Extract data from a .txt file

Hello! The data inside the .txt file has many sets of data separated by the same word. I would like to have so for each set of data one row with 2 columns, which will contain: the reference number after 'ref' and from the 26th row the value from the 4th column. So, for example from the 1st data set : 655158 100.0 I tried something with textscan function but i haven't reached the results.Thank you!

[Bob Thompson]

So you want one pair of numbers for each block of data? Or you want to have a set of data which contains the reference number, and then the first column from the block of data?

Ultimately, you might try using a for loop to run through each line, and then use the fgetl() command to draw each line as a value to be evaluated.

for I = 1:end
line = fgetl(openfile)
if line == something
data(I,1) = reference
elseif line == somethingelse
data(I,2) = numbers
end
end

I find that this is usually the best way for me to work through text files with mixed numbers and strings, but I'm sure somebody on here has a better method.

[irenne11]

I would like to have a pair of numbers for each block of data. I used fgetl, but after that i used strncmp to take the index of the text "ref", inside the second row, and somehow doesn't find. Maybe i don't use correctly this function:

while ~feof(fid)
  pos = ftell(fid);
  str = strtrim(fgetl(fid));
  if strncmp(str,'ref f',5)
    fseek(fid,pos,'bof');
    break
  else
    hdr{end+1} = str;
  end
end

Answer 1

[Jeff Miller]

This function loads all of the numbers into one big long list. You can then grab out the numbers you want if you know their sequential positions within the list (e.g., you want the 4th, 8th, 104th, 108th, etc numbers):

function outNums = GetNumsInAsciiFile(sFName)
% Extract the numbers from any ASCII file into a vector, ignoring the text.

% Load the whole file into an array of char:
fid = fopen(sFName, 'r');
c1 = fscanf(fid,'%c');
fclose(fid);

% Split the file into tokens delimited by white space:
Tokens = strsplit(c1);

% Convert the tokens into numbers (returns NaN for any token that is not a number):
outNums = str2double(Tokens);

% Get rid of the NaN's:
outNums(isnan(outNums)) = [];

end

[irenne11]

Thanks for the answer. It succeeded to extract all the data. But it still have to extract the data after "ref f".

Answer 2

[Image Analyst]

That looks like such a variable format that I don't think you can use a built-in function like importdata(), etc. You'll have to write your own custom reader from lower level functions like fgetl(), sscanf(), textscan(), etc. It's too much work for any of us to do for you (it's no 3 minute job!), so good luck.