Sorting cell for only dublicate values

Hi i have a cell containing strings which are only 3 characters. I wish to sort this cell so only values that appear twice or more will remain. I thought about using a logical array, and using fun2cell, but i cant seem to get it quite right.

[Image Analyst]

Give a small example, like

ca = {'abc', 'aac', 'cde', 'eee'}

and tell us what the output should be. Do you want only cells 2 and 4 to be in the output cell array?

[dpb]

@Image Analyst, I'd think from the Q? that the output array for that sample input would be the null set; there are none that are repeated. I guess it's possible he means the letters inside the strings...

Indeed, we would need more definition to be precise.

[Magnus Rasmussen]

Of course. I will try and give as much information as i can

BSW01_Rawtable is from the readtable function where import an excell sheet. "b" is one column from the excell sheet.

Here you see the content of b. I wish to sort it so every value appears twice or more. If there is less than 1 entry of value for example A05, i wish to have it sorted out of the list.

Hope this clarify things. Thanks for your quick replies. I know it would probably be faster to just sort it manually in the excell sheet, but i got more things of the same task and i value the learning experience. Apologies for my bad matlab language.

Answer 1

[dpb]

Alternate approach to Jan's using features of categorical arrays (which the IDs logically are)

% some dummy data of the same type as the example -- convert to categorical
>> CID=categorical(sort(cellstr(strcat(char('A'+randi([0 2],10,1)),num2str(randi(5,10,1),'%02d')))));  
>> CID
CID = 
  10×1 categorical array
     A02 
     A03 
     A04 
     B02 
     B02 
     B02 
     C02 
     C02 
     C04 
     C04 
>> 
% The engine
c=categories(CID);                      % a list of the categories (unique list of IDs)
isG1=(countcats(CID)>1);                % logical vector of those with >1 occurrences
ID=CID(ismember(CID,c(isG1)));          % select those out of the total list

The above yields

>> ID
ID = 
  7×1 categorical array
     B02 
     B02 
     B02 
     C02 
     C02 
     C04 
     C04 
>> 

which can be seen by inspection to be the duplicates in the original list.

[Magnus Rasmussen]

Yes this is perfect. Thank you very much.

Answer 2

[Jan]

Is the input really a cell containging strings? Or a cell string? Ort a string array?

% Test data with a cell string:
data = sprintfc('%03d', randi([0, 100], 1, 100));

data = sort(data(isMultiple(data)))

function T = isMultiple(A)
   [S, idx] = sort(A(:).');
   if iscellstr(A)
      m = [false, strcmp(S(1:nA - 1), S(2:nA))];   
   elseif isa(A, 'string')
      m = [false, (S(1:nA - 1) == S(2:nA))];
   else
      error(['Jan:', mfilename, ':BadInputType'], ...
         'Input type is not handled: %s', class(A));
   end
   
   ini    = strfind(m, [false, true]);
   m(ini) = true;        % Mark 1st occurence in addition
   T(idx) = m;           % Restore original order
end