Bug in saving .fig (but not .png or .pdf) from when specifiying YAxis.Exponent as -6

In my livescript (MATLAB R2023b), I programatically save some of my plots via saveas(). I'm currently saving as a .pdf, a .png, and a .fig.

When I first specify that that the plot should have a YAxis exponent of -6 (it defaults to -5), my plots look as expected in the produced/displayed MATLAB .fig, and in the saved .pdf and .png. BUT...when I open the saved .fig, it shows an exponent of 10^2147483647. Not 10^-6 as everything else does.

I've been able to reproduce this in a .m file as well, so it doesn't seem to be related to the .mlx.

This is how I'm saving the plots:

set(gcf, 'visible', 'on'); % have to do this bc livescripts sets off, but then the figs open as invisible later
saveas(gcf,plotFullFilename,'fig')
saveas(gcf,plotFullFilename,'png')
exportgraphics(gcf,allPlotsFilename,'append',true)

This is how I set the exponent:

axA.YAxis.Exponent = -6;
axB.YAxis.Exponent = -6;

If I comment out the exponent code and let it default to a yaxis exponent to -5, then the .fig saves with an exponent of -5.

But I don't want a y-axis exponent of -5. I want it in engineering notation, e.g. an exponent of -6. This looks like a bug to me. Any ideas?

[Voss]

How are you opening the saved .fig files? When I open them using openfig, none of the y-axis exponents are 10^2147483647. (Same result if I open them through the OS, e.g., right-click in Windows Explorer and select Open.)

fn = dir('*.fig');
for ii = 1:numel(fn)
    f = openfig(fn(ii).name,'visible');
end

[Nicole009]

I was just double clicking the .fig in the Windows Explorer window. Interesting if your method is a temporary workaround. Looks like it is a documented bug per the accepted answer.

Answer 1

[Steven Lord]

This sounds like Bug Report 3147076 which is listed as fixed in release R2023b Update 6.

[James Cooper]

Thanks for referencing it - I had the same issue and was pulling my hair out.

[Justus Gimsa]

@Steven Lord I still have this with the function (exportgraphics(gcf,....) in 2024a, that the exponent is gone - are people working on this? ;)

[Steven Lord]

This particular bug is listed as fixed in both release R2023b Update 6 and release R2024a. Please contact Technical Support with a specific example and work with them to determine if there's another bug that manifests the same way that this bug fix introduced or exposed.