[sage-devel] Re: Which docstrings should be doctested

PS: On 2013-05-09, Simon King wrote: > The disadvantage of this approach: It will only test stuff that can be > imported by Python---cdef methods do not appear in dir(O). And the worse disadvantage: This will only test importable modules. It will not test files. -- You received this message be

[sage-devel] Re: Which docstrings should be doctested

Hi! On 2013-05-08, Nicolas M. Thiery wrote: > I just made some tests (see attached file). Clearly Python uses > introspection. In spkg-check of my group cohomology spkg, I use the following basic approach, also relying on introspection: 1. Start with a list of items (initially: The modules of t

[sage-devel] GSL returning nonsense for divergent integrals

Hello, I've been trying to trace the problem that leads numerical_integral to often return nonsense for divergent integrals: sage: numerical_integral(x^7,1,oo) (4.930380657631324e-32, 8.707498697773713e-19) This is #14274 . GSL correctl

[sage-devel] Re: Sage 5.9 / ecl issue

Rob McMahon wrote: Solaris 10 SPARC, with libffi and libiconv installed on the system already for other purposes. Having worked around the iconv issue (special case for solaris in iconv*/src/src/Makefile.in, make now dies compiling ecl with: if [ -f CROSS-COMPILER ]; then \ touc

Re: [sage-devel] Re: Consistent failure of zn_poly in latest Sage

On Wed, May 8, 2013 at 7:33 PM, Francois Bissey wrote: > On 09/05/13 14:27, leif wrote: >> Kwankyu Lee wrote: >>> For example, can you stop everything on the computer and remote log >>> in to it from another machine, so you're hardly running anything on >>> it? If not, quit all possibl

Re: [sage-devel] Re: Consistent failure of zn_poly in latest Sage

On 09/05/13 14:27, leif wrote: > Kwankyu Lee wrote: >> For example, can you stop everything on the computer and remote log >> in to it from another machine, so you're hardly running anything on >> it? If not, quit all possible applications and try building again. >> By the way, how

[sage-devel] Re: Consistent failure of zn_poly in latest Sage

Kwankyu Lee wrote: For example, can you stop everything on the computer and remote log in to it from another machine, so you're hardly running anything on it? If not, quit all possible applications and try building again. By the way, how much RAM does this machine have (this may n

[sage-devel] Re: Consistent failure of zn_poly in latest Sage

Tom Roby wrote: 5.8 compiled fine on my laptop last week, another OS X 10.6.8 system with somewhat newer specs. I attributed it to the presence of Nicolas Thiery & Anne Schilling in the room during most of the compile [...] Oh, that's something I haven't thought of yet. Is it reproducible? A

Re: [sage-devel] Re: Consistent failure of zn_poly in latest Sage

On 09/05/13 12:02, John H Palmieri wrote: > > > On Tuesday, May 7, 2013 9:54:37 PM UTC-7, Kwankyu Lee wrote: > > For example, can you stop everything on the computer and remote > log in to it from another machine, so you're hardly running > anything on it? If not, quit al

Re: [sage-devel] New doctest continuation ....:

On Wed, May 8, 2013 at 2:58 AM, Jeroen Demeyer wrote: > The new doctesting framework allows doctests with continuation to be written > like > > sage: > : > : > > instead of the old > > sage: > ... > ... > > The new syntax is much better because it has correct alignment and it would > allow

Re: [sage-devel] sage/misc directory

Yes, it'd be nice if this were organized better, but it's unclear if that's worth the work involved and it's not that big of a pain point IMHO. On Wed, May 8, 2013 at 4:53 PM, Scott Mancuso wrote: > I was just familiarizing myself with the layout of the sage library, and I > came across the misc

[sage-devel] Re: Consistent failure of zn_poly in latest Sage

On Tuesday, May 7, 2013 9:54:37 PM UTC-7, Kwankyu Lee wrote: > > For example, can you stop everything on the computer and remote log in to >> it from another machine, so you're hardly running anything on it? If not, >> quit all possible applications and try building again. By the way, how much

[sage-devel] sage/misc directory

I was just familiarizing myself with the layout of the sage library, and I came across the misc folder. I noticed that it has about 150 items in it, which seems a bit much to me. Is this an acceptable way to organize Sage, or would it be appropriate to suggest a more refined organization (with t

Re: [sage-devel] Which docstrings should be doctested

Thanks, I added your testcases to #14272. -- 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 emai

Re: [sage-devel] Which docstrings should be doctested

On Thu, May 09, 2013 at 12:00:16AM +0200, Jeroen Demeyer wrote: > I should add that the doctest syntax you are using should work with > the standard Python doctest module. That is: if whatever syntax > you're using works with the Python doctesting framework, but not > with Sage, then that's a bug w

