Comment 7 for bug 7416

Message-ID: <email address hidden>
Date: Tue, 20 Jul 2004 05:38:21 -0400
From: Thomas Dickey <email address hidden>
To: Guido Guenther <email address hidden>
Cc: <email address hidden>
Subject: Re: Bug#260232: xterm: XTerm*metaSendsEscape no longer working

--dDRMvlgZJXvWKvBx
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Jul 20, 2004 at 09:13:59AM +0200, Guido Guenther wrote:
> On Mon, Jul 19, 2004 at 04:24:08PM -0400, Thomas Dickey wrote:
> > I don't think the problem is within xterm (it's been a while since I tw=
eaked
> > the logic for this). More likely something in the keyboard configurati=
on
> > has separated the definitions that you were relying upon. What I do to
> > debug this is to use xev and ensure that it's reporting "Meta_L" for the
> > key. If that's right, I generally compile a debug version of xterm
> > (configure --enable-trace) and check the initialization of the modifier=
s.
> Trace says about Meta:
> =20
> ~Meta<KeyPress>: insert-seven-bit()

Trace-parent.out has a lot of information - the fragment you're showing
tells what it can find about the translations resource. The piece I'm
interested in corresponds to the xmodmap settings, e.g., (starting with
"VTInitModifiers"):

looking for style 'Root'
VTInitModifiers
alt_left mask 0x8 is Mod1 modifier
alt_right mask 0x8 is Mod1 modifier
TranslationsUseKeyword(7ac10):#override

xterm looks to see which modifiers are associated with the Alt, Meta and
NumLock keys so that when it gets a keypress event, it can look at the
modifier information in that event and determine which of those keys
might have been used for modifying the key. (Technically, I should also
look for the control and shift keys, but it is unusual for someone to
change _those_ with xmodmap). In the example I just gave, there's no
Meta key associated with a modifier - so metaSendsEscape wouldn't work.

The trace is there of course for debugging - it's also possible that
xterm would find the modifiers but do something unexpected with them.
All of the logic dealing with the modifiers is in input.c, so it's not
hard to follow...

--=20
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net

--dDRMvlgZJXvWKvBx
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (SunOS)
Comment: For info see http://www.gnupg.org

iD8DBQFA/OT6tIqByHxlDocRAiCyAJ0ahC++QvduHOhrLsvlUn2kbxaSogCgn9Jy
L3SS0zhcqQ9RHjSlRbxCAGU=
=jh7i
-----END PGP SIGNATURE-----

--dDRMvlgZJXvWKvBx--