how can I convert comma with decimal points for a lot of textfiles?

Hello,

I would like to import 10 textfiles which have comma as decimal delimiter.

All text file look like:

Time Fn Ft

0,1 1,2 5,9

0,2 1,8 9,4

0,3 1,9 10,8

Does anyone know how to do import those textfiles and convert comma with decimal points?

Thank you for your help

Answer 1

[per isakson]

See Import data file with comma decimal point

[afrya]

Ok thanks

Answer 2

[Stephen23]

MATLAB high-level data reading functions (eg textscan, csvread , etc) only accept the period character (.) as the decimal radix point. This also applies to sscanf and the other string->numeric conversion functions.

Basically you need to convert the comma to a period before converting the strings to numerics. You can do this:

• externally in your favorite text editor, or
• within MATLAB by

1. reading the whole string using fileread
2. performing a regexprep or strrep operation
3. doing the numeric conversion using textscan or your function of choice.

Because the every file format is different and the presence of periods can make this complicated, you need to know exactly how the data is formatted before you can try this.

[afrya]

Thank you for your answer

Answer 3

[Geoff Hayes]

Afrya - if I copy and paste your three line example into a file called data.txt as

 0,1 1,2 5,9
 0,2 1,8 9,4
 0,3 1,9 10,8

I can then use the importdata function to read all the data into a cell array (cell array because your input data comes in as strings due to the comma) with the code

 data = importdata('data.txt','\t')

In the above, I'm assuming that each column is separated by a tab. Now, I just loop through each row in the matrix and replace the commas with a period as

 numericData = [];
 for k=1:size(data,1)
     numericData(k,:) = str2num(char(strrep(data(k,:),',','.')'));
 end

with numericData being populated as

 numericData =

    0.1000    1.2000    5.9000
    0.2000    1.8000    9.4000
    0.3000    1.9000   10.8000

Try the above and see what happens! (Note that if you have a header row in your data file, you should be able to ignore it. See importdata .

[afrya]

Thanks, however when I try to import my textfile, I get this error message:

Subscripted assignment dimension mismatch.

Error in test_removal_header (line 10) numericData(k,:) = str2num(char(strrep(data(k,:),',','.')'));

[Geoff Hayes]

Afrya - you'd have to attach your file (so that I can see what the problem is) or copy and paste the row from data that is generating the error.

[afrya]

You'll find the textfile that I am working with.

[Geoff Hayes]

Afrya - there appears to be nine header rows that you will want to ignore when importing the data. See the documentation for importdata that will describe how you can ignore these lines.

[afrya]

Ok thanks

Answer 4

[Roland Antal]

I know it's too late to answer that question because it's actually four years later but I found the solution right now, so maybe will still help others who will search for solution in future. Une notepad++ Ctrl+F find comma and than is there a replace all function which helps us. Good luck everybody!

[Walter Roberson]

This is the option Stephen suggested, "externally in your favorite text editor,"