Closed leshaker closed 7 years ago
Yeah, I agree. It would be great if whoever decides to pick up these bugs also writes a test for this functionality (since they'll be investigating how X functionality works anyway and likely have a good understanding of how to test it) and adds them to doTests in /Examples/. That way, future breakage can be reduced.
Also, doTests should be run (preferably on all three platforms) before pushing anything potentially disruptive (e.g. changes to c code, arLink, arCompile, arLoadX, arSaveX) to the master branch. It's happened more than once that things were broken which were already covered by a test. It's much easier to repair these mistakes when the new code is still being written / edited than to attempt to figure out what is going on after the fact.
this part in the function stringListChooser
if(isfield(S,'ar.ple'))
if(isfield(S.ar.ple,'chi2s'))
nple = sum(~cellfun(@isempty,S.ar.ple.chi2s));
plestr = [' #PLE=',sprintf('%3i ',nple)];
end
end
is never used because S.ar
is loaded from workspace_pars_only.mat
and contains only parameters but not the ple...
So far I found:
arLoadPars
with older workspaces:Error in stringListChooser>readParameterAnnotation (line 112) if(S.ar.config.fiterrors == 1)
Error in stringListChooser (line 13) [anno,vals{j}] = readParameterAnnotation(liste{j}); % vals contains npara, nfitted, chi2, ... and is useful when working with a breakpoint
Error in fileChooser (line 24) out = stringListChooser(file_list, default, zeigen);
Error in arLoadPars (line 49) [~, filename] = fileChooser(pfad, 1, true);
Warning: pleSave is deprecated. pleGlobals is now stored as ar.ple by using arSave
to be continued...