opening many files and changing code to match file name

Hello,

I was hoping that somebody might be able to help me with a problem that I have. I have 1000 files in the current directory called

RW1.a_process
RW2.a_process
RW3.a_process...
RW1000.a_process

I was wondering for each file how can I open it and change 2 lines in the code to be the same as the file name.

Here is an example http://dl.dropbox.com/u/54057365/All/screenshotmatlab.JPG

This file is called RW1.a_process

You will see RW1 in the two places marked red.

Is there anyway that I could do this for the remainder of the files? It would take me hours to do it manually.

Sincere thanks

John

[Walter Roberson]

Ah, this would be very easy if you were using OS-X or Linux...

[Jiro Doke]

What do the 2 lines in the code have currently? I assume you have some placeholder text, and you want to change those according to the file name?

[Rick Rosson]

Are the two lines of text always lines 2 and 15, or do they change location from one file to the next?

[John]

Hello Jiro and Rick,

I'm using a software that is based on the Matlab environment. However, in order to run a simulation a process file is necessary.

What I did was, I copied an original process file a 1000 times and changed the file names to RW1.a_process ... RW1000.a_process.

The process files are all the same, except for the two parts that I've marked in red. The first part is just the display name in the program and the second corresponds to separate .mat file which contains data for the simulation. They are named RW1.mat...RW1000.mat

I could do it manually but since I have a 1000+ files it would take days to complete.

So just to confirm the two lines of text are always lines 2 and 15.

Jiro - I hope this answers your question too.

Thank you

Regards

John

Answer 1

[Jiro Doke]

This is probably a little too specific to your case, i.e. it's looking for a particular text. I'm just using strrep but if it doesn't work, you may have to look at regexprep.

In general, I agree with Rick that you should do this from the beginning (using your original file), instead of creating 1000 copies, changing the file names, and then changing those 2 lines.

d = dir('*.a_process');
for id = 1:length(d)
    [p, name] = fileparts(d(id).name);
    fid = fopen(d(id).name, 'r');
    str = textscan(fid, '%s', 'delimiter', '\n', 'whitespace', '');
    fclose(fid);
    % Replace Strings
    newStr = strrep(str{1}, 'Run RW Cycle', ['Run ', name, ' Cycle']);
    newStr = strrep(newStr, 'RW.mat', [name, '.mat']);
    % Overwrite
    fid = fopen(d(id).name, 'w');
    fprintf(fid, '%s\n', newStr{:});
    fclose(fid);
end

[John]

This works great!

Answer 2

[Rick Rosson]

It may be easier to delete all of the files except for the original source file, and then use MATLAB to create all of the copies for you, but with the two specific lines customized on each copy.

Maybe, for example:

   for k = 2:1000

       fileName = ['RW'  num2str(k)];
       source = fopen( 'RW1.a_process','r');
       target = fopen([fileName '.a_process'],'w');

       Ln = 0;
       while ~feof(source)
           Ln = Ln + 1;
           aLine = fgetl(source);

           if Ln == 2
               aLine = sprintf(...);
           else
               if Ln == 15
                   aLine = sprintf(...);
               end
           end

           fprintf(target,'%s\n',aLine);

       end

       fclose(target);
       fclose(source);

   end

I realize that this code is not optimized, but it may help as a starting point.

HTH.

Rick