How to skip (variable) white space using textread

If I have a text file with different amount of white spaces in between the text and numbers, how can I get "textread" to ignore the (variable) whitespaces? For example see test(002).txt

Eventually I want to make a table with these variables to make plots and do statistics etc.

Answer 1

[Jeff Miller]

GetNumsInAsciiFile on File Exchange will probably do what you want: https://au.mathworks.com/matlabcentral/fileexchange/66692-getnumsinasciifile-sfname-

[Angelique Remmers]

To bad I have R2015b..

[Walter Roberson]

The code in that submission does not use anything newer than R2013a.

Answer 2

[Walter Roberson]

I suggest using

readtable()

I do not recommend textread() in any release from R14 onwards. (I do not mean R2014* -- I mean R14, MATLAB 7.0, release in June 2004.)

[Angelique Remmers]

If I try readtable, I get the error "input must be a row vector of characters". Even if I make a txt or dat file with the text from te example I get the same error.

filename = 'testfile.dat';
fileID = fopen(filename)
T = readtable(fileID)

[Angelique Remmers]

This is the text in the file:

Lastname,Gender,Age

Smith,M,38

Johnson,M,43

[Angelique Remmers]

I think it has to do with fopen. Since the fileID is -1, and I read this is when fopen cannot open the file..

[Walter Roberson]

You do not readtable() a file identifier, you readtable() a file name.

T = readtable(filename);

However if you are getting a fileID of -1 then it probably cannot find the file, and you might need to specify the directory name as well. For example,

[filename, filepath] = uigetfile('*.dat', 'Choose a data file');
if ~ischar(filename)
  return;   %user cancel
end
filename = fullfile(filepath, filename);
T = readtable(filename);

[Angelique Remmers]

Fopen now works correct, thanks!

The initial problem however is still there. I don't see how I can solve the problem with the inconsistent white spaces between the variables/numbers. The Delimiter is variable since the amount of white spaces differ, so cannot use this. If I use Format he reads the whole row as one string, in stead of one string and 4 numbers.

In my text file the first row contains 5 different words, with variables white spaces amounts in between. And the upcoming rows contain one word, followed by 4 numbers, also with variable white space amounts in between.

[Angelique Remmers]

Just found the MultipleDelimsAsOne! Thanks again.

[Walter Roberson]

readtable() will take care of this for you.

If you use textscan() then use a format of '%s%s%s%s', and do not bother using MultipleDelimsAsOne . Any amount of whitespace on input is skipped before a %s or numeric format item is processed.