[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [bug-ncurses] ANN: ncurses-5.9-20150110
From: |
Thomas Dickey |
Subject: |
Re: [bug-ncurses] ANN: ncurses-5.9-20150110 |
Date: |
Mon, 12 Jan 2015 05:25:41 -0500 |
User-agent: |
Mutt/1.5.20 (2009-06-14) |
On Mon, Jan 12, 2015 at 10:51:44AM +0100, Dr. Werner Fink wrote:
> On Mon, Jan 12, 2015 at 10:48:26AM +0100, Werner Fink wrote:
> > On Sat, Jan 10, 2015 at 08:42:41PM -0500, Thomas Dickey wrote:
> > >
> > > 20150110
> > > + add a step to generating ".map" files, to declare any remaining
> > > symbols beginning with "_" as local, at the last version node.
...
> > with
> > --with-versioned-syms
> > --with-hashed-db
> > I see
> >
> > [ 25s] ../lib/libtic.so: undefined reference to `_nc_db_put'
> > [ 25s] collect2: error: ld returned 1 exit status
> > [ 25s] Makefile:275: recipe for target 'infocmp' failed
> > [ 25s] make: *** [infocmp] Error 1
> > [ 25s] make: *** Waiting for unfinished jobs....
> >
> > and indeed the symbols
> >
> > _nc_db_get;
> > _nc_db_have_index;
> > _nc_db_put;
> > _nc_hashed_db;
> >
> > are local
>
> without --with-hashed-db see
thanks (it looks like I didn't factor in the libtic vs the hashed-db,
and had overlooked weak symbols). The goal of course is to categorize
the symbols which might be used in some configuration and suppress the
others...
I might be able to complete a build for this, this evening
--
Thomas E. Dickey <address@hidden>
http://invisible-island.net
ftp://invisible-island.net
signature.asc
Description: Digital signature