Extract specific field names and associated data from a structure

Dear all,

here is a screenshot of my structure I created using following code:

file = 'myfile';
struct = dir (fullfile (file, '*.xlsx'));

for i = 1:numel (struct);
    F = fullfile (file, struct(i).name);
    [struct(i).num, struct(i).txt, struct(i).raw] = xlsread (F); % insert .num, .txt. and .raw into the structure 
end

Attached you find the data for one .xlsx file. It is possible for me to extract the fields seperaly using:

x = struct(1).num (8:14, 3:end); % extract position 1 field .num; extract only row 8.14 and column 3:end

Know I want to extract all positions with 'bf_s' simultaniously using a if statement to analyse the data for each condition seperatly to calculate single and mean values.

Thank you for your support!

Regards, Jonas

Answer 1

[Stephen23]

Rather than using an IF it would be simpler to specify an appropriate filename for DIR:

P = 'myfile';
S = dir(fullfile(P,'*bf_s.xlsx')); % specify BF_S here
for k = 1:numel(S);
    F = fullfile(P,S(k).name);
    [S(k).num, S(k).txt, S(k).raw] = xlsread(F);
    S(k).ext = S(k).num(8:14,3:end);
end

You can loop over S or use the methods shown here:

https://www.mathworks.com/help/matlab/matlab_prog/access-multiple-elements-of-a-nonscalar-struct-array.html

Is there a particular reason why you want to use XLSREAD and not READMATRIX/READTABLE ?

[Jonas Bender]

Dear Stephen,

thanks for your support. When using readtable I get an error message: Too many output arguments and I can't fix that bug. Usually it is trivial. But I don't get it. Suggestions welcome.

So far I have to specify DIR for three filenames (bf_s, nb_s and ob_s) in three lines of code:

Is there a more clever solution?

Regards, Jonas

file = 'myfile';
struct = dir (fullfile (file, '*bf_s.xlsx')); Das % data #1

for i = 1:numel (struct);
    F = fullfile (file, struct(i).name);
    [struct(i).num, struct(i).txt, struct(i).raw] = readtable (F);
    struct(i).ext = struct(i).num (8:14,3:end);
end
...
%% further calculations
...
    
file = 'myfile';
struct = dir (fullfile (file, '*nb_s.xlsx')); % data #2

for i = 1:numel (struct);
    F = fullfile (file, struct(i).name);
    [struct(i).num, struct(i).txt, struct(i).raw] = readtable (F);
    struct(i).ext = struct(i).num (8:14,3:end);
end
...
% further calculations 
...
%     and so on

[Stephen23]

"Too many output arguments and I can't fix that bug. Usually it is trivial. But I don't get it."

Did you read the documentation for the functions that you are trying to use?

XLSREAD has three outputs.

READTABLE and READCELL and READMATRIX all have just one output.

So if you swap XLSREAD (like in your question and my answer) with three outputs for a function that only has one output, then obviously you can only call it with that one output that the function supports.

"So far I have to specify DIR for three filenames (bf_s, nb_s and ob_s) in three lines of code: Is there a more clever solution?"

If the code for for processing the files is exactly/mostly the same, then you can simply concatenate the DIR output together before the loop:

P = 'myfile';
S = vertcat(... Do NOT use STRUCT as a variable name!
    dir(fullfile(P, '*bf_s.xlsx')),...
    dir(fullfile(P, '*nb_s.xlsx')),...
    );
for k = 1:numel(S)
    .. etc.
end

Do NOT name your variable STRUCT, as this shadows the very important inbuilt STRUCT function.