Why is "findstr" not recommended compared to strfind.

Is it simply because you can flip the inputs in strfind however you want, (making it more flexible), or is it actually more resource intensive. I have findstr alot in my code and it works, but should I go back and edit it?

I don't know why it bothers to give me the warning "findstr is not recommended. try strfind instead"

Answer 1

[per isakson]

R2012a says: "Note: findstr will be removed in a future release. Use strfind instead." This is a reason to replace findstr in code, which will be around for some years in the future.

And strfind is somewhat faster

    n2  = 1e6;
    a   = 'a':'p';
    buf = a( randi( [1,16], [1,n2] ) );
    str = reshape( buf, 1, [] );
    pattern = a( randi( [1,16], [1,4] ) );
    tic, ix1 = strfind( str, pattern ); toc
    tic, ix2 = findstr( str, pattern ); toc
    all( ix1==ix2 )

returns

    Elapsed time is 0.001971 seconds.
    Elapsed time is 0.003049 seconds.
    ans =
         1