Complete.Org: Mailing Lists: Archives: freeciv-dev: August 2000:
[Freeciv-Dev] Re: [patch] Startup readline segfault (PR#491)
Home

[Freeciv-Dev] Re: [patch] Startup readline segfault (PR#491)

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: freeciv-dev@xxxxxxxxxxx, bugs@xxxxxxxxxxxxxxxxxxx
Cc: jacob@xxxxxxxxxxxxxxxxx
Subject: [Freeciv-Dev] Re: [patch] Startup readline segfault (PR#491)
From: David Pfitzner <dwp@xxxxxxxxxxxxxx>
Date: Thu, 10 Aug 2000 16:39:32 +1000 (EST)

On Sun, 6 Aug 2000 David Pfitzner <dwp@xxxxxxxxxxxxxx> wrote:

> Jacob Luna Lundberg <jacob@xxxxxxxxxxxxxxxxx> wrote:

> > Actually, the attached configure.in patch detects the problem and turns
> > off readline support when it won't work (unless you override with
> > --with-readline or you're cross-compiling).  It's a bit unclean because it
> > uses AC_TRY_RUN() but that seemed unavoidable to me.

Ok, I think I solved my problems with this - revised patch attached.

- Moved bulk of test to acinclude.m4
- Removed ncurses references except in comments.
- Terminal being left in bad state was apparently due to missing 
  rl_callback_handler_remove().  (Only needed in some cases, and
  Freeciv itself gets away without this, but the standalone test
  needs it.)
- The "wrong" result on my system was confusion on my part: I had 
  changed the sense of the test ("works" vs "bad") but forgot to 
  change the sense of the result message!

I haven't actually tested this on a faulty system, but hopefully
it works!

-- David

Attachment: check_rl_runtime.diff.gz
Description: GNU Zip compressed data


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