fieldfun: Apply a function to the matching fields of structures

Summary

S = fieldfun(fun,S1,...,SN) passes the values for each field in structures S1,...,SN to function fun and returns the result in scalar structure S. Each field F in S, i.e. S.(F), is the result of fun(S1.(F),...,SN.(F)).

fieldfun allows the organization of data by structure and field name while allowing the data to be easily processed according to the semantically meaninful fields, improving code readability. When only one structure is supplied, fieldfun can also be used as an alternative to structfun, the difference being that the output of structfun is an array, while fieldfun returns a structure, retaining the input format, again improving readability.

Inputs

• fun is either a function handle or a text scalar giving the function's name or operator. The function must take N input arguments, i.e. equal to the number of input structures, and return a value. fun is called once for each field F in S1,...,SN.
• S1 is a scalar structure or array of structures.
• S2,...,SN are scalar structures, arrays of structures, or variables of any other class. If S2,...,SN are structures, their fields and field order must match S1. If S2,...,SN are not structures, then they are converted to structures, with the value being assigned to each field.

Output

• S is a scalar structure with the same fields and field order as S1. The value of each field F in S, i.e. S.(F), is the result of fun(S1.(F),...,SN.(F)).

Example

treatmentGroup(1) = struct( 'name', "John", 'age', 30, 'vaccinated', true );
treatmentGroup(2) = struct( 'name', "Jane", 'age', 80, 'vaccinated', false );
controlGroup = struct( 'name', "Jim", 'age', 50, 'vaccinated', true );
allParticipants = fieldfun( @horzcat, treatmentGroup, controlGroup )

allParticipants = struct with fields:
        name: ["John"    "Jane"    "Jim"]
         age: [30 80 50]
  vaccinated: [1 0 1]

More examples available in examples.mlx / examples.pdf.

Compatibility, License and Citation

Created in 2022b. Compatible with MATLAB release 2019b and later. Compatible with all platforms.

Published under MIT License (see LICENSE.txt).

Please cite George Abrahams (https://github.com/WD40andTape/fieldfun, https://www.linkedin.com/in/georgeabrahams).