Environment
In MATLAB® Online™, you can change the colors of the MATLAB desktop using themes.
For example, to select a dark theme, on the Home tab, in the
Environment section, click 
Preferences. Select MATLAB > Appearance and set the Theme to Dark.
To further customize the colors of the MATLAB desktop, select MATLAB > Appearance > Colors. Then, change the colors in the Desktop tool colors, MATLAB syntax highlighting colors, and MATLAB output colors sections.
As part of this change, icons in the MATLAB Online desktop have an improved visual appearance.
For more information about changing the colors of the MATLAB desktop, see Change Desktop Colors.
You can change the text and background colors in the Live Editor by changing the MATLAB desktop tool colors.
To change the text and background colors:
On the Home tab, in the Environment section, click
Preferences.Select MATLAB > Colors
In MATLAB Online, select MATLAB > Appearance > Colors.
In the Desktop tool colors section, clear the Use system colors check box.
In MATLAB Online, the Use system colors check box is not available and this step can be skipped.
Use the Text and Background fields to change the colors. For example, select white for the text color and black for the background color.
The Live Editor automatically selects colors for titles and headings based on the selected background color. To further customize the colors of titles and headings, use settings. For more information, see matlab.fonts Settings.
You can use a subset of regular expressions to search for text that matches a pattern
in an open file in the Editor or Live Editor. To search using a regular expression, on
the Editor or Live Editor tab, in the
Navigate section, click 
Find. Then, in the find and replace dialog box, enter
a regular expression, and select the Regular Expression button
.
For example, to find all the words in a file that contain the letter
x, enter the expression \w*x\w* and select the
Regular Expression button .
For more information, see Find and Replace Text in Files and Go to Location.
Language and Programming
Data Analysis
Data Import and Export
Mathematics
Graphics
Create plots by passing a table directly to any of these functions: plot,
plot3, loglog, semilogx, semilogy, and polarplot. When you specify your data as
a table, Cartesian axis labels and the legend (if present) are automatically labeled
using the table variable names.
For example, create a table with the variables Temperature and
Humidity. Pass the table to the plot
function as the first argument, and specify the variables you want to plot.
Temperature = (40:70)'; Humidity = (50:80)' + randn(31,1); T = table(Temperature,Humidity); plot(T,"Humidity","Temperature")
When plotting tabular data, the default row labels of data tips created interactively
or with the datatip
function are the names of the table variables associated with the data point.
For example, create a table using the sample file patients.xls.
Then, plot the Systolic, Diastolic, and
Weight variables in a bubble chart. A data tip created with
datatip(b) displays three rows. The row labels are "Systolic",
"Diastolic", and "Weight".
tbl = readtable("patients.xls"); b = bubblechart(tbl,"Systolic","Diastolic","Weight"); datatip(b);
For scatter plots and bubble charts, data tips created interactively or with the
datatip
function include by default rows for visual properties such as size, color, or
transparency that are specified with vector data.
For example, create a scatter plot of random data and define the marker sizes as
vector sz. A data tip created with datatip(s)
displays three rows: X, Y, and
Size. The Size row in the data tip displays
the marker size specified by sz for the associated data point.
x = linspace(0,3*pi,200); y = cos(x) + rand(1,200); sz = linspace(1,100,200); s = scatter(x,y,sz); datatip(s);
The bubblechart, bubblechart3, polarbubblechart, and scatter3 functions now accept the same combinations of matrices and
vectors as the plot function does. As a result, you can visualize multiple data sets at
once rather than using the hold function between plotting commands.
Set the Alignment property of a Quiver object
to control how the arrows align with the data points. Valid values are
"head", "center", and
"tail". For example, plot a grid of vectors with the arrow heads
positioned at the data points. Specify a marker symbol to show the locations of the data points.
[X,Y] = meshgrid(0:6,0:6); U = 0.25*X; V = 0.5*Y; quiver(X,Y,U,V,"Alignment","head","Marker","o")
When you modify a figure in MATLAB Online using the Figure tab, MATLAB generates code that you can view, copy, and export. To view the generated code, select Show Code in the File section of the Figure tab. MATLAB generates code for these actions:
Adding a title, axis label, legend, color bar, grid, or annotation
Changing the text or line style
Using the pan, zoom, rotate, or data tip interactions
MATLAB does not currently generate code for the Select and Edit option in the Figure tab.
