Change line in a text file

Hello,

I have a text file where different reactions are written as well as their kinetics constants para1 para2 para3

For example:

A+ B => C + D para1 para2 para3

This text file is read by another software that solves a set of equations. I need to test out different values for each parameter para1 para2 and para3. These parameters are in the second to last line of the text file. It’s a very big text file and I don’t want matalb to read and then rewrite everything each time. I want to do this in a loop and for every iteration I test out a different value for each parameter (so the values written on the text file are not always the same). Does anyone know how I can do this?

Thanks

[Ameer Hamza]

Do you want to create several files with different values of para1, para2 and para3? Or do you want to modify the same file several times?

[FM]

I want to modify the same file several times

[Jan]

Please post a small but relevant example. The less we have to guess for the answer, the more likely is that it matches your needs.

[FM]

My text file is really big and the last lines are like this

...

A1+ B1 => C1 + D1 para11 para12 para13

A2+ B2 => C2 + D2 para21 para22 para23

A3+ B3 => C3 + D3 para31 para32 para33

END

The text file already exists and it has over 500 lines. I want to write a code that allows me to change the values of para31, para32 and para33.

This text file is read by another software that simulates gas phase chemistry, all I need matlab to do is modify the text file. I need to test different values for para31, para32 and para33

For example if I want to test 4 values for each parameter, I store the values in vectors P1 (for para31), P2(for para32) and P3 (for para33). P1, P2 and P3 would each have 4 values.

I want a code that does this

for i=1:4

Replace “A3+ B3 => C3 + D3 para31 para32 para33” by A3+ B3 => C3 + D3 P1(i) P2(i) P3(i) [Edited]

Launch software

end

[Jan]

Does the text file really contain additional blank lines? It would be enough if you post the last few lines of the file, as long as this clears uniquely, what the contents of the file is.

How can "the last lines" be identified? What is the contents of the variables P1, P2, P3?

[FM]

The last 3 lines of the text file are

C6H5+C2H2=C6H5C2H2 12E41 -9.26 16E3

C2H2 => 2C + H2 3E6 0 40E3

END

The line I want to change is "C2H2 => 2C + H2 3E6 0 40E3" I want to change the values 3E6 0 40E3, however the reaction itself C2H2 => 2C + H2 doesn't change so that could be a way of identifying the line

The vectors P1 P2 and P3 contain the values I want to test out, for example

P1= (2E6, 4E6, 5E6)

P2=(10, 20,0)

P3=(20E3,30E3 50E3)

[Jan]

The last 3 lines of the text file are

C6H5+C2H2=C6H5C2H2 12E41 -9.26 16E3

C2H2 => 2C + H2 3E6 0 40E3

END

These are 5 lines. Of course I can guess, that the empty lines do not belong to the real text. But why do you insert them at all? Please read Format in the forum to use formatting to improve the readability. Thanks.

 The line I want to change is "C2H2 => 2C + H2 3E6 0 40E3"

Formerly you said it is the line:

 A2+ B2 => C + D para21 para22 para23

Please, FM, do not assume that we can read your mind. Explain clearly and uniquely, what you want to replace.

[FM]

I apologize, the line I want to change is

A3+ B3 => C + D para31 para32 para33

Which corresponds to

C2H2 => 2C + H2 3E6 0 40E3

What I want to change is the values 3E6 0 40E3, the beginning of the line "C2H2 => 2C + H2" stays the same so for example after modifying the line once I would have

C2H2 => 2C + H2 2E6 10 20E3

The blank lines are there in the text file, I apologize if I said 3 lines instead of 5. I didn't think having blank lines in between the text would make a difference but I was wrong, I am sorry. I am not assuming you can read my mind, I have trouble explaining exactly what I need to do but I'm trying to do my best.

I use this code that you suggested (in a file named Rewrite)

CStr  = strsplit(fileread('Test.txt'), '\n');
P1=[2E6,4E6,5E6];
P2=[10,20,30];
P3=[20E3,30E6,50E6];

index = strncmp(CStr, 'C2H2 => 2C + H2 ', 16);

for i = 1:3
    CStr{index} = sprintf('C2H2 => 2C + H2 %d %d %d', P1(i), P2(i), P3(i));
    fid = fopen(fullfile(tempdir, 'Test2.txt'));
    if fid == -1
        error('Cannot open file.');
    end
    fprintf(fid, '%s\n', CStr{:});
    fclose(fid);
end

and this time the complete error message is

   Error using Rewrite (line 13)
   Cannot open file.

