Re: [PATCH 0/2] Skip ls-refs if possible for HTTP

From: Junio C Hamano <gitster@pobox.com>
Date: Aug 22, 2019, 16:27 UTC
Message-ID: <xmqqtva9p3ma.fsf@gitster-ct.c.googlers.com>
In-Reply-To: <cover.1566425828.git.jonathantanmy@google.com>

Jonathan Tan <jonathantanmy@google.com> writes:

Show 8 quoted lines

> This was discovered by one of my colleagues when using a partial clone.
> I thought I had resolved the problem with the commits mentioned in patch
> 1 (e70a3030e7 and ancestors), but apparently that is not the case (that
> only worked for native protocols). So here is a fix for HTTP.
>
> I'm not sure of the value of the test in patch 2, but that test does
> fail if I don't update fetch_refs_from_bundle() to first call
> get_refs_from_bundle() if it hasn't already been called.

This probably is totally off-tangent, but do any of these "let's advertise fewer" changes at the protocol level have to take into account the use of --prune option on the client side?

Show 10 quoted lines

> Jonathan Tan (2):
>   transport-helper: skip ls-refs if unnecessary
>   transport: teach all vtables to allow fetch first
>
>  t/t5607-clone-bundle.sh | 11 +++++++++++
>  t/t5702-protocol-v2.sh  | 13 +++++++++++++
>  transport-helper.c      | 38 ++++++++++++++++++++++++++++++++------
>  transport-internal.h    |  6 ------
>  transport.c             | 18 ++++++------------
>  5 files changed, 62 insertions(+), 24 deletions(-)