WOLFRAM

Wolfram Language & System Documentation Center

Cast[val,type]

converts val to the type type, for use in compiled code.

Cast[val,type,method]

converts val to the type type using the specified casting method.

Details
Details and Options Details and Options
Examples  
Basic Examples  
Scope  
Possible Issues  
See Also
Related Guides
History
Cite this Page

Cast

Cast[val,type]

converts val to the type type, for use in compiled code.

Cast[val,type,method]

converts val to the type type using the specified casting method.

Details

  • Cast is only available in compiled code.
  • Possible values of method in Cast[val,type,method] include:
  • "BitCast"reinterprets the bits representing val as having the type type
    "CCast"emulates the behavior of casting in C
    "ConversionCast"converts val to the type type
  • The default casting method is "ConversionCast".
  • Cast[val,type,"BitCast"] behaves as though val is written to memory and read back as type.
    The input type to a bitcast must have the same memory size as the output type.
  • Cast[val,type,"CCast"] only supports numeric types.
  • Cast[val,type,"ConversionCast"] can fail with an error.

Examples

open all close all

Basic Examples  (2)

Cast an "Integer64" to a "Real64" in a compiled function:

Wolfram Language code: cf = FunctionCompile[Function[Typed[arg, "Integer64"], Cast[arg, "Real64"] ]]
Wolfram Language code: cf[2]

Cast an expression to a packed array, suitable for use with other compiled functions:

Wolfram Language code: cf = FunctionCompile[Function[Typed[arg, "InertExpression"], Reverse[Cast[arg, "PackedArray"::["Real64", 1]]] ]]
Wolfram Language code: cf[{1, 2, 3}]

Some expressions cannot be converted into packed arrays:

Wolfram Language code: cf["test"]

Scope  (2)

Bitcast a pointer to an integer:

Wolfram Language code: cf = FunctionCompile[Function[{}, Module[{ptr, res}, ptr = LibraryFunction["malloc"][Typed[1, "UnsignedInteger64"]]; res = Cast[ptr, "UnsignedInteger64", "BitCast"]; LibraryFunction["free"][ptr]; res ] ]]
Wolfram Language code: cf[]

Cast from a larger type to a smaller one with a C-style cast:

Wolfram Language code: cf = FunctionCompile[Function[Typed[arg, "Integer64"], Cast[arg, "Integer32", "CCast"]]]
Wolfram Language code: cf[100]

As in C, the "CCast" method does not check for numerical overflows:

Wolfram Language code: cf[2 ^ 32]

Possible Issues  (1)

The default casting method "ConversionCast" does not currently support casting from larger types to smaller types:

Wolfram Language code: cf = FunctionCompile[Function[Typed[arg, "Integer64"], Cast[arg, "Integer32"]]]

Instead, "CCast" can be used:

Wolfram Language code: cf = FunctionCompile[Function[Typed[arg, "Integer64"], Cast[arg, "Integer32", "CCast"]]]

However, "CCast" does not check for numerical overflows:

Wolfram Language code: cf[2 ^ 32]
Wolfram Research (2022), Cast, Wolfram Language function, https://reference.wolfram.com/language/ref/Cast.html.

Text

Wolfram Research (2022), Cast, Wolfram Language function, https://reference.wolfram.com/language/ref/Cast.html.

CMS

Wolfram Language. 2022. "Cast." Wolfram Language & System Documentation Center. Wolfram Research. https://reference.wolfram.com/language/ref/Cast.html.

APA

Wolfram Language. (2022). Cast. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/ref/Cast.html

BibTeX

@misc{reference.wolfram_2026_cast, author="Wolfram Research", title="{Cast}", year="2022", howpublished="\url{https://reference.wolfram.com/language/ref/Cast.html}", note=[Accessed: 13-June-2026]}

BibLaTeX

@online{reference.wolfram_2026_cast, organization={Wolfram Research}, title={Cast}, year={2022}, url={https://reference.wolfram.com/language/ref/Cast.html}, note=[Accessed: 13-June-2026]}