However if I use this code instead, I don't get any errors

CStr  = strsplit(fileread('Test.txt'), '\n');
P1=[2E6,4E6,5E6];
P2=[10,20,30];
P3=[20E3,30E6,50E6];

index = strncmp(CStr, 'C2H2 => 2C + H2 ', 16);

for i = 1:3
    CStr{index} = sprintf('C2H2 => 2C + H2 %d %d %d', P1(i), P2(i), P3(i));
    fid = fopen('Test2.txt', 'w+');
    if fid == -1
        error('Cannot open file.');
    end
    fprintf(fid, '%s\n', CStr{:});
    fclose(fid);
end

I don't know what

fid = fopen(fullfile(tempdir, 'Test2.txt'));

does exactly so that's why I tried

fid = fopen('Test2.txt', 'w+');

instead.

Anyways thanks

[Jan]

It is still not clear to me, what "correspond" means here:

 A3+ B3 => C + D para31 para32 para33
 Which corresponds to
 C2H2 => 2C + H2 3E6 0 40E3

But I've forgot the 'w' flag in the fopen command. Without it, fopen tries to open the file for reading. When the file does not exist, the fopen fails. With 'w' the file is opened for writing. 'w+' allows reading and writing and is not useful here.

It is better to use the message in case of errors:

    [fid, msg] = fopen('Test2.txt', 'w');
    if fid == -1
        error('Cannot open file: %s', msg);
    end

I've fixed my answer and inserted the 'w' specifier.

By the way: Do not get confused by questions for clarifications. The purpose of this forum is to solve Matlab problems. It is the nature of problems, that not all details are clear initially. Even if I'm probing, the goal is to solve your problem.

[FM]

The text file contains chemical reactions and the data necessary to calculate their kinetic constants. At first to explain my problem I was trying to provide an example without actually giving the names of the components and the values of the parameters. That's why I used A1,A2,A3... and para31, para32,para33... but then when you asked for the actual content of the text file I posted the actual chemical reaction with the values of the kinetic constants

So in my example

A3+ B3 => C3 + D3 para31 para32 para33

A3, B3, C3 and D3 are chemical components. The actual reaction and the data necessary to calculate the kinetic constant is

C2H2 => 2C + H2 3E6 0 40E3

So when I say that "it corresponds" I mean

• A3 is C2H2
• there is no B3
• C is 2C (2 Carbon molecules)
• D is H2
• para1=3E6
• para2=0
• para3=40E3

I added the message in case of errors and everything works, thank you for your help

Answer 1

[Jan]

If the file has 500 lines only, importing the complete file and creating a new one is cheap an easy. Note that 500 lines of text are a lot of work for a human, but a cookie for a computer.

CStr  = strsplit(fileread(FileName), '\n');
index = strcmp(CStr, 'A2+ B2 => C + D para21 para22 para23');  % [EDITED]
for i = 1:4
  CStr{index} = sprintf('A2+ B2 => C + D %d %d %d', P1(i), P2(i), P3(i));
  fid = fopen(fullfile(tempdir, 'OutputFile.txt'), 'w');  % [EDITED 2]
  if fid == -1
    error('Cannot open file.');
  end
  fprintf(fid, '%s\n', CStr{:});
  fclose(fid);
  ... Launch software
end

Here I guess, that P1 etc is a numerical array. If this does not match, please explain the details.

[FM]

Hi, thanks Jan,

I tried out what you suggested but when I write the line

 index = strcmp(Str, 'A2+ B2 => C + D para21 para22 para23');

There are only 0 in logical variable index, it's as if it didn't find the charachters 'A2+ B2 => C + D para21 para22 para23' and it tells me

Expected one output from a curly brace or dot indexing expression, but there were 0 results.

I tried on another text file just by writing

Hello

Bye

And if I write

index = strcmp(Str, 'Hello');

it still sends me the same Expected one output from a curly brace or dot indexing expression, but there were 0 results.

[Jan]

This was a typo. See [EDITED]: Replace "Str" by "CStr". "Str" seems to be a variable left over from any other operations in you workspace. Prefer to store code in functions, such that the workspace is clean.

You can find the source of such problems using the debugger: https://www.mathworks.com/help/matlab/matlab_prog/debugging-process-and-features.html

[FM]

Yes, I had changed Str to CStr because otherwise it was an unknown variable, I still get the problem I explained above though

[Jan]

Now I'm guessing what "explained above" exactly means. Please post the code you are using and a copy of the complete error message.

If you did not meant

'A2+ B2 => C + D para21 para22 para23'

but a line, which starts with

'A2+ B2 => C + D '

use:

index = strncmp(CStr, 'A2+ B2 => C + D ', 16);

or whatever is suitable to identify the line you want to replace.