How to properly use textscan and regexp to sift through a poorly formatted text file?

I am trying to read in a text file that has different formatting for each of it's 1000+ lines into matlab so that I can pull very specific information out. I am able to read in the text file using textscan (skipping first 23 since I do not need that information), but I have not been able to use regexp to find the strings I am looking for. I think this is because the way each line is formatted (char) is not being recognized by regexp as consecutive characters. See how strangely my "proc" variable is formatted. I attached my text file.

fileID = fopen('FIST_fMRI_Task_Run1_adultpilot-8395-1 (1).txt');

inputtxt = textscan(fileID, '%s', 'delimiter', '\n', 'HeaderLines', 23);

proc = inputtxt{1}{3}

y = regexp(proc, '\w*[Proc]', 'match')

x =regexp(proc, 'FlexibilityProc', 'match')

proc =

         P r o c e d u r e :   F l e x i b i l i t y P r o c 

y = 

      'P'    'r'    'o'    'c'    'r'    'P'    'r'    'o'    'c'

x = 

       {}

[Guillaume]

I strongly suspect that

y = regexp(proc, '\w*[Proc]', 'match')

does not do what you want. It matches any number (including none) of letters, numbers or _ followed by either 'P', 'r', 'o' or 'c'.

Answer 1

[Walter Roberson]

Change

fileID = fopen('FIST_fMRI_Task_Run1_adultpilot-8395-1 (1).txt');

to

fileID = fopen('FIST_fMRI_Task_Run1_adultpilot-8395-1 (1).txt', 'n', 'UTF16-LE');

and ignore the warning that says UTF16 is not supported (it is writing of the files that is not supported.)

[Dina Dajani]

when I try to get fileID with the code you specified, it gives me a permission error.

>> txtfile = strcat(folder, '/FIST_fMRI_Task_Run1_adultpilot-8395-1 (1).txt');

>> fileID = fopen(txtfile, 'n', 'UTF16-LE'); Error using fopen Invalid permission.

>> fileID = fopen(txtfile)

fileID = 5

[Guillaume]

Walter missed the permission argument in the fopen call. It should have read:

fopen(xxx, 'r', 'n', 'UTF16-LE')

Note that your file appears to have a BOF marker that matlab reads as the 1st character of the 1st line. In your case, it does not matter as you ignore the first 23 lines anyway.

[Dina Dajani]

This worked, thank you