Determine which cells contain elements of another cell

I have the following cell aray:

fileNames = [{'.'}, {'..'}, {'1.bin'}, {'1_Report.txt'}, {'2.bin'}, {'2_Report.txt'}, {'3.bin'}, {'3_Report.txt'}, {'4.bin'}];

I want to determine which *.bin files have a cooresponding *.txt file, so the final result would be:

processedFiles = [{'1.bin'}, {'2.bin'}, {'3.bin'}];

or

processedIndex = [3, 5, 7];

This seems like it should be easy but i'm struggling and hoping for some help in the right direction

Mike

Answer 1

[Adam Danz]

Here's a simpler solution than my first proposal.

fileNames = [{'.'}, {'..'}, {'1.bin'}, {'1_Report.txt'}, {'2.bin'}, {'2_Report.txt'}, {'3.bin'}, {'3_Report.txt'}, {'4.bin'}];
binNums = regexp([fileNames{contains(fileNames, '.bin')}], '\d+', 'match');
txtNums = regexp([fileNames{contains(fileNames, '_Report.txt')}], '\d+', 'match');
hasMatch = ismember(binNums, txtNums); 

processedFiles = strcat(binNums(hasMatch), '.bin');          %  = [{'1.bin'}, {'2.bin'}, {'3.bin'}];
processedIndex = find(ismember(fileNames, processedFiles));  %  = [3, 5, 7];

[Mike King]

Did exactly what i wanted, Thanks Adam!

Answer 2

[Rik]

A few calls to cellfun, a regexp and ismember will do the trick:

fileNames = [{'.'}, {'..'}, {'1.bin'}, {'1_Report.txt'}, {'2.bin'}, {'2_Report.txt'}, {'3.bin'}, {'3_Report.txt'}, {'4.bin'}];

bin_match=regexp(fileNames,'\.bin');%get the indices where .bin is in the filename
bin_names=fileNames(~cellfun('isempty',bin_match));%remove other files
bin_match=bin_match(~cellfun('isempty',bin_match));%remove empty elements

%find the base filename and compose the matching txt filenames
base_filename=cellfun(@(x,y) x(1:(y-1)),...
    bin_names,bin_match,'UniformOutput',0);
txt_names=cellfun(@(x) {[x '_Report.txt']},base_filename);

%find out which txt file actually is in the list and keep the matching bin file
[~,b]=ismember(fileNames,txt_names);
processedFiles = bin_names(b(b~=0));

Answer 3

[Adam Danz]

fileNames = [{'.'}, {'..'}, {'1.bin'}, {'1_Report.txt'}, {'2.bin'}, {'2_Report.txt'}, {'3.bin'}, {'3_Report.txt'}, {'4.bin'}];
binNumMatch = regexp(fileNames, '\d+.bin', 'match'); 
binNum = [binNumMatch{:}];
fileNumMatch = regexp(fileNames, '\d+_Report.txt', 'match'); 
fileNum = [fileNumMatch{:}];
ismem = ismember(cellfun(@str2double, regexp(binNum, '\d+', 'match')), cellfun(@str2double, regexp(fileNum, '\d+', 'match'))); 

processedFiles = binNum(ismem);                              %  = [{'1.bin'}, {'2.bin'}, {'3.bin'}];
processedIndex = find(ismember(fileNames, processedFiles));  % = [3, 5, 7];

[Rik]

Interesting how our approaches are very similar, even if this problem could be tackled in a number of ways.

[Adam Danz]

Ha! That challenged me to think of an alternative.