how to initialise field name of struct array?

I want to set some fields of a struct array, like .name .age .id then I want to put data in one line. Is it possible to do something like this?

% initialise data struct with fields .name, .age, .id in this order, than put data like this:
data(1) = ... ('John',12,'ABC123') ... ;

The result what I want is this:

data(1).name = 'John';
data(1).age = 12; 
data(1).id= 'ABC123';

Answer 1

[Guillaume]

How is this question different from your previous one on the same subject?

data = cell2struct({'John', 12, 'ABC123'}, {'name', 'age', 'id'}, 2)

[Mr M.]

Sorry but the initialisation is important for me, because I don't want to write {'name', 'age', 'id'} in each row, this is the point!

[Guillaume]

You can initialise the whole array at once:

data = cell2struct({'John', 12, 'ABC123'; 'Adam', 15, 'DEF456'; 'Lucy', 7, 'XXX000'}, {'name', 'age', 'id'}, 2)

IIf your struct array already exists and you want to initialise another element, you can use fieldnames to retrieve the names of the fields:

data = struct('name', {}, 'age', {}, 'id', {}); %creates an empty struct with fields
data(1) = cell2struct({'John', 12, 'ABC123'}, fieldnames(data), 2);
data(2) = cell2struct({'Adam', 15, 'DEF456'}, fieldnames(data), 2);
data(3) = cell2struct({'Lucy',  7, 'XXX000'}, fieldnames(data), 2);

Answer 2

[Walter Roberson]

Only in the case where you initialize all of the fields for the structure.

% initialise data struct with fields .name, .age, .id in this order, than put data like this:
data(1) = struct('name', 'John', 'age', 12, 'id', 'ABC123');

Or you can use cell2struct as shown by Guillaume, which will have the same restriction that every entry you do this for must have the same fields in the same order.

If you want a more compact syntax then you should consider using a small function (could be anonymous) which expands the syntax; you should also consider using Object Oriented Programming, in which case you would be calling a Constructor for your class.

[Mr M.]

Sorry but the initialisation is important for me, because I don't want to write 'name', 'age', 'id' in each row, this is the main point!

[Walter Roberson]

Like I wrote "If you want a more compact syntax then you should consider using a small function (could be anonymous) which expands the syntax"

INI = @(name, age, id) struct('name', name, 'age', age, 'id', id);
data(1) = INI('John', '12', 'ABC123');

[Walter Roberson]

By the way: remember that the MATLAB editor has search and replace so you can put in unique short character sequences to save time writing, and then you can replace them with the expanded sequence. Like writing

   $S($N,'John', $A, 12, $I, 'ABC123');

and then doing a search and replace of $S -> struct, $N -> 'name', and so on.

[Mr M.]

ok, so it is not possible to avoid write them down more than one :) I think this is a very ugly text in script: ('name', name, 'age', age, 'id', id) . But as far as I understand, this is the best :(

[Walter Roberson]

Huh? I just showed you how to do it using a small function. If writing INI is too much then you can use a shorter name.

S = @(name, age, id) struct('name', name, 'age', age, 'id', id);

data(1) = S('John', 12, 'ABC123');
data(2) = S('Fred', 37, 'PQ19B');
data(3) = S('Kansola', 5, 'ZZ9 plural Z alpha');

[Guillaume]

Much of the skill of writing code is not only in making it work but also in making it readable for whoever is going to maintain it. As such, using shortcuts that make it easier to write the code but make it difficult to understand should be avoided.

Code it typically, write once, read many. The read many is what needs to be optimised, not the write once.