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

Re: [PATCH v4 2/4] receive-pack: Clean dead code from update_worktree()

From
Johannes Schindelin <johannes.schindelin@gmx.de>
Date
Nov 9, 2021, 16:16 UTC
Message-ID
<nycvar.QRO.7.76.6.2111091711450.54@tvgsbejvaqbjf.bet>
In-Reply-To
<20211109030028.2196416-2-andersk@mit.edu>
Hi Anders,
looks good, just one suggestion, see inlined.
On Mon, 8 Nov 2021, Anders Kaseorg wrote:
Show 27 quoted lines
> update_worktree() can only be called with a non-NULL worktree parameter,
> because that’s the only case where we set do_update_worktree = 1.
> worktree->path is always initialized to non-NULL.
>
> Signed-off-by: Anders Kaseorg <andersk@mit.edu>
> ---
>  builtin/receive-pack.c | 20 +++++---------------
>  1 file changed, 5 insertions(+), 15 deletions(-)
>
> diff --git a/builtin/receive-pack.c b/builtin/receive-pack.c
> index 49b846d960..cf575280fc 100644
> --- a/builtin/receive-pack.c
> +++ b/builtin/receive-pack.c
> @@ -1449,29 +1449,19 @@ static const char *push_to_checkout(unsigned char *hash,
>
>  static const char *update_worktree(unsigned char *sha1, const struct worktree *worktree)
>  {
> -	const char *retval, *work_tree, *git_dir = NULL;
> +	const char *retval, *git_dir;
>  	struct strvec env = STRVEC_INIT;
>
> -	if (worktree && worktree->path)
> -		work_tree = worktree->path;
> -	else if (git_work_tree_cfg)
> -		work_tree = git_work_tree_cfg;
> -	else
> -		work_tree = "..";

We might want to make sure that `worktree` and `worktree->path` are non-`NULL`, and otherwise call a `BUG()`.

> -
>  	if (is_bare_repository())

Okay, I lied, I have two suggestions. Shouldn't this be turned into `worktree->is_bare`?

Of course, `find_shared_symref()` will currently not return any worktree with non-zero `is_bare`...

Show 24 quoted lines
>  		return "denyCurrentBranch = updateInstead needs a worktree";
> -	if (worktree)
> -		git_dir = get_worktree_git_dir(worktree);
> -	if (!git_dir)
> -		git_dir = get_git_dir();
> +	git_dir = get_worktree_git_dir(worktree);
>
>  	strvec_pushf(&env, "GIT_DIR=%s", absolute_path(git_dir));
>
>  	if (!hook_exists(push_to_checkout_hook))
> -		retval = push_to_deploy(sha1, &env, work_tree);
> +		retval = push_to_deploy(sha1, &env, worktree->path);
>  	else
> -		retval = push_to_checkout(sha1, &env, work_tree);
> +		retval = push_to_checkout(sha1, &env, worktree->path);
>
>  	strvec_clear(&env);
>  	return retval;
> @@ -1579,7 +1569,7 @@ static const char *update(struct command *cmd, struct shallow_info *si)
>  	}
>
>  	if (do_update_worktree) {
> -		ret = update_worktree(new_oid->hash, find_shared_symref("HEAD", name));
> +		ret = update_worktree(new_oid->hash, worktree);

Makes sense. If `worktree` is `NULL`, `do_update_worktree` won't ever be set, and if `worktree` is not `NULL`, even before we remove that `is_bare_repository()` ternary, it is set to `find_shared_symref("HEAD", name)` (and `name` is not changed in the `update()` function).

Thanks, Dscho

Show 7 quoted lines
>  		if (ret)
>  			return ret;
>  	}
> --
> 2.33.1
>
>
Previous: Anders KaseorgNext: Anders Kaseorg
Message 8 of 34 in “receive-pack: Protect current branch for bare repository worktree”
  1. 2/2 receive-pack: Protect current branch for bare repository worktreeAnders Kaseorg, Nov 8, 2021
  2. Junio C HamanoNov 8, 2021
  3. Junio C HamanoNov 9, 2021
  4. Johannes SchindelinNov 9, 2021
  5. Anders KaseorgNov 9, 2021
  6. 1/4 fetch: Protect branches checked out in all worktreesAnders Kaseorg, Nov 9, 2021
  7. 2/4 receive-pack: Clean dead code from update_worktree()Anders Kaseorg, Nov 9, 2021
  8. Johannes SchindelinNov 9, 2021
  9. Anders KaseorgNov 9, 2021
  10. 3/4 receive-pack: Protect current branch for bare repository worktreeAnders Kaseorg, Nov 9, 2021
  11. Johannes SchindelinNov 9, 2021
  12. Anders KaseorgNov 9, 2021
  13. 1/4 fetch: Protect branches checked out in all worktreesAnders Kaseorg, Nov 9, 2021
  14. 3/4 receive-pack: Protect current branch for bare repository worktreeAnders Kaseorg, Nov 9, 2021
  15. Ævar Arnfjörð BjarmasonNov 10, 2021
  16. 4/4 branch: Protect branches checked out in all worktreesAnders Kaseorg, Nov 9, 2021
  17. Ævar Arnfjörð BjarmasonNov 10, 2021
  18. 2/4 receive-pack: Clean dead code from update_worktree()Anders Kaseorg, Nov 9, 2021
  19. Ævar Arnfjörð BjarmasonNov 10, 2021
  20. Johannes SchindelinNov 10, 2021
  21. Ævar Arnfjörð BjarmasonNov 10, 2021
  22. Johannes SchindelinNov 10, 2021
  23. Junio C HamanoNov 10, 2021
  24. Junio C HamanoNov 11, 2021
  25. Junio C HamanoNov 10, 2021
  26. Anders KaseorgNov 10, 2021
  27. 4/4 branch: Protect branches checked out in all worktreesAnders Kaseorg, Nov 9, 2021
  28. Johannes SchindelinNov 9, 2021
  29. Johannes SchindelinNov 9, 2021
  30. Anders KaseorgNov 9, 2021
  31. Junio C HamanoNov 9, 2021
  32. Junio C HamanoNov 9, 2021
  33. Anders KaseorgNov 9, 2021
  34. Johannes SchindelinNov 9, 2021

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.