Re: [PATCH/RFC 0/6] fetch with refspec
From: Stefan Beller <hidden>
Date: 2016-06-16 02:18:49
On Fri, Apr 15, 2016 at 12:19 PM, David Turner [off-list ref] wrote:
We've got a lot of refs, but pretty frequently we only want to fetch one. It's silly for the server to send a bunch of refs that the client is just going to ignore. Here are some patches that fix that. Let me know if this seems reasonable.
Thanks for working on this! I had a similar goal back then for non-http traffic and that series exploded in size[1] The issue at my attempt was non http traffic would require a protocol update such that the client speaks first to transport the refspec to the server. To make "client speaks first" happen, we'd need to have a protocol v2. So that attempt of mine stalled as it seemed like a huge thing. [1] WIP at https://github.com/stefanbeller/git/tree/protocol2-10 This series looks small and reasonable from a cursory read. Thanks, Stefan
(and I'll start in on another round of index-helper as soon as this is sent!) David Turner (6): http-backend: use argv_array functions remote-curl.c: fix variable shadowing http-backend: handle refspec argument transport: add refspec list parameters to functions fetch: pass refspec to http server clone: send refspec for single-branch clones Documentation/technical/protocol-capabilities.txt | 23 +++++++ builtin/clone.c | 16 ++++- builtin/fetch.c | 24 ++++++- builtin/ls-remote.c | 2 +- builtin/remote.c | 2 +- http-backend.c | 23 +++++-- remote-curl.c | 25 ++++--- t/t5552-http-fetch-branch.sh | 47 +++++++++++++ transport-helper.c | 44 ++++++++---- transport.c | 14 ++-- transport.h | 4 +- upload-pack.c | 81 ++++++++++++++++++++++- 12 files changed, 261 insertions(+), 44 deletions(-) create mode 100755 t/t5552-http-fetch-branch.sh -- 2.4.2.767.g62658d5-twtrsrc -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html