[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: emacsclient's option decoding code

From: Juanma Barranquero
Subject: Re: emacsclient's option decoding code
Date: Tue, 11 Nov 2008 00:33:16 +0100

On Mon, Nov 10, 2008 at 19:36, Chong Yidong <address@hidden> wrote:

>>  - (make-frame-on-display nil ...) is a synonym of (make-frame ...),
>> but only on Windows; I proposed patch (2). Does anyone see a problem
>> with it?
> These two patches look OK to me.

Well, it's worse than that, because with your change to `make-frame-on-display',

   (make-frame-on-display "")

works on Windows; in fact, that's what
`server-create-window-system-frame' uses, because on Windows,

  (frame-parameter nil 'display)  => ""

With my patch, "" would not be accepted, because it is obviously not
null. We'd have to use

        ((or (null display) (eq system-type 'windows-nt))
         (make-frame parameters))

which is progressively uglier.

All in all, I think accepting anything as DISPLAY for Windows is a bad
choice. If we're going that route, it'd be cleaner to accept only nil
as default display, and fix the callers to use nil for "" on Windows.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]