MATLAB: How to filter and arrange text in a file using textscan

textscan

Hello, I have a Log.txt file with data;
ITEM: TIMESTEP
1000
ITEM: NUMBER OF ATOMS
113
ITEM: BOX BOUNDS mm mm mm
-1.1 1.1
-1.00019 1.9
-0.75007 0.75007
id  type  x  y  z
1  1  0.450082  1.76687  -0.101338
2  1  -0.0949301  1.76202  -0.0205438
3  1  0.0403266  1.78975  0.0742209
4  1  0.195614  1.75958  0.0106906
5  1  0.227547  1.84632  -0.098714
1  1  0.144155  1.76642  0.108974
2  1  0.288044  1.85155  0.0387214
3  1  0.370195  1.81785  -0.10084
4  1  0.383223  1.78207  0.0387085
5  1  -0.285847  1.72996  -0.00240017
1  1  0.450142  1.77285  0.117999
2  1  -0.318942  1.74455  -0.0651567
3  1  -0.198368  1.76376  0.0788363
4  1  0.255091  1.74895  0.0369646
5  1  -0.128421  1.72431  -0.113898
1  1  0.486869  1.83805  0.0239884
2  1  0.0405222  1.80871  -0.10944
3  1  -0.359026  1.75984  -0.00066415
4  1  -0.436089  1.73765  -0.0688759
5  1  -0.15772  1.81532  -0.0857257
I wrote a code as given below to skip the first 8 lines and subsequently read the entire text and in the process leaving non-numbers.
fid = fopen('Log.txt','r');
%Advance 8 lines:
linesToSkip = 8;
for ii = 1:linesToSkip-1
    fgetl(fid);
end
%Process all remaining lines
tline =  fgetl(fid);
To_Excel_Data = []; %You should allocate if you know how large your data is
while ~feof(fid)
    tline = fgetl(fid);
    %Getting rid of non-numbers
    tline = regexprep(tline,'[^0-9\s+-.eE]','');
    To_Excel_Data = [To_Excel_Data; str2num(tline)];
end
fclose(fid);
The Output of the code is;
1  1  0.450082  1.76687  -0.101338
2  1  -0.0949301  1.76202  -0.0205438
3  1  0.0403266  1.78975  0.0742209
4  1  0.195614  1.75958  0.0106906
5  1  0.227547  1.84632  -0.098714
1  1  0.144155  1.76642  0.108974
2  1  0.288044  1.85155  0.0387214
3  1  0.370195  1.81785  -0.10084
4  1  0.383223  1.78207  0.0387085
5  1  -0.285847  1.72996  -0.00240017
1  1  0.450142  1.77285  0.117999
2  1  -0.318942  1.74455  -0.0651567
3  1  -0.198368  1.76376  0.0788363
4  1  0.255091  1.74895  0.0369646
5  1  -0.128421  1.72431  -0.113898
1  1  0.486869  1.83805  0.0239884
2  1  0.0405222  1.80871  -0.10944
3  1  -0.359026  1.75984  -0.00066415
4  1  -0.436089  1.73765  -0.0688759
5  1  -0.15772  1.81532  -0.0857257
Which is well and good. Now I am looking for help to extent the code with textscan (or an alternative) so that I can filter and group the data according to the identity of the first element in column 1 so that the output is;
1  1  0.450082  1.76687  -0.101338
1  1  0.144155  1.76642  0.108974
1  1  0.450142  1.77285  0.117999
1  1  0.486869  1.83805  0.0239884
        
        
2  1  -0.0949301  1.76202  -0.0205438
2  1  0.288044  1.85155  0.0387214
2  1  -0.318942  1.74455  -0.0651567
2  1  0.0405222  1.80871  -0.10944
        
        
3  1  0.0403266  1.78975  0.0742209
3  1  0.370195  1.81785  -0.10084
3  1  -0.198368  1.76376  0.0788363
3  1  -0.359026  1.75984  -0.00066415
        
        
4  1  0.195614  1.75958  0.0106906
4  1  0.383223  1.78207  0.0387085
4  1  0.255091  1.74895  0.0369646
4  1  -0.436089  1.73765  -0.0688759
        
        
5  1  0.227547  1.84632  -0.098714
5  1  -0.285847  1.72996  -0.00240017
5  1  -0.128421  1.72431  -0.113898
5  1  -0.15772  1.81532  -0.0857257
I look forward to your help. Thank you.

Best Answer

fid = fopen('LOG.txt','r') ;
S = textscan(fid,'%f','delimiter','\n','HeaderLines',9) ;
fclose(fid) ;
data = reshape(S{1},5,[])' ;
% seperate the data 
[c,ia,ib] = unique(data(:,1)) ;
for i = 1:length(c)
    idx = data(:,1)==c(i) ;
    iwant(:,:,i) = data(idx,:) ;
end
Related Question