how to reduce the information contained in each field in a struct array?

I have a very big struct array which can be summarized in something like this:

 A = [0     0     0     0     0     0;
      1    -1     0     0     0     0;
      0     1    -1     1     0     0;
      0     0     1    -1     0    -1;
      0     0     1     0    -1     0];

 structArray.A=A;
 structArray.B={'B1[a]' 'B2[b]' 'B3[a]' 'B4[a]' 'B5[b]'}; % correspond to the rows
 structArray.C={'C1' 'C2' 'C3' 'C4' 'C5' 'C6'}; % correspond to the columns
 structArray.D={'D1' 'D2' 'D3' 'D4' 'D5'}; % correspond to the rows
 structArray.E={'E1' 'E2' 'E3' 'E4' 'E5' 'E6'}; % correspond to the columns

If I use:

 coolData=regexp(structArray.B,'(?<=\[).+?(?=\])','match')';
 coolData=cellfun(@char, coolData);
 coolerData=strmatch('a',coolData);

I will obtain all the rows relevant to me (all the rows with an "a" in the structArray.B), in this example:

 coolerData =
     1
     3
     4

I want to delete the all the columns from A and the corresponding data on structArray.C and structArray.E (which represent the columns) where the column and the rows obtained in "coolerData" are == to zero (in this example where all the elements in A([1 3 4],columns) are equal 0, therefore, columns 1 and 5) so I can obtain:

 A = [ 0     0     0     0;
      -1     0     0     0;
       1    -1     1     0;
       0     1    -1    -1;
       0     1     0     0];

 structArray.A=A;
 structArray.B={'B1[a]' 'B2[b]' 'B3[a]' 'B4[a]' 'B5[b]'}; % correspond to the rows
 structArray.C={'C2' 'C3' 'C4' 'C6'}; % correspond to the columns
 structArray.D={'D1' 'D2' 'D3' 'D4' 'D5'}; % correspond to the rows
 structArray.E={'E2' 'E3' 'E4' 'E6'}; % correspond to the columns

And then I want to delete all the row and the corresponding data on structArray.B and structArray.D (which represent the row) where all the row has only zeros (in this example row 1), so I can obtain:

 A = [-1     0     0     0;
       1    -1     1     0;
       0     1    -1    -1;
       0     1     0     0];

 structArray.A=A;
 structArray.B={'B2[b]' 'B3[a]' 'B4[a]' 'B5[b]'}; % correspond to the rows
 structArray.C={'C2' 'C3' 'C4' 'C6'}; % correspond to the columns
 structArray.D={'D2' 'D3' 'D4' 'D5'}; % correspond to the rows
 structArray.E={'E2' 'E3' 'E4' 'E6'}; % correspond to the columns

I did that with a lot of FOR and IF but it looks ugly and take a lot of time (the size of the matrix I'm using it is 7867x12569) Do you have any suggestions, Thanks,

Answer 1

[Guillaume]

Function to filter a scalar structure:

function s = filterstruct(s)
   rowstocheck = ~cellfun(@isempty, regexp(s.B, '\[a\]', 'once'));
   colstodelete = all(s.A(rowstocheck, :) == 0);
   s.A(:, colstodelete) = [];
   s.C(colstodelete) = [];
   s.E(colstodelete) = [];
   rowstodelete = all(s.A == 0, 2);
   s.A(rowstodelete, :) = [];
   s.B(rowstodelete) = [];
   s.D(rowstodelete) = [];
end

To then apply to a structure array:

filteredarray = arrayfun(@filterstruct, structArray);

That is assuming that you do have a structure array. Your example shows a scalar structure, not a structure array.

[German Preciat Gonzalez]

Cool Thanks! It works perfectly!