How to extract data out of a text file from specific point in a string

Hello everyone,

I'm currently trying to extract data from a text file that is not formatted to have columns and the data needed is in the middle of a string of text (see file attached). Essentially I need to store the first number that appears in each line into a matrix, but when I use textscan, or fscanf, it results in an empty cell array. I have also tried to use the extractBetween function and this has given me the closest result I am looking for but it still extracts unwanted text. The code I tried is below.

fid=fileread('input.txt');
startPat = ": ";
endPat = "[";
str=extractBetween(fid,startPat,endPat);

Ideally the final result after extracting the data will be to put the 9 values into matrix (i.e A = [1.2 20 150 20000 60 2.05 8 398 0.045]), or have each value set as their own variable.

How would I go about doing this?

Answer 1

[Walter Roberson]

fid = readlines('input.txt');
A = rmmissing(double(regexp(fid,'[+-]?\d+(\.\d*)?', 'once', 'match')));

Note: this particular version of the code does not support floating point numbers, but it does support negative and positive numbers. It also does not support decimal point without a leading digit in front of it -- for example it would support 0.5 but not .5

[Mustafa Mohammad]

Thank you very much Walter, this works perfectly.

[Image Analyst]

Seems like it does support floating point numbers.

format long g
fid = readlines('input.txt');
A = rmmissing(double(regexp(fid,'[+-]?\d+(\.\d*)?', 'once', 'match')))

A = 9×1

1.0e+00 * 1.2 20 150 20000 60 2.05 8 398 0.045

Answer 2

[Image Analyst]

Try this. It's one way

format long g
textLines = readlines('input.txt');
index = 1;
for k = 1 : numel(textLines)
    if ~contains(textLines{k}, '[')
        continue;
    end
    numbers(index) = str2double(extractBetween(textLines{k}, ':', '['));
    index = index + 1;
end
numbers' % Show in command window.

ans = 9×1

1.0e+00 * 1.2 20 150 20000 60 2.05 8 398 0.045