Re: Windows binaries for emacs-28 status and TODOs

[H. Dieter Wilhelm]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: "H. Dieter Wilhelm" <dieter@duenenhof-wilhelm.de>
>> Cc: corwin@bru.st,  emacs-devel@gnu.org,  phillip.lord@russet.org.uk
>> Date: Wed, 16 Feb 2022 10:50:03 +0100
>> 
>> > They should have emacsclient, not Emacs.
>> 
>> Yes, I agree.
>> 
>> But then, the user has to know about emacsclient(w) AND how to set up
>> the server beforehand.
>
> No, the user should be able to select emacsclient, and the installer
> should produce the actual command using that.  Just like with the
> other editors they offer.
>
>> I believe it should also work for emacs.exe (which it is) and
>> runemacs.exe (which is not working, please see the attached image).
>
> Why is that a good idea?

To be prepared for all eventualities when a user might overwrite a
suggestion of emacsclient?  It took many years of Emacs usage before I
understood what emacsclient is.

>> Or, even better, suggesting a command line option for emacsclient(w)
>> in case the server is not running.
>
> That's what I had in mind.  (And no, emacsclientw is NOT the right
> choice here, since Git is a console executable.)

Ah right, just considered the installer test.

>> (Unfortunately I couldn't get emacsclient to
>> run with a command line option, probably a bug in the Git installer.:-/)
>
> Which seems to indicate that integrating emacsclient into Git is not
> as simple as its maintainer thinks.

Sigh, unfortunately yes. Asked the maintainer about the command line
issue, anyway.

And there's also GitGui to consider, which is not a console program
(regarding the choice of emacsclient and emacsclientw)..

Let's see, thank you

      Dieter
-- 
Best wishes
H. Dieter Wilhelm
Zwingenberg, Germany