Delete the rows with its first character other than a number and copy all the other rows into a new file.

I want to delete the rows with the first character other than a number and copy the other rows which has it's first as a number into a new file.

For ex . my file looks like this

Mach-1 asrt asrj ssrj . . . 0346 1346 2346 3334 4337 . . . aaerh baer daer frah . . . 1345 2346 3436 4346 . . . asdh bsth

I only wan the lines which starts with a number

[Michael Haderlein]

That sounds as if you want to solve this again, although it was extensively answered by Azzi Abdelmalek and me (<http://www.mathworks.de/matlabcentral/answers/143308-i-have-to-detect-the-startrow-and-endrow-from-a-file-automatically-from-a-txt-file)>. In case it's the same question, please respond in the other thread.

In case it's a different question because now your file has numbers and text mixed:

   fid=fopen(filename);
   curline=fgetl(fid);
   result=[];
   while ischar(curline)
      if ~isempty(curline) && ~any(isstrprop(curline,'alpha'))
         result=[result;str2num(curline)];
      end
      curline=fgetl(fid);
   end
   fclose(fid);

[Vinit]

 sir .... both the programs works perfectly for me.... i have   
one more problem which i will post it as another question.... Thank you for your time

Answer 1

[Azzi Abdelmalek]

fid = fopen('file.txt');
a=textscan(fid,'%s')
fclose(fid);
b=a{:};
c=b(cellfun(@(x) ~isempty(regexp(x,'^\d','match')),b))

To export the new data to a new file

fileID = fopen('file1.txt','w');
for k=1:numel(c)
  fprintf(fileID,'%s \n',c{k});
end
fclose(fileID);