Back to the main page.

Bug 2840 - diagnose and/or fix testscripts that fail to run properly

Status CLOSED FIXED
Reported 2015-02-12 15:27:00 +0100
Modified 2015-07-15 13:31:10 +0200
Product: FieldTrip
Component: core
Version: unspecified
Hardware: PC
Operating System: Mac OS
Importance: P5 normal
Assigned to: Robert Oostenveld
URL:
Tags:
Depends on:
Blocks:
See also: http://bugzilla.fcdonders.nl/show_bug.cgi?id=2835

Robert Oostenveld - 2015-02-12 15:27:33 +0100

I noticed the following test scripts that fail to run, i.e. they do not produce PASSED or FAILED but are somehow aborted prior to completion. Why is that? test_bug1913 test_ft_datatype_source test_ft_prepare_headmodel test_ft_prepare_sourcemodel test_old_trialdef test_tutorial_beamformer test_tutorial_beamformingextended20131122 test_tutorial_eventrelatedstatistics test_tutorial_fem


Robert Oostenveld - 2015-02-12 15:37:50 +0100

(In reply to Robert Oostenveld from comment #0) 1) test_bug1913 might be because of mex file incompatibility, it reads eeprobe data. The output log contains a matlab crash report. 2) test_ft_datatype_source is because it takes longer than the specified WALLTIME. 3) test_ft_prepare_headmodel is unclear 4) test_old_trialdef has an issue with xunit 5) test_tutorial_beamformer is unclear 6) test_tutorial_beamformingextended20131122 is unclear 7) test_tutorial_eventrelatedstatistics has an issue with xunit 8) test_tutorial_fem is unclear The "unclear" ones just don't seem to start up. I can see that MATLAB starts, that my startup.m gets parsed, and then nothing happens. This seems a torque cluster issue, ant not something we can solve.


Robert Oostenveld - 2015-02-12 15:45:17 +0100

The test_old_trialdef script has a figure in it that requires interaction to continue. The test_tutorial_eventrelatedstatistics has a missing "end". I fixed both and increased the WALLTIME in test_ft_datatype_source mac011> svn commit Sending test/test_ft_datatype_source.m Sending test/test_old_trialdef.m Sending test/test_tutorial_eventrelatedstatistics.m Transmitting file data ... Committed revision 10229. remaining are a few unclear ones, and test_bug1913 which is a matlab/mexfile crash.


Robert Oostenveld - 2015-02-12 15:47:12 +0100

test_bug1913 also crashes on matlab2014b and with matlab2012b on OS X 10.9.5.


Robert Oostenveld - 2015-02-12 15:52:25 +0100

(In reply to Robert Oostenveld from comment #3) note that this indicates a problem with the eeprobe mex files. Not something I want to solve here and now.


Robert Oostenveld - 2015-02-12 15:56:25 +0100

(In reply to Robert Oostenveld from comment #4) that closes this bug


Robert Oostenveld - 2015-07-15 13:31:10 +0200

closed several bugs at once that were recently fixed