How can I plot multiple discontinuous segments in different colors without using a for loop?

I'd like to plot my data in segments using NaNs to indicate discontinuity:

x = [1 2 NaN 3 4 NaN 5 6 NaN 7 8];
y = [8 7 NaN 6 5 NaN 4 3 NaN 2 1];
figure();
plot(x, y)

which makes this:

The problem is I am not sure how to make the segments different colors without chopping the data into pieces and using a for loop (not an option with my bigger data). For this example, using the default ColorOrder is fine; I just want each segment to change to the next color in whatever the order is.

Answer 1

[Geoff Hayes]

Tara - does your data set already have the NaN in it or are you constructing the x and y above with NaN inserted to indicate discontinuity? Or can you reformat the data however you want? For example, if I do

 x = [1 3 5 7; 2 4 6 8];
 y = [8 6 4 2; 7 5 3 1];
 figure();
 plot(x, y)

where each column of the x and y matrices belongs to the same data set, then I can get each segment coloured differently.

[Adam Danz]

Tara, if it's possible to put your (x,y) coordinates into a matrix like this, this solution is more efficient than my surface proposal.

[Tara Spencer]

Yes, I am able to take out the NaNs, so I went with this one.

Answer 2

[Adam Danz]

There is no way to assign >1 color to a line without breaking up the line. However, you could plot a surface instead and by making it 2D and thin, you could make it appear as a line. Here's a demo and the figure it produces.

% Your data
x = [1 2 NaN 3 4 NaN 5 6 NaN 7 8];
y = [8 7 NaN 6 5 NaN 4 3 NaN 2 1];

% Assign color based on NaN groups
col = cumsum(isnan(x)); 

% Make sure 3rd dimension is flat
z = zeros(size(x)); 

% Plot surface with thin edge to appear as line
surface([x;x],[y;y],[z;z], [col;col], 'FaceColor', 'none', ...
    'EdgeColor', 'interp', 'LineWidth', 2)

Afterwards, you can select the colormap . For example ()

colormap('hsv')