[sage-devel] Re: Error installing package gcc-5.4.0

[Travis Scrimshaw]

What version of Sage are you trying to install? I've found I need both

https://trac.sagemath.org/ticket/24088
https://trac.sagemath.org/ticket/24092

with 8.1.beta9.

It looks like you have gcc 7.2.0 on your system. Are g++ and gfortran the 
same version? Perhaps try rebuilding Sage (after running "make distclean") 
if all of those are version 7.2.0 to avoid building Sage's gcc.

Best,
Travis

On Friday, November 3, 2017 at 1:46:08 PM UTC+10, Jon King wrote:
>
> I'm running Ubuntu 17.04, and I've been having some issues installing 
> sage. I've gone through many bug fixes where "make" wouldn't do anything 
> but shoot an error. I just tried it again, and it was installing for about 
> 10-15 minutes when it crashed. Attached is the log file.
>

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

Re: [sage-devel] Sage Days next July in Zaragoza (Spain)

[Jeroen Demeyer]

On 2017-10-31 13:01, mmarco wrote:

So, would someone be interested in coming and giving a short course on
one of those subjects (or to suggest other subjects that you think could
be interesting)? Ideally, each course would consist on a few lectures
plus some practical sessions.


I might be able to do the Cython session. That should also include how 
to communicate with external libraries on the C/C++ level.


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

[sage-devel] Segfault in Expression.coefficients involving symbolic functions

[Eric Gourgoulhon]

Hi,

In Sage 8.1.beta8 and all previous versions (at least down to 7.5.1), we 
have

sage: s = function('f')(x) * x^2
sage: s.coefficients(x)  # OK
[[f(x), 2]]
sage: s = diff(function('f')(x), x) * x^2
sage: s.coefficients(x)  # Bam!
Erreur de segmentation (core dumped)

Is this a known bug?

Eric.

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

[sage-devel] Re: Sage Days next July in Zaragoza (Spain)

[Travis Scrimshaw]

> We will be organizing the next meeting of the spanish network of computer 
> algebra in Zaragoza (Spain); the proposed dates are July 4th-6th. As a 
> satellite event we plan to organize also a Sage Days (right before, or 
> right after the meeting), in the form of a school oriented to introducing 
> people to Sage development.
>

Thank you for organizing it this. 

>
> The idea is to cover those aspects of the Sage development workflow that 
> are not usually covered in tutorials. Things like:
>
> - The sage development workflow (Trac, git, doctests...)
> - The coercion model
> - Implementation of Parents and elements
> - The category framework
> - Cython
> - Communication with external packages (libraries and pexpect)
> - ...
>
>
> The idea is to help people (mainly graduate students and postdocs) to make 
> the step from "I have writen some sage code for my research, but I keep it 
> to my own, or just share it as a notebook" to "I contribute my code to the 
> Sage codebase".  
>
> We don't know the exact amount of funding that we will have available, but 
> it is likely that we can pay the expenses of the speakers (and maybe also 
> have some financial support for attendants).
>
> So, would someone be interested in coming and giving a short course on one 
> of those subjects (or to suggest other subjects that you think could be 
> interesting)? Ideally, each course would consist on a few lectures plus 
> some practical sessions.
>
> I am interested in coming. I could give a course for any of those on your 
list (although there are definitely people who are better at each of those 
than me).

Best,
Travis

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

[sage-devel] Re: Sage Days next July in Zaragoza (Spain)

[Eric Gourgoulhon]

Hi, 

Le mardi 31 octobre 2017 13:01:44 UTC+1, mmarco a écrit :
>
>
> The idea is to cover those aspects of the Sage development workflow that 
> are not usually covered in tutorials. 
>


This sounds a very good idea!

I could possibly take part in these Sage Days and give an introductory 
course on the first three items of your list. This may depend on the dates: 
due to another commitment in a summer school, only the slot before July 4th 
may work for me. Anyway, please keep me posted. 

Best wishes,

Eric.  

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

Re: [sage-devel] floor division for integer matrices

[Thierry]

Hi,

if you want to make explicit that you apply the operation to each entry,
you can still do:

sage: A = Matrix(ZZ,2,[3,3,3,3]) 
sage: A.apply_map(lambda x:x//3)
[1 1]
[1 1]
sage: _.parent()
Full MatrixSpace of 2 by 2 dense matrices over Integer Ring


This method works for many things, that might not make sense as an
operation on matrices:

sage: A.apply_map(arctan)
[arctan(3) arctan(3)]
[arctan(3) arctan(3)]

It also allows not to be confused when applying an operator that is well
defined on the matrix, but does not have the same meaning than applying it
on each entry:

sage: A.apply_map(exp)
[e^3 e^3]
[e^3 e^3]

sage: exp(A)
[1/2*e^6 + 1/2 1/2*e^6 - 1/2]
[1/2*e^6 - 1/2 1/2*e^6 + 1/2]

Ciao,
Thierry


On Wed, Nov 01, 2017 at 02:36:59PM -0700, Simon Brandhorst wrote:
> sage: A = Matrix(ZZ,2,[3,3,3,3])
> sage: A
> [3 3]
> [3 3]
> sage: A//3
> ---
> TypeError Traceback (most recent call last)
> ...
> 
> TypeError: unsupported operand parent(s) for //: 'Full MatrixSpace of 2 by 
> 2 dense matrices over Integer Ring' and 'Full MatrixSpace of 2 by 2 dense 
> matrices over Integer Ring'
> sage: A/3
> [1 1]
> [1 1]
> sage: (A/3).base_ring()
> Rational Field
> 
> 
> What I would like is a matrix over the Integers. Not an integral matrix 
> over the rationals.
> This seems so elementary that I wonder if I am missing something?
> 
> (A/3).change_ring(ZZ) 
> 
> is too ugly. Ticket?
> 
> And similar for p-adic integers, polynomial rings in one variable etc.
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.

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

[sage-devel] Re: floor division for integer matrices

[Simon Brandhorst]

Enabling // just for Euclidean rings is fine for me. 
Or let's say A//k is defined if // is defined on the entries of A? (which 
should include euclidean rings).
And if it is not then one has to do the ugly conversion by hand if one 
really wants it.

On Wednesday, November 1, 2017 at 11:30:59 PM UTC+1, Simon King wrote:
>
> On 2017-11-01, David Roe  wrote: 
> > I don't think you're missing anything, and I would support adding this 
> > feature to matrices. 
>
> I wouldn't support it. 
>
> If you have an integral domain R, then the quotient x/y for x,y in R 
> will live in the fraction field of R, whereas floor division x//y 
> (I think) is only defined if R is a euclidean ring (division x//y 
> with remainder x%y). 
>
> The situation that this thread is about is different. Here, 
> we have a module, not an integral domain nor a euclidean ring. 
> If M is a matrix and c a skalar, then M/c is just syntactical 
> sugar for (1/c)*M. 
>
> Should M//c become syntactical sugar for "conversion of (1/c)*M 
> into M.parent()"? I don't see why it should. 
>
> That said: I also don't say that I would oppose to making M//c work. 
> I just wouldn't support it... 
>
> Best regards, 
> Simon 
>
>

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