For example, you can create a surface plot with surf(peaks). Then,
interactively add a title and colorbar, zoom into the axes, and view generated code.
Polar axes now display tick values in degrees with degree symbols when the
ThetaAxisUnits property is set to
"degrees". For example, create a polar plot. By default, the
theta-axis displays the tick values with degree
symbols.
theta = 0:0.01:2*pi; rho = sin(2*theta).*cos(2*theta); polarplot(theta,rho)
This change clarifies which units are being used for the theta
tick values. You can use the ThetaAxisUnits property to display
the tick values in degrees or radians. To remove the degree symbols, change the tick
label format for the theta-axis:
pax = gca;
pax.ThetaAxis.TickLabelFormat = "%g";App Building
You can now create styles for table and tree UI components that specify an icon and a
text interpreter using the uistyle
function.
Specify the
Iconproperty of the style object to add icons to table cells and tree nodes.Specify the
IconAlignmentproperty of the style object to modify where the icon appears in relation to the text in table cells.Specify the
Interpreterproperty of the style object to format text or add links using HTML markup, or to add equations using TeX or LaTeX markup to table cells and tree nodes.Specify the
HorizontalClippingproperty of the style object to control whether long text is clipped on the left or the right in table cells and tree nodes.
Add a style to a UI component using the addStyle
function.
For example, this code creates two styles, one that specifies an icon and one that specifies the text interpreter as TeX, and applies the styles to columns of a table.
T = table(["x^2";"3x^3+1"],["Success";"Success"]); fig = uifigure(Position=[500 500 300 160]); t = uitable(fig,Position=[10 10 250 140],Data=T); s1 = uistyle(Interpreter="tex"); s2 = uistyle(Icon="success",IconAlignment="right"); addStyle(t,s1,column=1) addStyle(t,s2,column=2)
Grid layout managers with row heights or column widths of 'fit' now
resize to fit the contents of table, list box, and image UI components.
For example, when you create a table UI component inside a grid layout manager with a
row height or column width of 'fit', the height of the row or the
width of the column resizes to fit the data in the
table.
fig = uifigure(Position=[680 558 300 170]);
gl = uigridlayout(fig);
gl.RowHeight = {'fit'};
gl.ColumnWidth = {'fit'};
tbl = uitable(gl,Data=rand(3));
In R2021b, grid layout managers with row heights or column widths of
'fit' scaled to a fixed size when the row or column contained
a table, list box, or image UI component.
Table UI component — Row height and column width previously resized to 300 pixels.
List box UI component — Row height previously resized to display at most four items. The exact pixel value to display four items might vary depending on your settings.
Image UI component — Row height and column width previously resized to 100 pixels.
To display a table, list box, or image at its size in a release before R2022a, set
the corresponding elements of the RowHeight and
ColumnWidth properties of the
GridLayout object to their respective fixed sizes.
To specify error handling options when debugging code in App Designer, configure the
Run button by clicking Run
. You can choose to pause code execution when an error
occurs, when a warning occurs, or when a NaN or
Inf value is returned.
Additionally, error messages in App Designer now contain links to relevant files and functions. Use these links to navigate more easily to the documentation or to line numbers in your code when debugging your apps.
Compare and merge two versions of an app file in MATLAB
Online using the Comparison Tool. To open the Comparison Tool, click
Compare in the Designer tab of the App
Designer Toolstrip.
For more information, see Compare and Merge Apps.
When creating an app with a UIAxes component in App Designer,
the UIAxes component content is not displayed in Design
View when the component is partially or fully off the canvas.
Instead, the UIAxes component is shown as a placeholder image. To
see the content of the component, drag it fully onto the canvas.
You can still modify properties of the UIAxes component when it
is off the canvas, but you will not be able to see a visual reflection of those
changes in Design View until the component is dragged onto the
canvas.
