[Xcb] [patch] bugfix: xproto.xml request GetKeyboardMapping

[Jamey Sharp]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 11/30/07, Barton C Massey <bart at cs.pdx.edu> wrote:
> (I just want to say for the record that you could have
> done it from the M4 also, though. :-) :-)

Yup. If these representations aren't all automatically
interconvertible, then we've done something wrong. :-)

> I think the encoding given is either right, or only
> off-by-one.  The length of the reply in words should be
> equal to the number of keysyms in the reply, since keysyms
> are 32 bits long---but it looks like the description as
> given doesn't take the extra word for keysyms-per-keycode
> into account.

The length field in an X reply counts the number of words
that *follow* the 32-byte fixed portion. (Example: unlike
requests, the length of a reply can be zero, and often is.)
The keysyms-per-keycode field is part of the leading 32-byte
portion, so is not counted in the length field.

Jamey
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHUQ7gp1aplQ4I9mURAhgXAJ9mg/YdwpPYPLpS6xgOs8PLHfPbEQCeMu67
+Gt8Rc7x/D1JcZHtvZp+i4E=
=v9j+
-----END PGP SIGNATURE-----