On Wed, Jul 21, 2004 at 05:44:32AM -0400, Thomas Dickey wrote: > > which looks sane to me. Here's the (maybe) interesting part: When I > > press <ALT>, I see: > > > > Input keysym 0xFFE7, 0:'' 7bit > > Handle 7bit-key > > Is that <ALT> the same as one of your Meta_L or Alt_R keys? I mean the key labeled "alt" on the keyboard. On my Mac Keyboard: keycode 64, keysym 0xffe7. It has the modifier ALT_L assigned by default and I change that via: keysym Alt_L = Meta_L Alt_L which works according to xev. > xev could identify that. What your trace seems to indicate to me > is that the mod1 for Meta_L isn't having a real effect, so the literal > key is sent to xterm. That should show up in xev's trace (though xev > doesn't show the modifier information, it should show a "Alt_L" or "Meta_L"). When I do: xmodmap -e 'keycode 64 = Alt_L' (removing the Meta_L) *it works* again. So it seems it's actually not "Meta sends escape" anymore but rather "Alt sends escape". This is a change in behaviour which we should at least document somewhere in the debian package before closing the bug. I'm still having some problems with my <apple-key>-click = "middle click" emulation, which completely confuses xterm at the moment (and which showed up at the same time than the above problem), but I'll have to dig deeper into this before reporting this as a bug. Thanks _very_ much for your help, -- Guido
Attachment:
signature.asc
Description: Digital signature