() got an unexpected keyword argument 'dir'
yours
t.d.
 
 


-- 

-
French universities are on a permanent strike!
Have a look at the International Call:
http://math.univ-lyon1.fr/appel
-


begin:vcard
fn:Thierry Dumont
n:Dumont;Thierry
org;quoted-printable:CNRS - Universit=C3=A9 Lyon 1. / Villeurbanne France.;Institut Camille Jordan
adr:;;43 Bd du 11 Novembre;Villeurbanne Cedex;F;69621;France
email;internet:tdum...@math.univ-lyon1.fr
title;quoted-printable:Ing=C3=A9nieur de Recherche/Research Ingineer
tel;work:(33) 4 72 44 85 23
x-mozilla-html:FALSE
url:http://math.univ-lyon1.fr/~tdumont
version:2.1
end:vcard



smime.p7s
Description: S/MIME Cryptographic Signature