Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread Volker Braun
This is some more fallout from #16858. Jeroen, do you already have a followup ticket for numerical noise? On Saturday, September 27, 2014 9:51:46 PM UTC+1, Justin C. Walker wrote: On Sep 27, 2014, at 07:51 , Volker Braun wrote: As usual, get the updated develop git branch.

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread William A Stein
On Ubuntu 14.04 LTS Opteron, (a SageMathCloud node), I get numerical noise failures, as alluded to above: sage -t --long --warn-long 57.7 src/sage/rings/polynomial/polynomial_element.pyx # 2 doctests failed sage -t --long --warn-long 57.7 src/sage/rings/real_double.pyx # 2 doctests failed sage

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread John Cremona
On 28 September 2014 17:47, John Cremona john.crem...@gmail.com wrote: I am getting a doctest failure in src/doc/en/bordeaux_2008/elliptic_curves.rst and I wonder if others can confirm or deny. You should have sage: C = CremonaDatabase() sage: C[37] {'allcurves': {'a1': [[0, 0, 1, -1, 0],

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread Jeroen Demeyer
On 2014-09-28 14:46, Volker Braun wrote: This is some more fallout from #16858. Jeroen, do you already have a followup ticket for numerical noise? http://trac.sagemath.org/ticket/17063 -- You received this message because you are subscribed to the Google Groups sage-release group. To

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread Volker Braun
On Sunday, September 28, 2014 5:54:23 PM UTC+1, Stein William wrote: Also, all those by default warning slow doctests things in the log (on doing make ptestlong) are annoying... Good, I hope somebody will clean up some of the unnecessarily long doctests -- You received this message

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread Nathann Cohen
Good, I hope somebody will clean up some of the unnecessarily long doctests This is not the only occasion for which we need something between reporting an error or saying nothing. What about a bit red warning when running the tests on a file, that would show the list of very long doctests ?

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread William A Stein
On Sun, Sep 28, 2014 at 11:49 AM, Volker Braun vbraun.n...@gmail.com wrote: On Sunday, September 28, 2014 5:54:23 PM UTC+1, Stein William wrote: Also, all those by default warning slow doctests things in the log (on doing make ptestlong) are annoying... Good, I hope somebody will clean up

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread Volker Braun
The threshold is a certain fraction of the total doctest time, so it takes the relative speed of computers into account. Really, there is no point in having tests run for minutes just because you can. Also, if you want to debug things then disabling optimization and can quickly lead to

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread François Bissey
On Sun, 28 Sep 2014 09:53:42 William A Stein wrote: On Ubuntu 14.04 LTS Opteron, (a SageMathCloud node), I get numerical noise failures, as alluded to above: sage -t --long --warn-long 57.7 src/sage/rings/polynomial/polynomial_element.pyx # 2 doctests failed sage -t --long --warn-long 57.7

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread William A Stein
On Sun, Sep 28, 2014 at 1:03 PM, Volker Braun vbraun.n...@gmail.com wrote: The threshold is a certain fraction of the total doctest time, so it takes the relative speed of computers into account. When I did make ptestlong (with MAKE=make -j12 and 12 cores), it says sage -t --long

Re: [sage-release] Sage 6.4.beta4 released

2014-09-28 Thread François Bissey
Anyone else see doctest failure of the kind: sage -t --long /usr/share/sage/src/sage/symbolic/expression.pyx ** File /usr/share/sage/src/sage/symbolic/expression.pyx, line 9349, in sage.symbolic.expression.Expression.solve Failed