Split text file to smaller files

I have a text file build up like this:

$header1
a b c d 
e f g h
i j k l
$header2
1 b 2 a
3 c 4 d
..

I would like to split the file into smaller files where each file contains the information from one '$' to the next '$'. How would I go about doing this? Each file does not have a fixed amount of lines btw.

Answer 1

[Nicolai NoName]

Thanks for your answers. I ended up getting help from a friend and the result we ended up with was:

data=['C:\somewhere\somefile.txt];
FID=fopen(data,'rt');

lumberoflines = 100000;
for i=1:lumberoflines
    l = fgetl(FID);
    if strcmp(l(1),'$')
        target = l(2:end);
        l = fgetl(FID);
        exist FID2;
        if true(ans)
            fclose(FID2)
        end
        FID2 = fopen(target,'w');
    end
    fprintf(FID2,[l '\n']);
end
close all

This solution saved the individual files with names corresponding to the header.

[Guillaume]

Glad you've got something that works for you. However, that is some badly cobbled together code. In particular, the lines

exist FID2
if true(ans)

should be replaced by

if exist('FID2')

but the whole concept of relying on the existence of a variable to know if a file is open is very iffy. Similarly, using a magic constant for the number of lines is just asking for trouble.

Also, since you're opening the file for reading the file using 'rt', you should open the destination files using 'wt'. Otherwise, you may end up with different line endings (which may or may not be a problem depending on which program you then use with the files)

See my edited answer for a cleaner and faster way to do exactly what you want.

[MANISH R]

@Nicolai NoName If there are 2 $ symbols instead of 1 will this code work? I have tried it with 2 $ but error comes as Unrecognized function or variable 'Fid2' in line fprintf(Fid2,[l '\n']);

Answer 2

[Guillaume]

Personnally, I wouldn't bother with parsing the file, just read it in one go, then split it at the $:

wholefile = fileread('C:\somewhere\somefile.txt');
splitfiles = regexp(wholefile, '$[^$]+', 'match');
destnames = regexp(splitfiles, '(?<=$)[^\r\n]+', 'match', 'once');
for fileidx = 1:numel(splitfiles)
   fid = fopen(fullfile('C:\somewhere', destnames{fileidx}), 'w');
   fwrite(fid, splitfiles{fileidx});
   fclose(fid);
end

edited to extract destination name from $ expression

Answer 3

[KSSV]

fid = fopen('data.txt','r') ;
S = textscan(fid,'%s','delimiter','\n') ;
fclose(fid) ; 
S = S{1} ;
% find $ location 
idx = find(contains(S,'$')) ;
N = length(idx) ;
iwant = cell(N,1) ;
for i = 1:N-1
    iwant{i} = S(idx(i)+1:idx(2)-1) ;
end
iwant{N} = S(idx(N)+1:end) ;