Re: [sage-devel] Sagemath : wild experiments

Hi, Le 05/02/2015 02:16, Vincent Delecroix a écrit : I did succeed in building the debian packages rubiks, sagenb, singular and lcalc from debian-science. After that the pruner gives elliptic_curves will not come from debian! The debian package is sagemath-database-elliptic-curves. libgap w

[sage-devel] Re: sage -t src/sage/repl/notebook_ipython.py # 9 doctests failed

> > > make testlong failed. Build from source code 6.4.1 > > > sage -t src/sage/repl/notebook_ipython.py # 9 doctests failed > > > It would be useful to know your exact system and especially the content of one or two of the failing doctests. It's unlikely the ssl is related, but who knows?

Re: [sage-devel] You should rebuild using libgmp >= 5

Version 6.4.1 still shows this. I am assuming it should be fine. On Monday, August 11, 2014 at 6:57:25 AM UTC-7, kcrisman wrote: > > >> That is reassuring! It would be nice if we could prevent the warning >> appearing at all, as it clearly worries people. >> >> > +1 I see this all the time to

[sage-devel] make testlong failed

I just ran the test and it shows this below. It is a fresh install from source code. I read somewhere it is because OpenSSL is not installed. On my other machine the tests passed with OpenSSL installed.: sage -t --long src/sage/repl/notebook_ipython.py # 9 doctests failed Any concerns? I

[sage-devel] sage -t src/sage/repl/notebook_ipython.py # 9 doctests failed

Hey there. make testlong failed. Build from source code 6.4.1 sage -t src/sage/repl/notebook_ipython.py # 9 doctests failed I built the same from source code and ran fine at home. At home, I have installed OpenSSL. Is this the reason? I don't need OpenSSL. TT -- You received this

[sage-devel] "Sage REU" -- Fwd: [MathFaculty] reu program

Hi Sage Devs, There is an "REU" -- an NSF-funded summer research program for undergrads -- at University of Washington. I'm a co-PI on the grant, and frequently several of the students will work on projects related to number theory or Sage or SageMathCloud. If you know of an undergrad who might

Re: [sage-devel] Sagemath : wild experiments

I did succeed in building the debian packages rubiks, sagenb, singular and lcalc from debian-science. After that the pruner gives elliptic_curves will not come from debian! libgap will not come from debian! flintqs will not come from debian! polytopes_db will not come from debian! eclib will not c

Re: [sage-devel] Sagemath : wild experiments

Hi, Le 05/02/2015 00:09, Julien Puydt a écrit : Le 04/02/2015 23:57, Vincent Delecroix a écrit : The compilation did want to install the data pkg into /usr/share/sagemath (and I certainly do not want to compile as root). And indeed there is some stuff installed in there (right now: combinatoria

Re: [sage-devel] Sagemath : wild experiments

On 02/05/15 12:04, Julien Puydt wrote: Le 04/02/2015 23:40, François Bissey a écrit : On 02/05/15 11:30, Julien Puydt wrote: Le 04/02/2015 21:33, François Bissey a écrit : Maxima is a funny beast that can be compiled against several lisp at the same time. Just adding "--enable-ecl" to the

Re: [sage-devel] Sagemath : wild experiments

Le 04/02/2015 23:57, Vincent Delecroix a écrit : 2015-02-04 23:22 UTC+01:00, Julien Puydt : Hi, Le 04/02/2015 21:19, Vincent Delecroix a écrit : Hi Julien, hi all, I tried to use your scripts on sage-6.5.rc0 on debian jessie. Compilation failed on polybori (but much of the rest compiled fin

Re: [sage-devel] Sagemath : wild experiments

Le 04/02/2015 23:40, François Bissey a écrit : On 02/05/15 11:30, Julien Puydt wrote: Le 04/02/2015 21:33, François Bissey a écrit : Maxima is a funny beast that can be compiled against several lisp at the same time. Just adding "--enable-ecl" to the debian rule probably means that a maxima

Re: [sage-devel] Sagemath : wild experiments

2015-02-04 23:22 UTC+01:00, Julien Puydt : > Hi, > > Le 04/02/2015 21:19, Vincent Delecroix a écrit : >> Hi Julien, hi all, >> >> I tried to use your scripts on sage-6.5.rc0 on debian jessie. >> Compilation failed on polybori (but much of the rest compiled fine). I >> describe below what steps I ac

Re: [sage-devel] Sagemath : wild experiments

On 02/05/15 11:30, Julien Puydt wrote: Le 04/02/2015 21:33, François Bissey a écrit : Maxima is a funny beast that can be compiled against several lisp at the same time. Just adding "--enable-ecl" to the debian rule probably means that a maxima based on ecl will be built and installed alongsid

