Jeroen Demeyer wrote:
On 2014-05-09 14:08, David Loeffler wrote:
It's happened a few times that people have put tests in the sage/tests
directory which explicitly require that certain inputs return
NotImplementedError, or something similar. Surely we shouldn't have to
go through the rigmarole of contacting the original author and waiting
a year in order to add a previously un-implemented feature?
I'd say that, alongside the policy of not changing the output of tests
in sage/tests without a year's deprecation period and consultation
with the original author, we should have a parallel policy that all
tests added to sage/tests should be "forward-compatible", in the sense
that they shouldn't be broken by the addition of reasonably
foreseeable new features to Sage.
Also, sometimes floating-point accuracy of some function improves,
"breaking" a doctest.
Same for deprecation warnings, but see my other reply. The files there
don't have to be "read-only", but changes should be reviewed carefully,
and the authors should get notified when appropriate (which I think is
not the case for floating-point noise).
-leif
--
() The ASCII Ribbon Campaign
/\ Help Cure HTML E-Mail
--
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 email to sage-devel@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
