threads / patch / 62247

patcharchive: fix misleading error message

Subject: [PATCH] archive: fix misleading error message

## tl;dr

2 messages between Oct 3, 2024 and Oct 4, 2024. Diffs are folded; open one to read it.

replies: 1people: 2as markdown or json

René Scharfe· Oct 3, 2024, 15:51 UTC · lore

The error message added by 296743a7ca (archive: load index before pathspec checks, 2024-09-21) is misleading: unpack_trees() is not touching the working tree at all here, but just loading a tree into the index. Correct it.

Signed-off-by: René Scharfe <l.s.r@web.de>
---
Silly me, I stole the wrong message. :-/

 archive.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Show changes to archive.c +2 −2

diff --git a/archive.c b/archive.c
index 58f86bf75c..a7a92ff839 100644
--- a/archive.c
+++ b/archive.c
@@ -536,7 +536,8 @@ static void parse_treeish_arg(const char **argv,
 		opts.fn = oneway_merge;
 		init_tree_desc(&t, &tree->object.oid, tree->buffer, tree->size);
 		if (unpack_trees(1, &t, &opts))
-			die(_("unable to checkout working tree"));
+			die(_("failed to unpack tree object %s"),
+			    oid_to_hex(&tree->object.oid));

 		git_attr_set_direction(GIT_ATTR_INDEX);
 	}
--
2.46.2

Martin Ågren· Oct 4, 2024, 10:07 UTC · re: René Scharfe · lore

Re: [PATCH] archive: fix misleading error message

On Thu, 3 Oct 2024 at 17:51, René Scharfe <l.s.r@web.de> wrote:

Show 5 quoted lines

>
> The error message added by 296743a7ca (archive: load index before
> pathspec checks, 2024-09-21) is misleading: unpack_trees() is not
> touching the working tree at all here, but just loading a tree into
> the index.  Correct it.

Show 5 quoted lines

>                 init_tree_desc(&t, &tree->object.oid, tree->buffer, tree->size);
>                 if (unpack_trees(1, &t, &opts))
> -                       die(_("unable to checkout working tree"));
> +                       die(_("failed to unpack tree object %s"),
> +                           oid_to_hex(&tree->object.oid));

The existing instance of this message isn't translated, so it'll still be a new string to translate. Anyway, it's good to use the same wording here as elsewhere.

There's also a "failed to unpack trees", but here, we're indeed unpacking *one* tree.

Good catch!

Martin