Performance
Apps that involve continuous event-driven updates, such as animations implemented with
a timer, are more responsive to those events. To observe the improvement, the animation
must be created in an app or in a figure created with the uifigure function. For example, an app containing an animation
controlled by a timer object is more responsive when you call the timer’s
start and stop methods.
This code creates an app using a timer object to plot a random number every 0.01 second. If you run this app on a Windows 10, Intel Xeon CPU W-2133 @ 3.60 GHz test system, the animation responds more quickly when you click the Start/Stop button than in the previous release.
function mytimerapp uif = uifigure("CloseRequestFcn",@CloseRequest); uibutton(uif,"State","Position",[235 59 100 22], ... "ValueChangedFcn",@StateButtonChanged,"Text","Start/Stop"); ax = uiaxes(uif,"Position",[25 100 508 300],"XDir","reverse"); % Create initial plot line p = plot(ax,0:60,zeros(1,61)); % Create timer object RandTimer = timer(... "ExecutionMode","fixedRate", ... "Period",0.01, ... "BusyMode","queue", ... "TimerFcn",@RandTimerFcn); % Timer function function RandTimerFcn(~,~,~) % Generate a random number and update plot line ydata = p.YData; ydata = circshift(ydata,1); ydata(1) = rand; p.YData = ydata; end % State button callback function function StateButtonChanged(obj,~) switch obj.Value case 0 stop(RandTimer); case 1 % If timer is not running, start it if strcmp(RandTimer.Running,"off") start(RandTimer); end end end % Figure close request function function CloseRequest(~,~) % Stop timer, then delete timer and figure stop(RandTimer); delete(RandTimer); delete(uif); end end
Responsiveness is improved for panning, rotating, and zooming into a region of
interest within a plot. To observe the improvement, the plot must be created in an app
or in a figure created with the uifigure function. The improvement is most noticeable for plots with
large numbers of points, or those that involve complex effects such as lighting,
transparency, or texture maps. Systems equipped with modern GPUs are more likely to show
the improvement.
For example, create a 3-D scatter plot with 200,000 points. If you run this code on a Windows 10, Intel Xeon CPU W-2133 @ 3.60 GHz test system with an NVIDIA Quadro® P620 GPU, and then drag to rotate the plot, the rotation is smoother and responds more quickly to the drag gesture than in the previous release.
z = linspace(0,4*pi,200000);
x = 2*cos(z) + rand(1,200000);
y = 2*sin(z) + rand(1,200000);
f = uifigure; a = axes(f);
scatter3(a,x,y,z,"filled")
Responsiveness is improved for panning, rotating, and zooming into a region of
interest within a UIAxes or Axes object in MATLAB
Online that contains a chart created with yyaxis and a legend. For such charts, updates are faster, and
interactions are smoother in R2022a than in the previous release.
For example, create a UIAxes object with two
y-axes and a legend. If you run this code on a Windows 10, Intel
Xeon CPU E5-1650 v4 @ 3.60 GHz test system, when you drag the cursor to pan the
view of a UIAxes object, the axes pan faster and track the cursor more
closely in R2022a than in R2021b.
f = uifigure;
a = uiaxes(f);
plot(a,1:100,rand(1,100));
yyaxis(a,"right");
plot(a,1:100,linspace(1,10,100));
legend(a);
Animations show improved performance in apps when multiple figures are open. To
observe the improvement, all figures must be created with the uifigure function.
For example, this code opens 10 empty figures and an additional figure containing a plot. The plot displays one marker that traces the path of a line at every iteration of a loop. The loop is about 6x faster than in the previous release.
function movingmarker % Open 10 empty figures for n = 1:10 uifigure("Position",[10+n*3 10+n*3 200 200]); end % Create a figure and a plot with one marker uif = uifigure; x = linspace(0,10,200); y = sin(x); ax = axes(uif); p = plot(ax,x,y,"-o","MarkerIndices",1); % Move the marker along the sine wave tic for idx = 2:200 p.MarkerIndices = idx; drawnow end toc end
The approximate execution times for the loop are:
R2021b: 32.6 s
R2022a: 5.4 s
The code was timed on a Windows 10, Intel
Xeon CPU W-2133 @ 3.60 GHz test system by calling the
movingmarker function.