>>>>> "fb" == Frank Berthold <[EMAIL PROTECTED]> writes:
fb> Two windows popped up, where you'd expect to see text instead
fb> there were squares of the type that seem common when the system
fb> doesn't recognize the characters it's being fed.
I don't know why CMUCL isn't able to connect to the motifd over a Unix
socket. Maybe you could try tracing the system calls that it makes to
see which error is being received. Check that the unix socket that
CMUCL tries to connect to is the same as the motifd prints to stdout.
As for the squares instead of characters, I think that you have
another font problem. On my system, the motifd tries to open fonts named
"-adobe-helvetica-medium-o-normal--*-120-75-*"
"-adobe-helvetica-medium-r-normal--*-120-75-*"
"-adobe-helvetica-bold-r-normal--*-120-75-*"
"fixed"
You can set the first three using the variables ENTRY-FONT-NAME,
ITALIC-FONT-NAME and HEADER-FONT-NAME in the interface package.
You can trace the font-related X11 traffic corresponding to a given
application using xmond. From a terminal window, run
% xmonui | xmond
then select the font-related calls in the middle-left "Selected
Requests -> Detail" slider. In another terminal window, start the
motifd as previously, then start CMUCL on X11 display number 1:
% motifd -nounix &
% DISPLAY=localhost:1 cmucl
then tell CMUCL where the motifd is running as previously. When the
graphical inspector starts up, you should see output from xmond:
............REQUEST: OpenFont
sequence number: 0002
request length: 000e
font-id: FNT 01c00002
length of name: 002c
name: "-adobe-helvetica-medium-o-normal--*-120-75-*"
............REQUEST: OpenFont
sequence number: 0003
request length: 000e
font-id: FNT 01c00003
length of name: 002c
name: "-adobe-helvetica-medium-r-normal--*-120-75-*"
Extended request: opcode 152, sequence number 000f
Unexpected reply, sequence number: 000f. No expected replies.
............REQUEST: OpenFont
sequence number: 0013
request length: 0005
font-id: FNT 01e00001
length of name: 0006
name: "cursor"
Presumably if these calls fail you will see something different.
--
Eric Marsden <URL:http://www.laas.fr/~emarsden/>
