How can I use a plot legend to toggle an element's visibility?

Hi,

I am trying to plot multiple items on the same plot, but I would like to implement a quick method to hide or show a single item, e.g. by clicking on it on the plot legend. I have tried with the clickable legend (as described here: https://www.mathworks.com/matlabcentral/fileexchange/21799-clickablelegend-interactive-highlighting-of-data-in-figures), but this functionality seems to be unsupported in more recent MATLAB releases. I have also tried to include a callback function (as in this example: https://blogs.mathworks.com/pick/2016/03/25/interactive-legend-in-r2016a/), again with no results.

Thanks in advance for your help

[dpb]

Have you tried <the example in the doc> of precisely that functionality?

[alessandro marino]

I have, with no successful results

[dpb]

Show us a SMALL working example that illustrates the issue; while I haven't done it, it's hard to imagine the doc example doesn't work...which specific release are you using?

[dpb]

I dunno if can do this interactively here or not...

function hitcallback_ex1(src,evnt)
  if strcmp(evnt.Peer.Visible,'on')
    evnt.Peer.Visible = 'off';
  else 
    evnt.Peer.Visible = 'on';
  end
end

plot(rand(4));
l = legend('Line 1','Line 2','Line 3','Line 4');
l.ItemHitFcn = @hitcallback_ex1;

No, can't do that here, apparently; there's a whole lot about using this interface i've yet to learn/understand what can/cannot do...

Anyways, putting the above function into an m-file and executing the example code works as advertised here on R2020b...I presume it'll work on more recent releases as well...can toggle each and every line in the plot from the legend label associated...

[alessandro marino]

Hi @dpb,

Thanks a lot, I tried to copy your code and it works! it looks like simpe stuff, but it was my first time studying MATLAB functions, so thanks a lot for your help.

I noticed that, when hiding one function, the y-axis adjusts to maximise the window on the remaining functions. I added the ylim command this way:

plot(rand(4));
ylim("manual");
l = legend('Line 1','Line 2','Line 3','Line 4');
l.ItemHitFcn = @hitcallback_ex1;

Not sure if ther is a more elegant way of doing it, but it works so far.

Thanks a lot again

[Davindra Usov]

have you tried switch cases with a tiled layout. Search switch on matlab help

[dpb]

@Davindra Usov -- what's the issue you're raising? I don't follow; I just created a tiled layout and put a legend on each of two axes/tiles and assigned the above callback function to each. Both work as advertised with no need to do anything else...

% from example for tiledlayout...
x = linspace(0,30);
y1 = sin(x/2);
y2 = sin(x/3);
y3 = sin(x/4);

% Plot into first tile twice... (doc does three, but two's plenty to illustrate)
tiledlayout('flow')
nexttile
plot(x,y1)
hLg(1)=legend('Y1');        % add a legend; save the handle...
nexttile
plot(x,y2)
hLg(2)=legend('Y2');        % add a legend; save the handle...
hLg(1).ItemHitFcn = @hitcallback_ex1;
hLg(2).ItemHitFcn = @hitcallback_ex1;

After this, clicking on the legend entry in each toggles visibility of each as expected...

set(hLg,{'ItemHitFcn'},{@hitcallback_ex1})

is a more elegant syntax to set all handles to the one callback function.

Answer 1

[Aditya Srikar]

Hi Alessandro

You can modify legend properties to control the appearance and behaviour of a Legend object. By changing each property, you can modify certain aspects of the legend.

Consider the following example

plot(rand(3))
lgd = legend('a','b','c');
c = lgd.TextColor;
lgd.TextColor = 'red';

In the above example, the text color of legend is set to red.

To toggle visibility of legend on click event, you can use the property ItemHitFcn, which associates a callback function to legend and the callback function executes when you click legend items.

Refer the documentation for more details :

https://in.mathworks.com/help/matlab/ref/matlab.graphics.illustration.legend-properties.html

Hope it helps !

[alessandro marino]

Thank you @Aditya Srikar, this helped, together with this comment by @dpb

Thank you all