Re: [PATCH 0/2] archive, apply: do not truncate files of 4GiB or more
- From
Patrick Steinhardt <ps@pks.im>
- Date
- Nov 19, 2026, 09:53 UTC
- Message-ID
- <av7HMbnylHiz_MVz@pks.im>
- In-Reply-To
- <20261119065711.827536-1-gitter.spiros@gmail.com>
On Thu, Nov 19, 2026 at 07:57:11AM +0100, Elia Pinto wrote:
Show 25 quoted lines
> Hi, > > A gentle ping on this two-patch series, which has had no comments so > far. It still applies cleanly to current master (e1ec09b23c) and > merges cleanly with 'seen'. > > 1/2 (archive) stands on its own. 2/2 (apply) has a question I should > have raised in the cover letter. It fixes the file written to the > working tree, but with --index, and so with "git am", the result still > goes through odb_write_object(), whose length is an "unsigned long" > until "use size_t for a few more object and buffer sizes" [1] lands. > Where "long" has 32 bits while size_t has 64, the blob written to the > index is then still truncated, while the working tree file is now > complete. > > I see two ways to handle that: > > - pass the size through cast_size_t_to_ulong() in add_index_file(), > so that "git apply --index" dies instead, as 5/5 of the loose object > series [2] does for notes, and drop the cast once [1] is in; > > - hold back 2/2 until [1] lands, as Patrick suggested for a similar > case in the pack-receive series [3]. > > Which one would you prefer?
I'd prefer the latter. So how about you send a new, rebased version of this series once it has landed?
Patrick