[sage-devel] Re: Linear programming docs and problem

2010-08-25 Thread Nathann Cohen
Hello !!! Well, I'm sorry to say it like that, but as David mentionned this document is really outdated now, which mean I have been thinking about rewriting all of it. Actually, I already wrote an introduction to graphs and LP using Sage, which should soon be available as a part of the (french) Sa

Re: [sage-devel] Does cliquer really depend on SCons ?

2010-08-25 Thread Dr. David Kirkby
On 08/26/10 01:24 AM, Mitesh Patel wrote: On 08/25/2010 07:08 PM, David Kirkby wrote: spkg/standard/deps shows: $(INST)/$(CLIQUER): $(BASE) $(INST)/$(SCONS) $(INSTALL) "$(SAGE_SPKG) $(CLIQUER) 2>&1" "tee -a $(SAGE_LOGS)/$(CLIQUER).log" but as far as I can tell, there is no such depen

Re: [sage-devel] Does cliquer really depend on SCons ?

2010-08-25 Thread Mitesh Patel
On 08/25/2010 07:08 PM, David Kirkby wrote: > spkg/standard/deps shows: > > $(INST)/$(CLIQUER): $(BASE) $(INST)/$(SCONS) > $(INSTALL) "$(SAGE_SPKG) $(CLIQUER) 2>&1" "tee -a > $(SAGE_LOGS)/$(CLIQUER).log" > > > but as far as I can tell, there is no such dependency. > > Cliquer is a stand

[sage-devel] Does cliquer really depend on SCons ?

2010-08-25 Thread David Kirkby
spkg/standard/deps shows: $(INST)/$(CLIQUER): $(BASE) $(INST)/$(SCONS) $(INSTALL) "$(SAGE_SPKG) $(CLIQUER) 2>&1" "tee -a $(SAGE_LOGS)/$(CLIQUER).log" but as far as I can tell, there is no such dependency. Cliquer is a standard bit of C, with a simple makefile. Does anyone know why cliq

Re: [sage-devel] Linear programming docs and problem

2010-08-25 Thread David Kirkby
On 25 August 2010 17:23, Ryan Hinton wrote: > First, I'm hoping someone (e.g. Nathann Cohen) will notice trac #9801 > which makes a few corrections to the linear programming part of the > Sage Constructions document.  I've fixed one problem, but I'm left > with two more. http://www.sagemath.org/d

Re: [sage-devel] Could someone retest SYMPOW on Cygwin, to close #9166

2010-08-25 Thread Dr. David Kirkby
On 08/25/10 09:49 PM, Mike Hansen wrote: On Wed, Aug 25, 2010 at 1:29 AM, Dr. David Kirkby wrote: It is 99% identical to the SYMPOW of mine that Mike tested before, but there are a couple of minor changes. Assuming that patch above, builds and passes tests on Cygwin, then #9166 can be closed a

[sage-devel] Numerical integration of ODEs update

2010-08-25 Thread Ryan Hinton
I saw in http://groups.google.com/group/sage-devel/browse_thread/thread/bf3a2494d547230a/3f890b6a4cb79fc3?hl=en&ie=UTF-#3f890b6a4cb79fc3 that T. Dumont was planning on wrapping a set of modern numerical methods for ODEs for Sage. Has anyone worked on this? -- To post to this group, send an email

Re: [sage-devel] Could someone retest SYMPOW on Cygwin, to close #9166

2010-08-25 Thread Mike Hansen
On Wed, Aug 25, 2010 at 1:29 AM, Dr. David Kirkby wrote: > It is 99% identical to the SYMPOW of mine that Mike tested before, but there > are a couple of minor changes. Assuming that patch above, builds and passes > tests on Cygwin, then #9166 can be closed as fixed, so one less things to > worry

[sage-devel] Linear programming docs and problem

2010-08-25 Thread Ryan Hinton
First, I'm hoping someone (e.g. Nathann Cohen) will notice trac #9801 which makes a few corrections to the linear programming part of the Sage Constructions document. I've fixed one problem, but I'm left with two more. 1. The maximal matching example code does not like my fix. sage: g = graphs.

Re: [sage-devel] Re: graphviz / dot2tex support

2010-08-25 Thread Dr. David Kirkby
On 08/24/10 07:18 PM, Nicolas M. Thiery wrote: Thanks Dima, that's at least a piece of feedback :-) So, if no one seems to care, does this mean that this is implicitly generally accepted as an optional package? Vote: [ ] Yes! [ ] No! Cheers, Nicolas -- Nico

Re: [sage-devel] Re: graphviz / dot2tex support

2010-08-25 Thread Florent Hivert
> > > > Speaking of which: I never got feedback on my request to put this spkg > > > > in optional/experimental: > > > > > >http://groups.google.com/group/sage-devel/browse_thread/thread/ea1dca... > > > > > your wish to have it as standard most probably is not possible, due to > > > licensing of gr

[sage-devel] Re: two polynomial questions

2010-08-25 Thread kcrisman
On Aug 25, 4:02 am, "Johan S. R. Nielsen" wrote: > > 2. If one has a non-symbolic polynomial currently, it won't plot with > > the new plotting syntax. > > > plot(f,0,5) # works, old-school Sage > > plot(f,(x,0,5)) # doesn't work, new-school Sage > > plot(f,x,0,5) # doesn't work, though sort of

[sage-devel] Could someone retest SYMPOW on Cygwin, to close #9166

2010-08-25 Thread Dr. David Kirkby
http://boxen.math.washington.edu/home/kirkby/patches/sympow-1.018.1.p8.spkg which is merged in the latest alpha fixes the SYMPOW issues on Solaris x86 and OpenSolaris x86. It is 99% identical to the SYMPOW of mine that Mike tested before, but there are a couple of minor changes. Assuming that

Re: [sage-devel] Failed to build Atlas in Sage 4.5.2 on Xenon processor.

2010-08-25 Thread Dr. David Kirkby
On 08/25/10 02:01 AM, William Stein wrote: On Tue, Aug 24, 2010 at 1:38 PM, William Stein wrote: On Mon, Aug 23, 2010 at 8:25 PM, Jonathan Hanke wrote: Hi, I'm trying to build sage 4.5.2 on a Sun Fire X4450 server (Intel Xenon processors) running Redhat linux (Enterprise edition), and get an

[sage-devel] Re: two polynomial questions

2010-08-25 Thread Johan S. R. Nielsen
> 2. If one has a non-symbolic polynomial currently, it won't plot with > the new plotting syntax. > > plot(f,0,5) # works, old-school Sage > plot(f,(x,0,5)) # doesn't work, new-school Sage > plot(f,x,0,5) # doesn't work, though sort of makes sense it shouldn't > since x isn't a symbolic variable n

[sage-devel] Sage 4.5.3.alpha2 - all tests pass on 'fulvia' - Solaris x86

2010-08-25 Thread Dr. David Kirkby
As I remarked yesterday Sage 4.5.3.alpha2.tar was the first ever Sage to build and pass all doc tests on OpenSolaris. Since Skynet became available against yesterday, I built and tested on there, and again all tests pass. sage -t -long devel/sage/sage/symbolic/integration/integral.py

[sage-devel] Re: two polynomial questions

2010-08-25 Thread Johan S. R. Nielsen
> >  > 1. There is no way to get a symbolic interpolated polynomial de novo > >  > without going through polynomial rings, e.g. all these steps: > >  > > >  > pts = [(1,2),(2,3),(3,2),(4,3),(5,2),(6,3)] > >  > R.=QQ[] > >  > f = R.lagrange_polynomial(pts) > >  > SR(f) > >  > Everything you're doin