A discussion started on sage-release, which I think is best on sage-devel.
I believe this ticket
http://trac.sagemath.org/sage_trac/ticket/11277
is a very bad idea, as it disables a doctest which is known to fail on one
platform. IMHO, the doctest should be permitted to fail, and not simply removed.
The whole idea of doctests is to indicate bugs - hiding them by disabling the
test is not a great idea in my opinion.
A couple of comments on sage-release of note are:
========= From Jeroen Demeyer ===========
Given that it has a Trac ticket, we are not "hiding the problem". After
disabling the test as in this ticket, the SVD bug is no different from
most other Sage bugs: known in Trac but doesn't cause doctest failures.
If we want to use your logic, every known Sage bug should have a doctest
demonstrating that bug.
===========End of Jeroen Demeyer's comment==========
Following up from comments, Jason Grout wrote
============== Jason's comment =======
Which is I believe how some other projects work. First they write the doctest
that fails, then they write the code to fix the doctest. I think there is value
in that approach (it keeps the bugs more visible to everyone, and it leads to us
doctesting (with a "pass") something that does actually work on most platforms).
You've got a good point that all isn't lost since the bug is listed on trac,
so it is still technically known, at least by those that have seen the ticket
and remember it.
Jason
=============== End of Jason's comment ==========
I think Jason has very good points there.
Another issue I have with Jeroen's comment is that whilst this test is now
disabled on all platforms, it was only failing on one platform. So now, even if
it did fail on another platform we would not know about it, since the test is
disabled.
It would need some re-write of the test infrastructure, but I think it would be
useful if we had
1) Passed
2) Failed
3) Expected failures
4) Unexpected passes
#3 would keep in the public eye known bugs.
#4 would indicate a bug that may have been fixed as the side-effect of some
other change.
Anyway, I really dislike the idea of disabling a test just because it fails on
one platform.
Comments?
Dave
--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
