git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH] builtin/repack.c: only collect fully-formed packs

From
Junio C Hamano <gitster@pobox.com>
Date
Jun 12, 2023, 21:00 UTC
Message-ID
<xmqq4jnc9z16.fsf@gitster.g>
In-Reply-To
<20230607101617.ges6tnMry4E52lDGld43QgtNUsIS4YQq6w-t71hEfkQ@z>
Taylor Blau <me@ttaylorr.com> writes:
> Fix the above by teaching `collect_pack_filenames()` to only collect
> packs with their corresponding `*.idx` files in place, indicating that
> those packs have been fully staged.

Hmph, after following the problem description (which by the way was very well written), I expected to see "collect_pack_filenames() is taught to use get_all_packs() and then enumerate the names from the in-core packed_git structures---this way there is no possibility for the two methods to produce inconsistent results".

But that is not what is going on. Rather, it is "we still keep using a separate readdir() loop, but made sure that the criteria to include a packfile in the result is identical to the logic currently used by the other function", whose implication is that they can diverge again. I am somewhat puzzled.

Will queue.  Thanks.
Message 1 of 1 in “Re: [PATCH] builtin/repack.c: only collect fully-formed packs”
  1. Junio C HamanoJun 12, 2023

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.