Use string to index variables

Hello,

So far I have data that depend on several parameter, something like that

%data first index is the year 1 for 2021 2 for 2022
%data second index is the person 1 for Ana 2 for Bob
%data third index is 1 the size 2 the age
data[1,2,2]=43; %the age of BOB in 2021

I want something like this

data['2021','Bob','age']=43;

And ideally if I want to add someone else I could just do

data['2021','Camille','age']=55; %adding a new person

Is there any syntaxe in matlab that allows me to do that ?

[Stephen23]

"Is there any syntaxe in matlab that allows me to do that ?"

You could use the fieldnames of a structure:

data.y2021.Bob.age = 43;
data.y2021.Camille.age = 55;

but a much better would be to use a table or non-scalar structure (and not force meta-data into fieldnames), e.g.:

S(1).year = 2021;
S(1).name = 'Bob';
S(1).age  = 43;
S(2).year = 2021;
S(2).name = 'Camille';
S(2).age  = 55;

[Matteo Bonhomme]

This doesn't work for me because I'd like to do stuff like this

for i=['Bob','Alice']
    dostuff(i)=data['2021',i,'age']
end

This works with my equivalence table with something like this but is way less lisible

for i=[1,2]

dostuff(i)=data[1,i,2]

end

[Stephen23]

Note that in MATLAB square brackets are a concatenation operator, so your code

['Bob','Alice']

is exactly equivalent to

'BobAlice'

"This doesn't work for me because I'd like to do stuff like this"

It works for me:

data.y2021.Bob.age = 43;
data.y2021.Alice.age = 55;
for name = ["Bob","Alice"]
    data.y2021.(name).age
end

ans = 43

ans = 55

https://www.mathworks.com/help/matlab/matlab_prog/generate-field-names-from-variables.html

If you have a recent enough version, you could possibly use a dictionary:

https://www.mathworks.com/help/matlab/dictionary.html

https://blogs.mathworks.com/matlab/2022/09/15/an-introduction-to-dictionaries-associative-arrays-in-matlab/

Note that meta-data is data, and most tasks are easier to solve in MATLAB when data is stored in variables.

[Matteo Bonhomme]

It worked !!!

Thank you :)