Re: [sage-devel] Which docstrings should be doctested

On 05/08/2013 11:55 PM, Nicolas M. Thiery wrote: Hi Jeroen, David, Thanks for the quick feedback, and for improving doctests reports! Do you agree with the following summary of the situation: - Doctests in strings not directly attached to a class or a function definition, like those

Re: [sage-devel] Which docstrings should be doctested

Hi Jeroen, David, Thanks for the quick feedback, and for improving doctests reports! Do you agree with the following summary of the situation: - Doctests in strings not directly attached to a class or a function definition, like those used with global options or for testing attribute

Re: [sage-devel] Which docstrings should be doctested

On 05/08/2013 08:51 PM, Nicolas M. Thiery wrote: This is not anymore possible. Finally, with the new global options, it's natural to feed documentation to the option parser as strings, and those strings contain examples. We would want them to be tested! (see sage.combinat.partitions). I'm hoping

Re: [sage-devel] Which docstrings should be doctested

> Is there a compelling reason for why the new doctesting insists on > taking indentation into account? And in general in not taking each and > every multiline string in the file as a docstring where to get > doctests from? > The reason is to be able to identify what function is being tested: when

[sage-devel] Re: Which docstrings should be doctested

Hey, Also I'd also like to add that things like class Foo: __doc__ = \ """ EXAMPLES:: sage: 1/0 """ are not tested (which is why anything in a global options documentation is not being tested). Best, Travis -- You received this message because you are subscribed to

[sage-devel] ARM Sage 5.9

Hi, I finally compiled sage 5.9 for ARM ; a bdist is available at the usual place [~jpuydt/bdist/sage-5.9-armv7l-Linux.tar.gz]. The compilation was more eventful than usual (I had to restart it more), but still ended well. I'm hopeful getting a newer m4rie in sage will help. I'll post the t

[sage-devel] Which docstrings should be doctested

Dear doctest framework team, Let me start by thanking the doctest team for the new framework! Even just from a speed point of view, it's great. On Mon, May 06, 2013 at 08:34:43PM -0700, Ben Hutz wrote: > Thanks that helped. Actually, it does appear that four tests are not > bei

Re: [sage-devel] Re: About the maxima spkg

On 09/05/13 05:45, Julien Puydt wrote: > Le 08/05/2013 19:37, Francois Bissey a écrit : >> On 09/05/13 05:33, kcrisman wrote: >>> >>> >>> On Wednesday, May 8, 2013 1:02:52 PM UTC-4, Snark wrote: >>> >>> Hi, >>> >>> does someone know how the SAGE_CLISP_DISABLE_READLINE_HACK hack >>> works?

Re: [sage-devel] Re: About the maxima spkg

Le 08/05/2013 19:37, Francois Bissey a écrit : On 09/05/13 05:33, kcrisman wrote: On Wednesday, May 8, 2013 1:02:52 PM UTC-4, Snark wrote: Hi, does someone know how the SAGE_CLISP_DISABLE_READLINE_HACK hack works? ?? We haven't used Clisp in quite a while. Maybe this is just cr

Re: [sage-devel] Re: About the maxima spkg

On 09/05/13 05:33, kcrisman wrote: > > > On Wednesday, May 8, 2013 1:02:52 PM UTC-4, Snark wrote: > > Hi, > > does someone know how the SAGE_CLISP_DISABLE_READLINE_HACK hack works? > > > ?? We haven't used Clisp in quite a while. Maybe this is just cruft? > I am suspecting he wan

[sage-devel] Re: About the maxima spkg

On Wednesday, May 8, 2013 1:02:52 PM UTC-4, Snark wrote: > > Hi, > > does someone know how the SAGE_CLISP_DISABLE_READLINE_HACK hack works? > > ?? We haven't used Clisp in quite a while. Maybe this is just cruft? -- You received this message because you are subscribed to the Google Groups

[sage-devel] Re: Building Sage 5.9 fails on Mercurial

On Wednesday, May 8, 2013 5:07:41 AM UTC-4, Volker Braun wrote: > > Can you paste a (simple) command line session that exhibits the problem? If you're referring to the problem with IPython vomiting the symbol list with print commands, I just figured out what the problem was: I had tab chara

[sage-devel] gsoc application

Hello all My name is Remus Barbatei, I am a student at University of Kaiserslautern, Germany, Department of Electrical & Computer Engineering where I am currently doing my masters program in embedded systems I am interested in doing the easy server deployment project as part of google sum

