Call Functions in C++ Compiled Library

The publisher of your MATLAB^® interface to a C++ library provides you with instructions for installing the interface file and any dependent library files, if required. The publisher might give you dependent library files, ask you to install libraries from an external source, or provide a link to all relevant files. If the publisher created a toolbox using MATLAB Add-Ons, then this information might be found in the Getting Started Guide available through the Options menu for the toolbox in the Add-On Manager. If you need more information or cannot find a Getting Started Guide, then contact the publisher. For details about add-ons, see Manage Add-Ons.

The name of the interface file for library libname is libnameInterface.ext, where ext is platform-specific — .dll on Windows^®, .so on Linux^®, or .dylib on macOS.

Set Run-Time Path

MATLAB looks for the interface file on the MATLAB path and the dependent library files on the system path or run-time search path (rpath). If the publisher gives you dependent library files, you can put them in the same folder as the interface file. Alternatively, to add them to the system path, see Set Run-Time Library Path for C++ Interface. For information about locating dependent libraries, see Missing or Incorrectly Installed Run-Time Libraries.

Set MATLAB Path

Call addpath on the folder containing the interface file.

Display Help

The MATLAB help and doc functions provide help for members of the library. For example, to display help for function funcname in library libname, type:

help clib.libname.funcname

Call Function

To call a function funcname in C++ library libname with input arguments arg1,arg2,... and output argument retVal, use the MATLAB clib namespace. MATLAB automatically loads the library when you type:

retVal = clib.libname.funcname(arg1,arg2,...)

After MATLAB loads the library, you can use tab completion to view the members of the clib namespace.

MATLAB Type to C++ Type Mapping

When you pass MATLAB data as arguments to C++ methods or functions, MATLAB converts the data into types that best represent the data to the C++ language.

Each row in this table shows a MATLAB type followed by the possible C++ argument matches, from left to right in order of closeness of the match. The MATLAB types (except cell arrays) can be scalar (1-by-1) arrays or matrices. The C++ types can be scalar values or arrays. The phrase cpparray refers to these C++ array types:

