GNU bug report logs - #29156
25.3; eshell/kill does not understand -<signal>

Previous Next

Full log

Message #27 received at 29156 <at> debbugs.gnu.org (full text, mbox):

From: Noam Postavsky <npostavs <at> gmail.com>
To: 29156 <at> debbugs.gnu.org
Cc: Eric Skoglund <eric <at> pagefault.se>, Pierre Neidhardt <ambrevar <at> gmail.com>
Subject: Re: bug#29156: 25.3;
 eshell/kill does not understand -<signal>, [PATCH] Make eshell/kill
 handle -<signal> and -<SIGNALNAME>
Date: Sat, 17 Mar 2018 10:54:46 -0400

> From: Eric Skoglund <eric <at> pagefault.se>

>> More importantly, could you explain a bit how your change works/why it's
>> correct?
>
> I'm guessing you mean in the commit message and not here in the bug
> report right?

Either, both :)

(If you send a new patch with the explanation in the commit message,
it's automatically in the bug report as well anyway)

>> Not sure [about] this `make-symbol' call, should it rather be
>> `intern'?  (Maybe we should update signal-process take a string as
>> well a symbol.)
>
> Saw that Eli also agreed, for future reference when do we want to use
> `make-symbol` vs `intern`?

Usually, `make-symbol' is only for macros, where you want a symbol that
is not `eq' to any other.  Sometimes it's handy for making a unique
object at run-time.  In this case, since we only care about the symbol
name, it doesn't really matter, it's just a bit surprising to see it
because it's usually a mistake in a non-macro context.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.