Keyboard

Keyboard handling is an area that needs improving.
We currently use the core X11 api, but we should be moving to Xkb, or maybe the ​xkbcommon wrapper (#371).

Solved Bugs

• When connecting over high latency links, use the --no-keyboard-sync option to prevent keys from repeating. This switch may cause other problems though.
• US layout and most common layouts should work ok.
• If starting xpra from an environment which has non-standard input methods, this can interfere, see #286

Unsolvable Bugs

On Posix systems, if your distribution does not provide setxkbmap -query then you should really upgrade it, there are bugs which are simply unfixable when one uses outdated software.

Pending Bugs

• Supporting multiple layouts and switching layouts reliably and/or manually: #230, #166, #86
• Keyboard sync may end up pressing the wrong key: #222
• Input grabs: #139

Reporting Bugs

First, please check the bugs above to see if any match your problem or seems related. Failing that, make sure to read the reporting bugs guidelines and generally you will need to include (select those that apply):

• active keyboard layout(s)
• input methods
• keyboard related configuration setup/files
• keyboard type
• whether the bug is also present with/without --no-keyboard-sync
• setxkbmap -print and setxkbmap -query (both directly in the client if it supports those commands and in the xpra session)
• xmodmap -pke and xmodmap -pm (again on both)
• xkbprint -label name $DISPLAY
• xev output of the misbehaving key events
• the ​XPRA_KEYBOARD_DEBUG environment variable can increase logging verbosity (quite a bit with a value of "1" and a lot more with a value of "2")
• using the test script ​gtk_view_keyboard.py (named GTK_Keyboard_Test.exe on MS Windows) can also be useful sometimes