box array ([])
array buffer (*)
std::vector

MATLAB Argument (Scalar or Array)	C++ Parameter Type (Scalar or Array) Types Other Than Object Closest Type <———————————————————————> Least Close Type
`logical`	`bool` `cpparray bool`	`int8_t` `cpparray int8_t` `uint8_t` `cpparray uint8_t`	`int16_t` `cpparray int16_t` `uint16_t` `cpparray uint16_t`	`int32_t` `cpparray int32_t` `uint32_t` `cpparray uint32_t`	`int64_t` `cpparray int64_t` `uint64_t` `cpparray uint64_t`	`float` `cpparray float`	`double` `cpparray double`
`double`	`double` `cpparray double`	`float` `cpparray float`	`int64_t` `cpparray int64_t` `uint64_t` `cpparray uint64_t`	`int32_t` `cpparray int32_t` `uint32_t` `cpparray uint32_t`	`int16_t` `cpparray  int16_t` `uint16_t` `cpparray  uint16_t`	`int8_t` `cpparray int8_t` `uint8_t` `cpparray uint8_t`	`std::complex<double>` `cpparray std::complex<double>` `std::complex<float>` `cpparray std::complex<float>` `std::complex<int64_t>` `cpparray std::complex<int64_t>` `std::complex<uint64_t>` `cpparray std::complex<uint64_t>` `std::complex<int32_t>` `cpparray std::complex<int32_t>` `std::complex<uint32_t>` `cpparray std::complex<uint32_t>` `std::complex<int16_t>` `cpparray std::complex<int16_t>` `std::complex<uint16_t>` `cpparray std::complex<uint16_t>` `std::complex<int8_t>` `cpparray std::complex<int8_t>` `std::complex<uint8_t>` `cpparray std::complex<uint8_t>`
`single`	`float` `cpparray float`	`double` `cpparray double`	`std::complex<float>` `cpparray std::complex<float>`	`std::complex<double>` `cpparray std::complex<double>`
`[u]int8`	`[u]int8_t` `cpparray [u]int8_t`	`[u]int16_t` `cpparray  [u]int16_t`	`[u]int32_t` `cpparray [u]int32_t`	`[u]int64_t` `cpparray [u]int64_t`	`float` `cpparray float`	`double` `cpparray double`	`std::complex<[u]int8_t>` `cpparray std::complex<[u]int8_t>` `std::complex<[u]int16_t>` `cpparray std::complex<[u]int16_t>` `std::complex<[u]int32_t>` `cpparray std::complex<[u]int32_t>` `std::complex<[u]int64_t>` `cpparray std::complex<[u]int64_t>` `std::complex<float>` `cpparray std::complex<float>` `std::complex<double>` `cpparray std::complex<double>`
`[u]int16`	`[u]int16_t` `cpparray  [u]int16_t`	`[u]int32_t` `cpparray [u]int32_t`	`[u]int64_t` `cpparray [u]int64_t`	`float` `cpparray float`	`double` `cpparray double`	`std::complex<[u]int16_t>` `cpparray std::complex<[u]int16_t>` `std::complex<[u]int32_t>` `cpparray std::complex<[u]int32_t>` `std::complex<[u]int64_t>` `cpparray std::complex<[u]int64_t>` `std::complex<float>` `cpparray std::complex<float>` `std::complex<double>` `cpparray std::complex<double>`
`[u]int32`	`[u]int32_t` `cpparray [u]int32_t`	`[u]int64_t` `cpparray [u]int64_t`	`float` `cpparray float`	`double` `cpparray double`	`std::complex<[u]int32_t>` `cpparray std::complex<[u]int32_t>` `std::complex<[u]int64_t>` `cpparray std::complex<[u]int64_t>` `std::complex<float>` `cpparray std::complex<float>` `std::complex<double>` `cpparray std::complex<double>`
`[u]int64`	`[u]int64_t` `cpparray [u]int64_t`	`float` `cpparray float`	`double` `cpparray double`	`std::complex<[u]int64_t>` `cpparray std::complex<[u]int64_t>` `std::complex<float>` `cpparray std::complex<float>` `std::complex<double>` `cpparray std::complex<double>`
complex `double`	`std::complex<double>` `cpparray std::complex<double>`	`std::complex<float>` `cpparray std::complex<float>`	`std::complex<int64_t>` `cpparray std::complex<int64_t>` `std::complex<uint64_t>` `cpparray std::complex<uint64_t>`	`std::complex<int32_t>` `cpparray std::complex<int32_t>` `std::complex<uint32_t>` `cpparray std::complex<uint32_t>`	`std::complex<int16_t>` `cpparray std::complex<int16_t>` `std::complex<uint16_t>` `cpparray std::complex<uint16_t>`	`std::complex<int8_t>` `cpparray std::complex<int8_t>` `std::complex<uint8_t>` `cpparray std::complex<uint8_t>`
complex `single`	`std::complex<float>` `cpparray std::complex<float>`	`std::complex<double>` `cpparray std::complex<double>`
complex `[u]int8`	`std::complex<[u]int8_t>` `cpparray std::complex<[u]int8_t>`	`std::complex<[u]int16_t>` `cpparray std::complex<[u]int16_t>`	`std::complex<[u]int32_t>` `cpparray std::complex<[u]int32_t>`	`std::complex<[u]int64_t>` `cpparray std::complex<[u]int64_t>`	`std::complex<float>` `cpparray std::complex<float>`	`std::complex<double>` `cpparray std::complex<double>`
complex `[u]int16`	`std::complex<[u]int16_t>` `cpparray std::complex<[u]int16_t>`	`std::complex<[u]int32_t>` `cpparray std::complex<[u]int32_t>`	`std::complex<[u]int64_t>` `cpparray std::complex<[u]int64_t>`	`std::complex<float>` `cpparray std::complex<float>`	`std::complex<double>` `cpparray std::complex<double>`
complex `[u]int32`	`std::complex<[u]int32_t>` `cpparray std::complex<[u]int32_t>`	`std::complex<[u]int64_t>` `cpparray std::complex<[u]int64_t>`	`std::complex<float>` `cpparray std::complex<float>`	`std::complex<double>` `cpparray std::complex<double>`
complex `[u]int64`	`std::complex<[u]int64_t>` `cpparray std::complex<[u]int64_t>`	`std::complex<double>` `cpparray std::complex<double>`
`char`	`cpparray char` `char16_t` `cpparray char16_t` `char32_tcpparray char32_t` `wchar_t` `cpparray wchar_t`	`std::string` `std::wstring` `std::u16string` `std::u32string`
`string` scalar See Call Function with Enum Arguments.	`enum`
`string` array cell array of character vectors	`std::vector<std::string>` `std::vector<std::wstring>` `std::vector<std::u16string>` `std::vector<std::u32string>`	`char * []`
MATLAB object of C++ type `MyClass`	C++ object of type `MyClass`
MATLAB object of MATLAB class	Unsupported

Call Function with Default Arguments

If a C++ function is defined with default arguments, then you can call the function without providing one or more trailing arguments. The function help shows the default value. For example, if the type of arg is double and its default value is 100, then help displays:

clib.libname.funcname(arg)
    Input Arguments
      arg             double = 100

These statements produce the same result:

clib.libname.funcname
clib.libname.funcname(100)

This statement is also correct, although your result might be different:

clib.libname.funcname(99)

MATLAB supports default arguments for scalar integer and floating point types.

Call Function with Enum Arguments

You can pass a string to a C++ function or method in C++ library libname that accepts clib.libname.enum as an input argument.

For example, suppose that you have a MATLAB interface to C++ library libname with a function getValue that takes clib.libname.Days as input:

help clib.libname.getValue

getValue -  clib.libname.getValue Representation of C++ function getValue.

  RetVal = clib.libname.getValue(D)
    Input Arguments
      D              clib.libname.Days  

    Output Arguments
      RetVal         int32

To display the values for clib.libname.Days, type:

doc clib.libname.Days

Enumeration Summary
Fri	
Mon	
Sat	
Sun	
Thu	
Tue	
Wed

You can call getValue with either clib.libname.Days.Sun or the string "Sun".

clib.libname.getValue(clib.libname.Days.Sun)
clib.libname.getValue("Sun")

ans = 106

Access Violation Errors

When you make calls to the library, make sure that your library was compiled in Release mode, not Debug mode. For help with other common problems, see Troubleshooting MATLAB Interface to C++ Library Issues.