I do not see the general difference between scanning a file line by line or in blocks of 10 lines.
fid = fopen(FileName, 'r');
if fid == -1, error('Cannot open file: %s', FileName);
k = 0;
C = {};
D = cell(1, 10);
proceed = true;
while 1
for m = 1:10
s = fgetl(fid);
if ~ischar(s)
proceed = false;
D = D(1:m);
break;
end
D{m} = s;
end
k = k + 1;
C{k} = D;
end
Instead of storing D in C, you can parse it how ever you like.
This wooden technique is efficient, when you do not have to keep all lines in the memory at once (omit everything concerning "C" then). textscan looks much nicer, but is not necessarily faster.
