Times extracted from TimeSeries are wrong (and erratic)

[Marie Rognes]

Something seems to be messed up in the TimeSeries mesh_times() and vector_times(). Adding

print series.mesh_times()
print series.vector_times()

at the end of demo/undocumented/time-series/python/demo_time-series.py gives random values, most of which are of the order 1e-312, that change with each run.

[Marie Rognes (meg-simula)]

MER: I'm upping the importance on this one, because I needed it yesterday ;-)

[Johan Hake (johan-hake)]

Well, then it should be fixed.

Could you please add a unit test for these two functions? Maybe compare with the output of TimeSeries.str(True)?

[Marie Rognes (meg-simula)]

On 06/15/2011 07:41 PM, Johan Hake wrote:
> Well, then it should be fixed.
>
> Could you please add a unit test for these two functions? Maybe compare
> with the output of TimeSeries.str(True)?
>

Thanks!

Unit test under way -- once buildbots go green.

--
Marie

> ** Changed in: dolfin
> Status: Confirmed => Fix Committed
>