Alternatives to the eval Function - MATLAB & Simulink (original) (raw)

Why Avoid the eval Function?

Although the eval function is very powerful and flexible, it is not always the best solution to a programming problem. Code that calls eval is often less efficient and more difficult to read and debug than code that uses other functions or language constructs. For example:

• MATLAB® compiles code the first time you run it to enhance performance for future runs. However, because code in an eval statement can change at run time, it is not compiled.
• Code within an eval statement can unexpectedly create or assign to a variable already in the current workspace, overwriting existing data.
• Concatenated character vectors within an eval statement are often difficult to read. Other language constructs can simplify the syntax in your code.

For many common uses of eval, there are preferred alternate approaches, as shown in the following examples.

Variables with Sequential Names

A frequent use of the eval function is to create sets of variables such as A1, A2, ..., An, but this approach does not use the array processing power of MATLAB and is not recommended. The preferred method is to store related data in a single array. If the data sets are of different types or sizes, use a structure or cell array.

For example, create a cell array that contains 10 elements, where each element is a numeric array:

numArrays = 10; A = cell(numArrays,1); for n = 1:numArrays A{n} = magic(n); end

Access the data in the cell array by indexing with curly braces. For example, display the fifth element of A:

ans = 17 24 1 8 15 23 5 7 14 16 4 6 13 20 22 10 12 19 21 3 11 18 25 2 9

The assignment statement A{n} = magic(n) is more elegant and efficient than this call to eval:

eval(['A', int2str(n),' = magic(n)']) % Not recommended

For more information, see:

• Create Cell Array
• Structure Arrays

Files with Sequential Names

Related data files often have a common root name with an integer index, such as myfile1.mat through myfileN.mat. A common (but not recommended) use of the eval function is to construct and pass each file name to a function using command syntax, such as

eval(['save myfile',int2str(n),'.mat']) % Not recommended

The best practice is to use function syntax, which allows you to pass variables as inputs. For example:

currentFile = 'myfile1.mat'; save(currentFile)

You can construct file names within a loop using the sprintf function (which is usually more efficient than int2str), and then call the save function without eval. This code creates 10 files in the current folder:

numFiles = 10; for n = 1:numFiles randomData = rand(n); currentFile = sprintf('myfile%d.mat',n); save(currentFile,'randomData') end

For more information, see:

• Choose Command Syntax or Function Syntax
• Import or Export a Sequence of Files

Function Names in Variables

A common use of eval is to execute a function when the name of the function is in a variable character vector. There are two ways to evaluate functions from variables that are more efficient than using eval:

• Create function handles with the @ symbol or with the str2func function. For example, run a function from a list stored in a cell array:
  examples = {@odedemo,@sunspots,@fitdemo};
  n = input('Select an example (1, 2, or 3): ');
  examples{n}()
• Use the feval function. For example, call a plot function (such as plot, bar, or pie) with data that you specify at run time:
  plotFunction = input('Specify a plotting function: ','s');
  data = input('Enter data to plot: ');
  feval(plotFunction,data)

Field Names in Variables

Access data in a structure with a variable field name by enclosing the expression for the field in parentheses. For example:

myData.height = [67, 72, 58]; myData.weight = [140, 205, 90];

fieldName = input('Select data (height or weight): ','s'); dataToUse = myData.(fieldName);

If you enter weight at the input prompt, then you can find the minimum weight value with the following command.

For an additional example, see Generate Field Names from Variables.

Error Handling

The preferred method for error handling in MATLAB is to use a try, catch statement. For example:

try B = A; catch exception disp('A is undefined') end

If your workspace does not contain variable A, then this code returns:

Previous versions of the documentation for the eval function include the syntax eval(expression,catch_expr). If evaluating the expression input returns an error, then eval evaluates catch_expr. However, an explicit try/catch is significantly clearer than an implicit catch in an eval statement. Using the implicit catch is not recommended.