extraction of numeric data with presence of text data

I want to extract x,y co-ordinate and amplitude data from the txt file but i am not getting any idea how to do so.

for eg.:- * Gp# 1[8,4] x,y = -14.46, 0 dT[7] Src Amplitude = 36.0

then i want output as

x y amplitude

1. -14.46 0 36.0

2.

3. and so on

Answer 1

[dpb]

fid=fopen('abc.txt','r');
abc=textscan(fid,'%s',inf,'Delimiter','\n','HeaderLines',2,'commentstyle','1','texttype','string');
fid=fclose(fid);
abc=string(abc{:});
abc=abc(contains(abc,'Gp#'));
xy=str2double(split(extractBetween(abc,'x,y = ',' dT'),','));
ampl=str2double(extractAfter(abc,'Amplitude = '));
tABC=table(xy(:,1),xy(:,2),ampl,'VariableNames',{'x','y','amplitude'});

results in

>> head(tABC)
ans =
  8×3 table
      x        y      amplitude
    ______    ____    _________
    -14.46    0.00      36.00  
    -14.46    0.00      35.00  
     13.39    0.00      34.00  
    -18.75    0.00      32.00  
     19.82    0.00      36.00  
    -20.89    0.00      32.00  
     19.82    0.00      50.00  
     -5.89    0.00      36.00  
>> 

[Abhijit Sardar]

whenever i am running the program i am getting this error

"Error using contains

First argument must be a string array, character vector, or cell array of character vectors."

how can i solve it? My input file is the same.

[dpb]

Oh...missed copying a line from command window...insert

abc=string(abc{:});

before the contains statement...

[Abhijit Sardar]

also what if i want to extract data below Gp# ie. data in rows 2,4,6,8......

[dpb]

Then you'll have to read the entire file either piecewise w/ textscan between sections or line-by-line and parse each line for being data or the Gp# line and treat accordingly.

NB: you can't use the "trick" I did of treating the leading "1" in data lines as a comment to reduce the file size on input only looking for the other strings to do that.

Probably just readlines to start with the whole file as a string array then contains to find/extract the locations of the headers and processing between them in a loop will be as good as any.

[Abhijit Sardar]

I will try don't know if i can do it since i am very new to matlab but really thanks for the help till this point.

[dpb]

Well, give it a go and post your code and specific Q? if you get stuck...