Re: [sage-devel] Sagemath : wild experiments

Le 04/02/2015 21:33, François Bissey a écrit : Maxima is a funny beast that can be compiled against several lisp at the same time. Just adding "--enable-ecl" to the debian rule probably means that a maxima based on ecl will be built and installed alongside whatever debian is already installing

Re: [sage-devel] Sagemath : wild experiments

Hi, Le 04/02/2015 21:19, Vincent Delecroix a écrit : Hi Julien, hi all, I tried to use your scripts on sage-6.5.rc0 on debian jessie. Compilation failed on polybori (but much of the rest compiled fine). I describe below what steps I achieved * sqlalchemy is not anymore a pkg, so one dependen

Re: [sage-devel] Sagemath : wild experiments

Maxima is a funny beast that can be compiled against several lisp at the same time. Just adding "--enable-ecl" to the debian rule probably means that a maxima based on ecl will be built and installed alongside whatever debian is already installing by default. Everything that communicates with m

Re: [sage-devel] Sagemath : wild experiments

Sorry I messed up the attachment. Here is the missing one... 2015-02-04 21:19 UTC+01:00, Vincent Delecroix <20100.delecr...@gmail.com>: > Hi Julien, hi all, > > I tried to use your scripts on sage-6.5.rc0 on debian jessie. > Compilation failed on polybori (but much of the rest compiled fine). I >

Re: [sage-devel] Sagemath : wild experiments

Hi Julien, hi all, I tried to use your scripts on sage-6.5.rc0 on debian jessie. Compilation failed on polybori (but much of the rest compiled fine). I describe below what steps I achieved * sqlalchemy is not anymore a pkg, so one dependency less. I just had to remove the corresponding line in t

Re: [sage-devel] Sagemath : wild experiments

Hi, Le 04/02/2015 17:44, John Foster a écrit : This appears to be a fairly good start toward a system of installation that I have been interested in for several years. using the packages already available in Debian or other linux would go a LONG way towards mainstreaming Sage. I have the proper

[sage-devel] Re: Can a Parent have different Element classes?

Le mercredi 4 février 2015 15:18:26 UTC+1, Volker Braun a écrit : > > Of course nobody can stop you from setting the element's parent manually. > Though thats clearly not recommended. > > Existence of coercions, say, depend on the parent only so the two elements > better have the same interface

Re: [sage-devel] Sagemath : wild experiments

This appears to be a fairly good start toward a system of installation that I have been interested in for several years. using the packages already available in Debian or other linux would go a LONG way towards mainstreaming Sage. I have the proper environment to experiment in & am willing to d

[sage-devel] Re: Can a Parent have different Element classes?

Hi Marc, Le mercredi 4 février 2015 16:55:39 UTC+1, Marc Mezzarobba a écrit : > > > More examples: > > sage: parent(-infinity) is parent(infinity) > True > sage: type(infinity) > > sage: type(-infinity) > > > sage: two = RLF(2) > sage: type(two) > > sage: type(exp(two)) > > sage:

[sage-devel] Re: Can a Parent have different Element classes?

Thierry wrote: >> parent(A) is parent(B) >> but >> type(A) is not type(B) > > Here is an example: > > sage: A = cos(x) > sage: B = units.charge.coulomb > sage: parent(A) is parent(B) > True > sage: type(A) is not type(B) > True More examples: sage: parent(-infinity) is parent(infinity) True sag

[sage-devel] book on automating math

Look up, on Amazon, SIGMA elvey The author, John Elvey, died recently. He did, however, interact with research groups at Waterloo and at Berkeley, some years ago. I know that the Berkeley group was unable to do much with this, and I suspect a similar result at U Waterloo. Maybe Sage-ists woul

Re: [sage-devel] FiniteEnueratedSets: iterator called twice when creating a list

> Is there a trac-ticket for the behaviour of the code [1] for > FiniteEnumeratedSets? The __iter__ function is called twice when > creating a list (or a tuple). This shouldn't be; there is only a warning > given in the docstring of .list(). I did not see any related ticket appear recently. If you

[sage-devel] FiniteEnueratedSets: iterator called twice when creating a list

Is there a trac-ticket for the behaviour of the code [1] for FiniteEnumeratedSets? The __iter__ function is called twice when creating a list (or a tuple). This shouldn't be; there is only a warning given in the docstring of .list(). Daniel [1] Copied from a warning in sage/categories/finite_enum

[sage-devel] Sagemath : wild experiments

Hi, here is a more precise explanation of my experiments. First, let's see how I do a testing run : !!! BEWARE the following commands clean part of your home dir then install cruft in your home dir!!! (1) rm -rf sage-exp .local/lib/python2.7/site-packages/sage* (cleaning the previous attempt)

[sage-devel] Re: "cd man" crashes building ncurses

I wouldn't be surprised if you can break pretty much anything with a suitable CDPATH... On Wednesday, February 4, 2015 at 6:31:30 AM UTC-5, Denis Sunko wrote: > > > I have recently built 6.4.1 from sources on three different computers. One > of the compilations crashed while building the ncurs

[sage-devel] Re: Can a Parent have different Element classes?

