# [PATCH] archive: release refname after use

2 messages from 2020-11-14 to 2020-11-17. Participants: René Scharfe, Jeff King.
Thread: https://gitlist.dev/t/54646

## René Scharfe, 2020-11-14 22:01

Subject: [PATCH] archive: release refname after use
Message-ID: <7692d2f3-0a3b-91cd-c454-9725b0a395a7@web.de>
URL: https://gitlist.dev/e/7692d2f3-0a3b-91cd-c454-9725b0a395a7%40web.de

```
parse_treeish_arg() uses dwim_ref() to set refname to a strdup'd string.
Release it after use.  Also remove the const qualifier from the refname
member to signify that ownership of the string is handed to the struct,
leaving cleanup duty with the caller of parse_treeish_arg(), thus
avoiding a cast.

Signed-off-by: René Scharfe <l.s.r@web.de>
---
 archive.c | 1 +
 archive.h | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/archive.c b/archive.c
index 3c1541af9e..da89d292c2 100644
--- a/archive.c
+++ b/archive.c
@@ -662,6 +662,7 @@ int write_archive(int argc, const char **argv, const char *prefix,
 	rc = ar->write_archive(ar, &args);

 	string_list_clear_func(&args.extra_files, extra_file_info_clear);
+	free(args.refname);

 	return rc;
 }
diff --git a/archive.h b/archive.h
index 82b226011a..43028ffbf3 100644
--- a/archive.h
+++ b/archive.h
@@ -8,7 +8,7 @@ struct repository;

 struct archiver_args {
 	struct repository *repo;
-	const char *refname;
+	char *refname;
 	const char *prefix;
 	const char *base;
 	size_t baselen;
--
2.29.2

```

## Jeff King, 2020-11-17 00:28

Subject: Re: [PATCH] archive: release refname after use
Message-ID: <20201117002839.GB13516@coredump.intra.peff.net>
URL: https://gitlist.dev/e/20201117002839.GB13516%40coredump.intra.peff.net
In-Reply-To: <7692d2f3-0a3b-91cd-c454-9725b0a395a7@web.de>

```
On Sat, Nov 14, 2020 at 11:01:04PM +0100, René Scharfe wrote:

> parse_treeish_arg() uses dwim_ref() to set refname to a strdup'd string.
> Release it after use.  Also remove the const qualifier from the refname
> member to signify that ownership of the string is handed to the struct,
> leaving cleanup duty with the caller of parse_treeish_arg(), thus
> avoiding a cast.

Yeah, I looked at the surrounding code and agree that this is the right
thing to do.

I wondered if it might make  sense for the initialization and clearing
of this "archive_args" structure to go in a pair of matched functions,
but it probably isn't worth the trouble. write_archive() is the only
place where we will allocate and deallocate such a struct (the only
reason it is a type at all is to pass the set of values around the
stack).

-Peff

```