2009/7/9 Manoj Srivastava <[email protected]>:
> Hi,
>
> This was reported by a Debian user. Please retain
> [email protected] in your responses, so that the Debian
> BTS may have a copy of your response.
>
> ICCCM specifies WM_NAME, as an uninterpreted string, in the
> latin-1 domain, I think, EWMH specifies _NET_WM_NAME as utf-8; also
> Note that freedesktop.org WM spec already defines
> _NET_WM_ICON_NAME
> _NET_WM_NAME
> both of type UTF8_STRING, to be used in preference to WM_ICON_NAME and
> WM_NAME. (http://freedesktop.org/Standards/wm-spec/1.3/ar01s05.html
> and http://standards.freedesktop.org/wm-spec/wm-spec-1.3.html#id2506989)
But _NET_WM_NAME is used via ewmh_name.c:EWMH_WMName() -- this gets
called from add_window.c:setup_window_name().
The EWMH_WMName() function uses FiconvUtf8ToCharset() which is our own
wrapper around iconv to convert the given UTF8 string to the current
charset the system is set to -- be it Latin-1 or indeed UTF8.
Has this been verified with FVWM 2.5.28 (CVS, basically)? The test
case works fine here:
[n6ta...@shuttle][~]% xprop -id 0x4200008 | grep NET_WM_NAME | \
perl -ne 'while (/(0x[0-9a-f]{2})/g) { printf "%c", hex $1 }'
urxvt%
-- Thomas Adam