[sage-devel] About the maxima spkg

Hi, does someone know how the SAGE_CLISP_DISABLE_READLINE_HACK hack works? Snark on #sagemath -- 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

[sage-devel] Re: New doctest continuation ....:

Hi Jeroen, On 2013-05-08, Jeroen Demeyer wrote: > On 05/08/2013 04:23 PM, Nathann Cohen wrote: >> As the patchbot already reacts when new patches are added to a ticket, >> would it be possible to ask it to "just check" that the new patch is >> correct, and either to have a special color for that

[sage-devel] Re: Consistent failure of zn_poly in latest Sage

As I communicated a couple of weeks ago, I hit this same error on my desktop OS X 10.6.8 system. It's now happened consistently, even when I unload my system, with both Sage5.8 and 5.9rc1. So far, no luck installing sage on box. On the other hand, 5.8 compiled fine on my laptop last week, a

Re: [sage-devel] Re: New doctest continuation ....:

On 05/08/2013 05:05 PM, Nathann Cohen wrote: We will probably have to rebase them anyway if you merge your patch which changes all doctests. I'm not going to change all doctests, that would indeed be too problematic. Didn't you say that it was possible to automatically update all patches uploa

Re: [sage-devel] Re: New doctest continuation ....:

Hello ! > And would you require the use of ":"? Yep. > * If it's optional, people are going to ignore it. Then it shouldn't be. Otherwise you will be the one complaining on tickets, and I prefer to anger a bot than to anger you :-P > * If it's mandatory, people will complain that a

Re: [sage-devel] Re: New doctest continuation ....:

On 05/08/2013 04:23 PM, Nathann Cohen wrote: As the patchbot already reacts when new patches are added to a ticket, would it be possible to ask it to "just check" that the new patch is correct, and either to have a special color for that or even to post a message on the trac ticket saying that th

[sage-devel] Re: New doctest continuation ....:

> Any ideas? > Jeroen. > As the patchbot already reacts when new patches are added to a ticket, would it be possible to ask it to "just check" that the new patch is correct, and either to have a special color for that or even to post a message on the trac ticket saying that the ... should b

[sage-devel] Re: Sage 5.9 / ecl issue

Solaris 10 SPARC, with libffi and libiconv installed on the system already for other purposes. Having worked around the iconv issue (special case for solaris in iconv*/src/src/Makefile.in, make now dies compiling ecl with: if [ -f CROSS-COMPILER ]; then \ touch ecl_min; \

Re: [sage-devel] Re: Sage 5.9 / iconv issue

On 07/05/2013 17:52, Jean-Pierre Flori wrote: It vaguely remembers me of a problem we had on Cygwin: http://trac.sagemath.org/sage_trac/wiki/CygwinPort#TestingSage5.5.rc1onWindows764bits (Although the problem there is that if you want to use libintl, then it will fail because it will not find

[sage-devel] Re: Coefficients of power series

The behavior of .list() for power series also produces bugs in power series comparison. I filed a ticket and patch for this at http://trac.sagemath.org/sage_trac/ticket/9457 but ran into unexpected problems in some of the elliptic curves code, because the p-adics in Sage are derived from the g

Re: [sage-devel] Use fork when several algorithms are available.

In my opinion, this looks like a bad excuse to skip proper tuning (especially for the ticket you refer to)... -- 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

[sage-devel] New doctest continuation ....:

The new doctesting framework allows doctests with continuation to be written like sage: : : instead of the old sage: ... ... The new syntax is much better because it has correct alignment and it would allow expected output to start with ... if we disallow the old syntax. It is also

[sage-devel] Re: Building Sage 5.9 fails on Mercurial

Can you paste a (simple) command line session that exhibits the problem? -- 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.

[sage-devel] Use fork when several algorithms are available.

A recent post reffers to ticket #14551, which proposes to use different algorithms for the same purpose, given the fact that sometimes one of them is much faster, while some other times it is just the opposite. This situation is not uncommon in sage. Think for instance in symbollic integration wit

Re: [sage-devel] Re: [sage-combinat-devel] use Sage!

Yo ! > - Graph theory (mostly done by Nathann?) > - Linear programming (mostly done by Nathann?) The current thematic tutorial is a translation f an early versin of the book chapter. http://www.sagemath.org/doc/thematic_tutorials/linear_programming.html But I don't think I ever wrote something s

Re: [sage-devel] Finite field nth_root too slow / stuck

This is due to a changed algorithm in #7931 which was aimed at large n. In your case just factoring x^3-3 modulo p is doable, while the algorithm Sage is using requires a primitive root modulo p. This is now http://trac.sagemath.org/sage_trac/ticket/14551. Thanks for catching it! David On Tue,