dsseewe@SUPER.ORG (Daniel Seewer) (11/28/89)
Hope someone can provide me with some answers for the following questions: 1) What are the "proper" procedures to follow when a client request a font using XLoadQueryFont() and it is not there? Is there a font that is always guaranteed to be there? Or is this acceptable behavior? 2) Where can I find the source for a client that is actually ICCCM compliant? I am in the process of reading the document about ICCCM an was curious about what the bare minimum requirements for a client to be "ICCCM compliant" would be. 3) What can the following problem be related to? I have a client that runs without any problems on SUN 3/** and every SUN 4 I have tried except for the SUN 4/330 (or it might be a 4/430 if there is such a machine). The server is one distributed by SUN with Openwindows and uses cg6 (or cg9). As you can tell I really know my hardware. Any way when I allocate the Pixmap there seems to be garbage in the pixmap. That is even if the pixmap is not drawn in and then displayed it will have stuff in the pximap. The confusing part is that it works on all color SUNS using the mit X11 server. When the code is recompiled on the SUN 4/330it has the previously mentioned problem. Could it be the background pixmap? I am using the athena widget set X11R3. Am I doing something wrong? I hope these questions can generated some comments that may lead to solutions! THanks in advance!!! Dan Seewer (dsseewe@super.org)
rws@EXPO.LCS.MIT.EDU (Bob Scheifler) (11/28/89)
What are the "proper" procedures to follow when a client
request a font using XLoadQueryFont() and it is not there?
Sounds pretty application dependent to me.
Is there a font that is always guaranteed to be there?
When a GC is created it contains a default font, although you can't
assume very much about it.
Any way
when I allocate the Pixmap there seems to be garbage in the
pixmap.
This is correct. The initial contents of a pixmap are undefined. See
page 399 of the Digital Press book, for example.
The confusing part
is that it works on all color SUNS using the mit X11 server.
A misfeature of the MIT server that is corrected in R4.lemke@lupine.UUCP (Dave Lemke) (11/29/89)
From: Daniel Seewer <uunet!super.org!dsseewe> Date: Mon, 27 Nov 89 22:56:40 EST 3) What can the following problem be related to? I have a client that runs without any problems on SUN 3/** and every SUN 4 I have tried except for the SUN 4/330 (or it might be a 4/430 if there is such a machine). The server is one distributed by SUN with Openwindows and uses cg6 (or cg9). As you can tell I really know my hardware. Any way when I allocate the Pixmap there seems to be garbage in the pixmap. That is even if the pixmap is not drawn in a nd then displayed it will have stuff in the pximap. The confusing part is that it works on all color SUNS using the mit X11 server. When the code is recompiled on the SUN 4/330it has the previously mentioned problem. Could it be the background pixmap? I am using the athena widget set X11R3. Am I doing something wrong? xnews (OpenWindows) doesn't clear a pixmap at creation. the protocol doesn't require it, and its often unnecessary. the MIT server does clear them at creation. the safe thing to do is clear the pixmap to the background color (XFillRect it) before drawing into it. [ i was told R4 will stop clearing pixmaps at creation, but i haven't checked. ] Dave Lemke ARPA: lemke@ncd.com Network Computing Devices, Inc. UUCP: uunet!ncd!lemke
keith@EXPO.LCS.MIT.EDU (Keith Packard) (11/29/89)
> the MIT server does > clear them at creation. the safe thing to do is clear the pixmap to > the background color (XFillRect it) before drawing into it. [ i was > told R4 will stop clearing pixmaps at creation, but i haven't checked. ] Actually, only some of the MIT servers cleared pixmaps. The server for the DEC GPX (QDSS) didn't, while the monochrome and color frame buffer servers all did. Dave is correct about R4 though, none of the new servers clear pixmaps. Keith Packard MIT X Consortium