Re: Moving objects and incremental redisplay

• To: rm@cast.uni-linz.ac.at, CLIM@BBN.COM
• Subject: Re: Moving objects and incremental redisplay
• From: Scott McKay <SWM@stony-brook.scrc.symbolics.com>
• Date: Wed, 29 Sep 1993 12:34 -0400
• In-Reply-To: <199309291113.AA14317@arnold.cast.uni-linz.ac.at>

    Date: Wed, 29 Sep 1993 07:13 EDT
    From: rudolf mittelmann <rm@cast.uni-linz.ac.at>

    In message <19930928175757.9.SWM@SUMMER.SCRC.Symbolics.COM> Scott McKay writes:
    > ...
    > CLIM's incremental redisplay facility operates at too low a level to be
    > as efficient as the 4 steps above.  CLIM doesn't know, for example, that
    > you just moved one little object.  It has to go figure it out by looking
    > at the whole output record tree.
    > ...

    I really cannot find CLIM to be THAT bad... ;-)

    The whole record tree has to be traversed only in worst case.  If the
    programmer is careful to supply enough information via hierarchical use
    of updating-output macros, only small parts of the tree need to be
    investigated. Isn't it?

Yes.  Using UPDATING-OUTPUT supplies CLIM with information that allows
it to cut branches of the output record tree when it is searching for
things to redisplay.  But even with careful use of UPDATING-OUTPUT,
there are many cases where it is possible to do much better when the
programmer has domain-specific knowledge.  For example, a spreadsheet.

I guess the point I am trying to make with incremental redisplay is
that, for many things, it is adequate.  Not terrific, but adequate.
For some subset of those things, incremental redisplay can be quite
good.  But for other things, you can often do much better by doing the
redisplay yourself.

• Re: Moving objects and incremental redisplay
• From: rudolf mittelmann <rm@cast.uni-linz.ac.at>

• Prev: Presentation action for getting text into an X cut_buffer0
• Next: tracking-pointer clauses?
• Index: CLIM Mailing list (1-Jan-93 to 31-May-94)
• Thread: CLIM Mailing list Index (1-Jan-93 to 31-May-94)