processing a file but keeping blank lines intact

Hello,

I have a script that takes a file which contains 2 columns of data (in scientific notation), then subtracts a constant value from the first column, in order to commensate for a glitch in our machine which makes the measuremetns. A new file is then produced with the new values.

However, the original file contains blank lines throughout the sequence, and the new file does not.

How do I add a line into the script that tells it to keep blank lines?

Is this possible?

Thank you!

[dpb]

Almost anything is possible, but to have a klew as to how to do your particulary anything we'd have to see at least the processing code you have...

The issue revolves about having numeric data and blanks together in a numeric array -- they can't coincide. Probably the best way to go at it would be to read with a function that has the facility to indicate missing values -- they'll then be NaN which are easy enough to deal with; then to rewrite the file convert to cellstr array and replace the NaN with the empty string.

Alternatively, if you were to process line by line, you could simply test that the given line is empty and echo it back out to the file unchanged while processing those which do have data before rewriting them.

Again, "it all depends"; as per usual, it would help to see a small(ish) sample file that contains the problem/feature for folks to poke at ...

[Christina Verhagen]

Thank you for the responses!

Here are some images of both the way the data should look (with blank lines) and the script I am working on.

Sorry new here!

[Walter Roberson]

Please post code rather than pictures of code. We cannot execute pictures of code.

[Voss]

@Christina Verhagen: I've modified my answer to work with your example file, please try to run it and see if it does what you want.

Answer 1

[Voss]

Here's one way:

input_file = 'file.txt';
output_file = 'file_modified.txt';

constant_offset = 100;

input_format = "%f,%f";        % modify this to match your file's format for reading
output_format = '%+.6E,%+.6E'; % modify this to try to match your file's format for writing

type(input_file); % show the original file's contents

    Field         Moment
     (0e)         (emu)
+661.6140E+00,+125.8343E-06

+329.1367E+00,+85.14747E-06

+661.6153E+00,+126.2483E-06

+324.2799E+00,+84.55759E-06
+329.0182E+00,+85.26904E-06

L = readlines(input_file);

has_numbers = false(numel(L),1);
M = zeros(0,2);
for ii = 1:numel(L)
    vals = textscan(L(ii),input_format);
    vals = [vals{:}];
    if numel(vals) == 2
        has_numbers(ii) = true;
        M(end+1,:) = vals;
    end
end

M(:,1) = M(:,1)-constant_offset;
L(has_numbers) = compose(output_format,M);
writelines(L,output_file);

type(output_file); % show the modified file's contents

    Field         Moment
     (0e)         (emu)
+5.616140E+02,+1.258343E-04

+2.291367E+02,+8.514747E-05

+5.616153E+02,+1.262483E-04

+2.242799E+02,+8.455759E-05
+2.290182E+02,+8.526904E-05

[Christina Verhagen]

Awesome! That did it!

Thank you so much for your help!

[Voss]

You're welcome!

Answer 2

[Walter Roberson]

ConstantOffset = as appropriate;

filename_in = 'as appropriate';
filename_out = 'as appropriate preferably not the same';

C = readcell(filename_in);
mask = cellfun(@(V) isnumeric(V) && ~isempty(V), C(:,1));
C(mask) = cellfun(@(V)V-constant_offset, C(mask,1), 'uniform', 0);

writecell(C, filename_out);

[Christina Verhagen]

This works! Sorry for uploading a picture. I will show code next time.

Thank you for your help! I truly appreciate your time!