Newsgroups: comp.windows.x
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sdd.hp.com!think.com!snorkelwacker.mit.edu!bloom-beacon!dont-send-mail-to-path-lines
From: mouse@lightning.mcrcim.mcgill.EDU (der Mouse)
Subject: Re:  A wish for R5
Message-ID: <9105090623.AA02805@lightning.McRCIM.McGill.EDU>
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Date: 9 May 91 06:23:19 GMT
Lines: 42

> [I]t would be very helpful ([...]) to have another X call for putting
> bitmaps.  Quoting from the man page for XDrawString,

>	"the drawable is only modified where the font character
>	 has a bit set to 1."

> What I am suggesting would be an XDrawBitmap() call to provide the
> same functionality for drawing bitmaps.

You can get this effect by setting the bitmap as the clip-mask in a GC
and doing a FillRectangle operation.  You can get a similar effect
(with, unfortunately, more potential for visual anomalies) by doing two
CopyPlane operations with appropriate values in the GCs:

#1: foreground=0, background=~0, function=GXand
#2: foreground=desired fg, background=0, function=GXor

However, this second way may be faster, because many servers'
implementation of clip-masks is grossly inefficient when the clip-mask
does not decompose neatly into rectangles (and most font glyphs don't).

If your pixel values satisfy one of the relationships

	(fg & bg) == bg
	(fg & bg) == fg

then you can get away with using just one CopyPlane operation.  (This
seems like a special case, but I would guess, and it is a guess, that
it's common enough to be worth checking for.  Certainly every 1-bit
display qualifies.)

> Or, allowing clients to download fonts would also suffice for this
> purpose.

I got the impression from Jim Getty's talk at the X conference last
January that R5 will contain much-improved font mechanisms that are
capable of addressing this desire.  I certainly hope so.

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu