On 23 Jul 2025, at 13.10, Eli Zaretskii <eliz@gnu.org> wrote:

From: Bror Winther <bbw@nobad.coffee>
Date: Tue, 22 Jul 2025 20:01:36 +0200
Cc: 79058@debbugs.gnu.org

How does that work when the client frame is a tty? Not the sense of emacs in the terminal but
using --eval.
I’m asking to understand why we waiting in exactly that scenario is fitting.

Sorry, I don't understand the question. Even if the client frame is a
tty, it is the Emacs server that displays there, not emacsclient.
emacsclient doesn't display anything on that frame, it just sends the
command to the server.

Let me see if I understand correctly and thank you for your patience with me.
I was under the impression that using --eval would not create a client frame
so your statement confuses me a bit.

No, --eval doesn't disable creation of a client frame. Try this:

$ emacsclient -t --eval '(message "foo")'

This creates a client frame on the current terminal, and shows the
message "foo" there.

Right, sorry for the confusion with a -t og -c option that makes sense but with

just --eval alone I don’t think the server should wait. Would you be open to a

solution that ignores the wait if the caller client does not have a child frame.

I have not researched if it is possible but would like to know if it is even worth

spending time on it.

I can’t wrap my head around the necessity of the wait in the case of --eval.
When you call --eval and encounter an error there is no output in the terminal
during the wait time. So let me rephrase my question, what are we waiting for?

The expression inside --eval could signal an error. That error is
shown on the client frame. If we delete the client frame immediately,
the error message will only be shown on the stderr stream of
emacsclient, and if that is not visible, the user won't have any
opportunity of seeing the message. So we wait for the user to see the
message on the client frame before we delete it.

Thank you for the clarification.

On 23 Jul 2025, at 13.17, Eli Zaretskii <eliz@gnu.org> wrote:

From: Bror Winther <bbw@nobad.coffee>
Date: Tue, 22 Jul 2025 20:06:15 +0200
Cc: 79058@debbugs.gnu.org

Sorry I can see I used some confusing terms here. I wrote client frame in tty. I have since
found that there is no client frame and thus I understand why you didn’t understand the
question. I’m sorely concerned with using the --eval option only, not in in combination with
-c or any other option that would create a client frame. Apologies for the confusion.

OK, but then your situation is very special, whereas the wait targets
much more general use cases: it makes sure we give the user ample
opportunity to see any error messages before they disappear from the
screen.