[PATCH 0 of 5] RFC: Implement O(n + m) algorithm for findrenames()

[Benoit Boissinot]

On Mon, Mar 08, 2010 at 11:19:12AM +1100, David Greenaway wrote:
> On Sun, Mar 07, 2010 at 06:01:06PM +0100, Mads Kiilerich wrote:
> 
> > Would it make sense to make a first cheap scan only
> > looking for 100% match? Perhaps by using the cheap "len" as a coarse
> > hash value? How much improvement will this give in your example?

Ideally there are a bunch of other heuristics to apply (I don't remember
if git does that as well) like filename matching (file is moved from a
directory from another, check identical filenames first).

Benoit
-- 
:wq