sage: stats = json.load(open('/home/vbraun/.sage/timings2.json'))
sage: sorted([(v['walltime'], k) for k,v in stats.items() if 'failed' not
in v])[-20:]
[(103.71118903160095,
u'sage.combinat.rigged_configurations.tensor_product_kr_tableaux'),
(105.40626382827759, u'sage.combinat.root_system.plot'),
(105.58554887771606, u'sage.functions.bessel'),
(106.15052008628845, u'sage.plot.graphics'),
(108.91321992874146, u'sage.schemes.elliptic_curves.isogeny_small_degree'),
(113.79727506637573, u'sage.matroids.catalog'),
(117.22659015655518, u'sage.symbolic.random_tests'),
(118.7537989616394, u'sage.combinat.ncsf_qsym.ncsf'),
(125.5258629322052, u'sage.combinat.backtrack'),
(126.43836998939514,
u'sage.schemes.hyperelliptic_curves.hyperelliptic_finite_field'),
(127.17231702804565, u'sage.plot.plot'),
(128.25516295433044, u'sage.calculus.riemann'),
(135.7807331085205,
u'sage.combinat.rigged_configurations.rigged_configurations'),
(149.45724391937256, u'sage.tests.book_schilling_zabrocki_kschur_primer'),
(158.25851893424988, u'sage.combinat.sf.macdonald'),
(171.6492350101471, u'sage.combinat.crystals.alcove_path'),
(203.31193494796753, u'sage.schemes.elliptic_curves.ell_rational_field'),
(214.14894604682922,
u'sage.combinat.cluster_algebra_quiver.mutation_type'),
(271.6959960460663, u'sage.combinat.similarity_class_type'),
(375.5583908557892,
u'sage.combinat.root_system.non_symmetric_macdonald_polynomials')]
We don't need to change anything in these right away, hopefully the warning
will nag the authors into shortening their test ;-) Maybe with the
exception of sage.plot.graphics, none of these are core functionality
making it very likely that they just demonstrate what can be done instead
of being necessary for coverage.
On Tuesday, July 29, 2014 4:20:09 AM UTC-4, John Cremona wrote:
>
> I would not be surprised if some of the offending long tests are by me ("
> just because the author wants to show off the capabilities of their code")
> so can you post a list of the longest ones?
>
> The other factor is that some modules are themselves very long (thousands
> of lines) withough -- necessarily -- having long individual tests. Though
> probably they do.
>
> John
>
>
> On 29 July 2014 03:58, Volker Braun <[email protected] <javascript:>>
> wrote:
>
>> A small handful of modules takes the majority of the doctest walltime,
>> see e.g. this lopsided distribution:
>>
>> <http://boxen.math.washington.edu/home/vbraun/t/doctest.png>
>> I'm proposing to
>>
>> * turn the existing sage -t --warn-long parameter into a warning (instead
>> of an error)
>>
>> * enable it by default with a limit of 60s on a modern computer, adjusted
>> for slowness of the computer running the tests.
>>
>> This is implemented in http://trac.sagemath.org/ticket/16609. Please
>> review.
>>
>> --
>> 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 [email protected] <javascript:>.
>> To post to this group, send email to [email protected]
>> <javascript:>.
>> Visit this group at http://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 [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
