> > Depending on who wrote the code, they may have extracted the rows > as hashrefs rather than arrays; that can be a 10x slowdown right > there. [I have no idea why so many people are so addicted to storing > rows in hashes, but it is always a significant slowdown; and > array slices are no more complicated than hash slices!]
I have not done perl code for a while now, but most perl coders, already suffering from a complex of coding in an unreadable language do not prefer to make it worse by using array which is position dependent and hence reading it can be a nightmare when large number of cols are selected. Also isn't array_ref even better than array, since it avoids copying the data to your local array in the code.
