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

Re: [PATCH 3/5] describe: catch unborn branch in describe_blob()

From
Junio C Hamano <gitster@pobox.com>
Date
Aug 18, 2025, 21:19 UTC
Message-ID
<xmqqy0rggwep.fsf@gitster.g>
In-Reply-To
<20250818210154.GC1024556@coredump.intra.peff.net>
Jeff King <peff@peff.net> writes:
Show 8 quoted lines
> --- a/builtin/describe.c
> +++ b/builtin/describe.c
> @@ -518,10 +518,16 @@ static void describe_blob(const struct object_id *oid, struct strbuf *dst)
>  {
>  	struct rev_info revs;
>  	struct strvec args = STRVEC_INIT;
> +	struct object_id head_oid;
>  	struct process_commit_data pcd = { *null_oid(the_hash_algo), oid, dst, &revs};

An unrelated tangent, but it seems that we are copying the object name for the first member of this struct, even though 1/5 changed the second one.

> +	if (repo_get_oid(the_repository, "HEAD", &head_oid))
> +		die(_("cannot search for blob '%s' on an unborn branch"),
> +		    oid_to_hex(oid));

Makes sense. I briefly wondered if there is really a point in doing the traversal only from HEAD, but this topic is not about enhancing and making the "describe <blob>" more useful, but it is a strict improvement. When you first mentioned "resolve HEAD ourselves", I somehow expected you to ask the ref subsystem to resolve HEAD, but this should do fine, thanks to ref_rev_parse_rules[].

Show 19 quoted lines
> diff --git a/t/t6120-describe.sh b/t/t6120-describe.sh
> index 470631d17d..feec57bcbc 100755
> --- a/t/t6120-describe.sh
> +++ b/t/t6120-describe.sh
> @@ -415,6 +415,14 @@ test_expect_success 'describe an unreachable blob' '
>  	test_grep "blob .$blob. not reachable from HEAD" actual
>  '
>  
> +test_expect_success 'describe blob on an unborn branch' '
> +	oldbranch=$(git symbolic-ref HEAD) &&
> +	test_when_finished "git symbolic-ref HEAD $oldbranch" &&
> +	git symbolic-ref HEAD refs/heads/does-not-exist &&
> +	test_must_fail git describe test-blob 2>actual &&
> +	test_grep "cannot search .* on an unborn branch" actual
> +'
> +
>  test_expect_success ULIMIT_STACK_SIZE 'name-rev works in a deep repo' '
>  	i=1 &&
>  	while test $i -lt 8000
Previous: Jeff KingNext: Jeff King
Message 19 of 28 in “Potential Null Pointer Dereference detected by static analysis tool”
  1. ChengAug 13, 2025
  2. Phillip WoodAug 13, 2025
  3. Jeff KingAug 14, 2025
  4. Phillip WoodAug 15, 2025
  5. René ScharfeAug 17, 2025
  6. Jeff KingAug 18, 2025
  7. Jeff KingAug 18, 2025
  8. René ScharfeAug 18, 2025
  9. Jeff KingAug 18, 2025
  10. Jeff KingAug 18, 2025
  11. 0/5 fix segfault and other oddities describing blobsJeff King, Aug 18, 2025
  12. 1/5 describe: pass oid struct by const pointerJeff King, Aug 18, 2025
  13. Junio C HamanoAug 18, 2025
  14. 2/5 describe: error if blob not foundJeff King, Aug 18, 2025
  15. Junio C HamanoAug 18, 2025
  16. Patrick SteinhardtAug 19, 2025
  17. René ScharfeAug 19, 2025
  18. 3/5 describe: catch unborn branch in describe_blob()Jeff King, Aug 18, 2025
  19. Junio C HamanoAug 18, 2025
  20. Jeff KingAug 18, 2025
  21. 4/5 describe: handle blob traversal with no commitsJeff King, Aug 18, 2025
  22. Patrick SteinhardtAug 19, 2025
  23. Jeff KingAug 19, 2025
  24. Patrick SteinhardtAug 20, 2025
  25. 4/5 describe: handle blob traversal with no commitsJeff King, Aug 20, 2025
  26. 5/5 describe: pass commit to describe_commit()Jeff King, Aug 18, 2025
  27. Patrick SteinhardtAug 19, 2025
  28. Jeff KingAug 19, 2025

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.