How to read multiple text files in a folder and create a cell array?

Hi, i have a folder with 432 text files named 00001.txt, 000002.txt...etc. I will like to create a matrix or a cell array with the information of these files. I used the import tool with the first file for creating a function like this:

function Untitled1 = importfile1(filename, startRow, endRow) %IMPORTFILE Import numeric data from a text file as a matrix.

Then i used a for loop for importing data from each text file into the cell array:

numFiles = 432; startRow = 2; endRow = inf; myData = cell(1,numFiles);

for fileNum = 1:numFiles fileName = sprintf('00000%02d.txt',fileNum); myData{fileNum} = importfile(fileName,startRow,endRow); end

But it shows the following message:

Error using textscan Invalid file identifier. Use fopen to generate a valid file identifier.

Error in importfile (line 37) dataArray = textscan(fileID, formatSpec, endRow(1)-startRow(1)+1, 'Delimiter', delimiter, 'HeaderLines', startRow(1)-1, 'ReturnOnError', false);

I will appreciate any help, thanks

[Stephen23]

What is the function importfile? I cannot find any reference to it in the current online documentation. Did you write this function yourself, or obtain it from someone?

[Sean de Wolski]

@Stephen, Magic 8 ball says importfile is the default name given to functions generated with the import tool.

Answer 1

[Stephen23]

Why not just read the files out of the folder using dir, like this:

>> S = dir('*.txt');
>> N = {S.name};

where N is a cell array of all of the files that match the given filename string. Then you don't need to worry about generating the filenames yourself.

You might also like to compare the other file-reading functions, such as csvread, dlmread and textscan. These might be simpler to use in your loop, and a little faster too. You might also like to read MATLAB's own advice on importing a sequence of files:

http://www.mathworks.com/help/matlab/import_export/process-a-sequence-of-files.html

and the two examples given here:

http://matlab.wikia.com/wiki/FAQ#How_can_I_process_a_sequence_of_files.3F

Answer 2

[Sean de Wolski]

Instead of sprintf, use num2str to build a string that always has the same length

num2str(ii,'%05i')

(or 6i, your example is inconsistent)

compare:

>> num2str(2,'%05i')
ans =
00002
>> num2str(243,'%05i')
ans =
00243

[Angela Ayobi]

Thank you so much i already fixed that part, but it didn't work.

I tried to run this separately:

Untitled = importfile('000001.txt', 1, 6);

and it showed the same message, is there any problem in the function importfile?

Thanks,

[Stephen23]

Note that sprintf / fprintf do also work with fixed width formats:

>> fprintf('%06i\n',[1,10,100,1000])
000001
000010
000100
001000
>> sprintf('%06i',999)
ans = '000999'

Answer 3

[Sean de Wolski]

If the files are all the exact same format, you might have some luck with datastore as well (new in r2014b).

doc datastore

Answer 4

[Angela Ayobi]

thanks to all of you, it worked!