[PATCH] btrfs-progs: snapshot fix user message

[PATCH] btrfs-progs: snapshot fix user message

[Anand Jain]

The fstests depend on the output message of the create snapshot command,
and if it's changed, the tests fail. Bring back the original messages,
as they are also grammatically correct.

Fixes: 5f87b467a9e7 ("btrfs-progs: subvolume: output the prompt line only when the ioctl succeeded")
Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 cmds/subvolume.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/cmds/subvolume.c b/cmds/subvolume.c
index 11bb5f560ad5..6516ea981d4d 100644
--- a/cmds/subvolume.c
+++ b/cmds/subvolume.c
@@ -779,11 +779,11 @@ static int cmd_subvolume_snapshot(const struct cmd_struct *cmd, int argc, char *
 
 	if (readonly)
 		pr_verbose(LOG_DEFAULT,
-			   "Create readonly snapshot of '%s' in '%s/%s'\n",
+			   "Create a readonly snapshot of '%s' in '%s/%s'\n",
 			   subvol, dstdir, newname);
 	else
 		pr_verbose(LOG_DEFAULT,
-			   "Create snapshot of '%s' in '%s/%s'\n",
+			   "Create a snapshot of '%s' in '%s/%s'\n",
 			   subvol, dstdir, newname);
 
 out:
-- 
2.39.3

Re: [PATCH] btrfs-progs: snapshot fix user message

[David Sterba]

On Tue, Mar 19, 2024 at 07:41:29PM +0530, Anand Jain wrote:
> The fstests depend on the output message of the create snapshot command,
> and if it's changed, the tests fail. Bring back the original messages,
> as they are also grammatically correct.

I removed the 'a' for consistency with other messages. That fstests
depend too much on what applications return is not a new thing and the
breakage will happen once the user-facing error messages get updated as
they get the priority.

Re: [PATCH] btrfs-progs: snapshot fix user message

[Anand Jain]

On 3/26/24 03:06, David Sterba wrote:
> On Tue, Mar 19, 2024 at 07:41:29PM +0530, Anand Jain wrote:
>> The fstests depend on the output message of the create snapshot command,
>> and if it's changed, the tests fail. Bring back the original messages,
>> as they are also grammatically correct.
> 
> I removed the 'a' for consistency with other messages. That fstests
> depend too much on what applications return is not a new thing and the
> breakage will happen once the user-facing error messages get updated as
> they get the priority.

  Alright, I wasn't aware it was an intentional change. We need to update
  the fstests test cases. I'll take care of it.

Thanks, Anand