Re: [External] Re: [PATCH 1/1] revision: don't set parents as uninteresting if exclude promisor objects

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Aug 13, 2024 at 12:09 AM Junio C Hamano <gitster@xxxxxxxxx> wrote:
> Perhaps another simpler approach may be to use is_promisor_object()
> function and get rid of this initial marking of these objects in
> prepare_revision_walk() with the for_each_packed_object() loop,
> which abuses the UNINTERESTING bit.  The feature wants to exclude
> objects contained in these packs, but does not want to exclude
> objects that are referred to and outside of these packs, so
> UNINTERESTING bit whose natural behaviour is to propagate down the
> history is a very bad fit for it.  We may be able to lose a lot of
> existing code paths that say "if exclude_promisor_objects then do
> this", and filter objects out with "is_promisor_object()" at the
> output phase near get_revision().

I tried to go down this route. I removed the for_each_packed_object()
loop and filter promisor commits in get_revision_1() instead.
However, this only filtered promisor commits, not promisor trees and
objects. A combined approach would be keeping the
for_each_packed_object() loop, but only mark non-commit objects
as UNINTERESTING there, and filter promisor commits in
get_revision()?





[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux