Timing instability in doctest reporter

[vbraun]

Random doctest failure on the buildbot:

sage -t --long src/sage/doctest/reporting.py
**********************************************************************
File "src/sage/doctest/reporting.py", line 265, in sage.doctest.reporting.DocTestReporter.report
Failed example:
    DTR.report(FDS, False, 0, (sum([len(t.examples) for t in doctests]), D), "Good tests")
Expected:
        [... tests, 0.00 s]
Got:
        [103 tests, 0.01 s]
**********************************************************************
File "src/sage/doctest/reporting.py", line 275, in sage.doctest.reporting.DocTestReporter.report
Failed example:
    DTR.report(FDS, False, 0, (sum([len(t.examples) for t in doctests]), D), "Doctest output including the failure...")
Expected:
        [... tests, 1 failure, 0.00 s]
Got:
        [103 tests, 1 failure, 0.01 s]
**********************************************************************
File "src/sage/doctest/reporting.py", line 289, in sage.doctest.reporting.DocTestReporter.report
Failed example:
    DTR.report(FDS, False, 0, (sum([len(t.examples) for t in doctests]), D), "Good tests")
Expected:
        1 unlabeled test not run
        4 long tests not run
        5 magma tests not run
        2 other tests skipped
        [... tests, 0.00 s]
Got:
        1 unlabeled test not run
        4 long tests not run
        5 magma tests not run
        2 other tests skipped
        [103 tests, 0.01 s]
**********************************************************************

obviously we shouldn't doctest that the time is really 0.00s

Component: doctest framework

Keywords: days57

Author: Volker Braun

Branch/Commit: 2d73de6

Reviewer: Jeroen Demeyer

Issue created by migration from https://trac.sagemath.org/ticket/16109

[vbraun]

Author: Volker Braun

[vbraun]

Branch: u/vbraun/timing_instability_in_doctest_reporter

[vbraun]

Changed keywords from none to days57

[vbraun]

Commit: 2d73de6

[vbraun]

New commits:

2d73de6

do not test that elapsed time is 0.00s

[jdemeyer]

comment:4

Could you please tell me which buildbot this is? I am asking because I have never seen this error before and I want to make sure that doctesting didn't actually become slower.

[jdemeyer]

Reviewer: Jeroen Demeyer

[jdemeyer]

comment:5

Obviously, the patch makes sense, but it's strange that this error suddenly appears.

[vbraun]

comment:6

I've seen it once so far. I don't remember which buildbot but one of the UW machines. I think there were a lot of background jobs running.

[vbraun]

Changed branch from u/vbraun/timing_instability_in_doctest_reporter to 2d73de6