CLIM mail archive

[Prev][Next][Index][Thread]

Re: Semantics of flipping inks



>From: Lawrence.G.Mayka@att.com
>Date: Sat, 23 Apr 94 15:13:56 CDT
>Subject: Semantics of flipping inks
>
>CLIM 2.0 on Allegro CL 4.2
>CLIM 2.0-beta on LispWorks 3.2
>
>I'm having some difficulty with the semantics of flipping inks.  
>  [...]
>For <color> I used, in turn, CLIM:+RED+, CLIM:+GREEN+, CLIM:+BLUE+,
>CLIM:+YELLOW+, CLIM:+MAGENTA+, and CLIM:+CYAN+.  The resulting
>rectangles were of the following colors:
>Circle		Rectangle
>------		 ---------
>red		gray
>green		black
>blue		black
>yellow		gray
>magenta		red
>cyan		light green

My experience using flipping inks has been even more frustrating than this.
The resulting color *changes* each time I run the system.  (My hypothesis
is that the exact color that ended up on the screen had something to do
with the number of slots already already in use in the X-color-map at the
time that my code started to run --but this is just a guess.)

We were trying to use large rectangles drawn in flipping-ink to signify
selection of a region.  As a work-around I tried "graying-out" the region
by over-laying a tiling-pattern that had alternating pixels drawn with
flipping-ink and transparent-ink.  However, this brought up an error that
this is not supported yet...  (We ended up drawing a bold outline in
flipping-ink.)

(On the other hand *if* you can get by with two colors, you can always set
the foreground and background inks of the window and flip between those
two.) 

>Is there any simple mental model I can use to predict the behavior of
>flipping inks?

If you find out anything useful, please be sure to let the rest of us know.
However, I suspect that the answer is "no".  This would seem to be one of
the unavoidable consequences of running what is intended to be a fully-
portable software platform on a number of widely varying host color
systems. 

N


Main Index | Thread Index