bug-gnubg
[Top][All Lists]
Advanced

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

Re: [Bug-gnubg] Compile: Windows and python


From: Jim Segrave
Subject: Re: [Bug-gnubg] Compile: Windows and python
Date: Mon, 18 Aug 2003 08:01:42 +0200
User-agent: Mutt/1.4.1i

On Sun 17 Aug 2003 (22:56 +0200), Holger wrote:
> At 22:04 15.08.03 +0200, Jim Segrave wrote:
> >On Fri 15 Aug 2003 (21:04 +0200), Holger wrote:
> >> At 23:21 14.08.2003 +0200, Nardy Pillards wrote:
> >> >Just a guess: change the order of the libs ?
> >> >
> >> >This is mine for the no-gui (yep, gtk not needed there, i know):
> >> >
> >> >LIBS     = $(LIBPATH) -levent -lxml2 -lintl.dll -lgdbm.dll \
> >> >                -lglib-2.0.dll -lgdk.dll -lgtk.dll -liconv -lpng -lz \
> >> >                -lreadline -lpython22 -lwsock32 -lwinmm -lm
> >> >
> >> >with lwinmm and lm as closing libraries.
> >> 
> >> No, I even tried, but it would've been very strange if the order of any 
> >> library should matter.
> >
> >Just as a point, it can matter. If two libraries define a global with
> >the same name, then the one linked in will depend on the linker and
> >the order in which the libraries are named. Generally, I believe most
> >linkers will use the first definition encountered (and I'll bet Posix
> >specifies what should happen)
> 
> But wouldn't this give a linker error or at least a warning?

It should not do so. It's a useful way to override some functions in a
library with custom ones without having to supply a complete
replacement library.

-- 
Jim Segrave           address@hidden





reply via email to

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