Of course nobody can stop you from setting the element's parent manually. Though thats clearly not recommended. Existence of coercions, say, depend on the parent only so the two elements better have the same interface. As for the "identity map" example, if that were a subclass then you would a

Re: [sage-devel] Can a Parent have different Element classes?

Le mercredi 4 février 2015 15:03:20 UTC+1, Thierry (sage-googlesucks@xxx) a écrit : > > > Here is an example: > > sage: A = cos(x) > sage: B = units.charge.coulomb > sage: parent(A) is parent(B) > True > sage: type(A) is not type(B) > True > > (though UnitExpression inherits from Expressi

[sage-devel] Re: Can a Parent have different Element classes?

Another example, still in the homset context, is isomorphisms. They typically require a specific method, __invert__(), which has no meaning for generic homset elements. So, what is the general policy regarding the treatment of isomorphisms in Sage ? For automorphisms, things are different, beca

Re: [sage-devel] Can a Parent have different Element classes?

Hi, On Wed, Feb 04, 2015 at 11:57:48AM +0100, Jeroen Demeyer wrote: > Is it possible in Sage to have a Parent which has different Element > classes? I.e. two Elements A and B such that > > parent(A) is parent(B) > but > type(A) is not type(B) > Here is an example: sage: A = cos(x) sage: B = un

Re: [sage-devel] Re: Can a Parent have different Element classes?

(just to say that it would be cool to add this information to the doc if you figure out the answer to this question) Nathann On 4 February 2015 at 14:40, Eric Gourgoulhon wrote: > Hi, > > I don't have the answer, but a related discussion is in this thread. > > I am often facing this problem, typ

[sage-devel] Re: Can a Parent have different Element classes?

Hi, I don't have the answer, but a related discussion is in this thread . I am often facing this problem, typically when the parent is a homset and one would like to implement a specific Element class to deal with the identity ma

[sage-devel] "cd man" crashes building ncurses

I have recently built 6.4.1 from sources on three different computers. One of the compilations crashed while building the ncurses package. I was able to trace this to a "cd man" command issued during the build process. What happened was that my cdpath on that machine pointed to a directory whic

[sage-devel] Can a Parent have different Element classes?

Is it possible in Sage to have a Parent which has different Element classes? I.e. two Elements A and B such that parent(A) is parent(B) but type(A) is not type(B) -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and

Re: [sage-devel] Error in gap-4.7.5 when installing sage 6.4.1

Well I’ll do some look up because it is strange that MPILIBS has been caught by stuff. François > On 4/02/2015, at 22:10, rafaellopezfernan...@gmail.com wrote: > > Problem solved: the problem was in my MPI installation. For any reason that I > cannot remember > I included -pcmpio in the MPILIBS

Re: [sage-devel] Error in gap-4.7.5 when installing sage 6.4.1

*Problem solved*: the problem was in my MPI installation. For any reason that I cannot remember I included -pcmpio in the MPILIBS environment variable. I have removed it, and now make is doing its work without problems. > -- You received this message because you are subscribed to the Google G

Re: [sage-devel] Error in gap-4.7.5 when installing sage 6.4.1

What about starting from /scratch? In any case this shouldn’t be there: -L/scratch/ADF/adfhome/bin/platform_mpi/MPICH2.0/lib/linux_amd64 -lpcmpio -Wl,-rpath-link,/scratch/ADF/adfhome/bin/platform_mpi/MPICH2.0/lib/linux_amd64 -lmpich -ldl Could we have the full build log of gap (found under /scr

Re: [sage-devel] Error in gap-4.7.5 when installing sage 6.4.1

I must precise that I seeked for "*pcmpio*" in my /usr directory and all the subdirs in the tree starting on it (find /usr -iname "*pcmpio*") -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving email

Re: [sage-devel] Error in gap-4.7.5 when installing sage 6.4.1

I have checked the flags that you suggest and they are all empty. At least, when I "echo" them (echo $LDFLAGS, etc) in the Konsole in which I am running make, empty space is returned. I started sage installation from scratch (running first configure and then make). I have also seeked anythin