On Tue, Jul 13, 2021 at 5:47 PM Junio C Hamano <gitster@xxxxxxxxx> wrote: > > Elijah Newren <newren@xxxxxxxxx> writes: > > > On Mon, Jul 12, 2021 at 8:09 AM Derrick Stolee <stolee@xxxxxxxxx> wrote: > >> > >> Since the steps that find inexact renames without the quadratic > >> algorithm are heuristics, perhaps this portion could instead be > >> called "exhaustive rename detection" or even "expensive rename > >> detection"? It perhaps implies more directly that the limit exists > >> as a way to prevent an expensive operation. > > > > The name "exhaustive rename detection" seems reasonable to me. I'll > > resubmit using that term and see what folks think. > > Funny. In a sense, since computing content similarity is _more_ > precise way to find a pair that is a likely rename than the more > recent heuristics, the more expensive "exhaustive" one can be called > "more precise rename detection", even though it is tempting to use > "inexact" to call it, too ;-) I'm afraid I'm not following. The more recent heuristics use content similarity as well; in fact, they use the exact same function with a higher threshold: estimate_similarity(). The exhaustiveness of the quadratic portion comes from comparing each file to more other files, not in using a different type of comparison.
