Path: utzoo!attcan!uunet!mailrus!cornell!rochester!uhura.cc.rochester.edu!elmo
From: elmo@uhura.cc.rochester.edu (Eric Cabot)
Newsgroups: comp.lang.pascal
Subject: Re: (R)window() problems in turbo 5.5
Message-ID: <9622@ur-cc.UUCP>
Date: 26 Sep 90 14:37:38 GMT
Organization: University of Rochester, Rochester NY
Lines: 29

In article <90268.130854SPRAGGEJ@QUCDN.BITNET> SPRAGGEJ@QUCDN.QueensU.CA
(John G. Spragge) writes:
>I don't think it's a bug... more of a feature.
>

and futhermore:
>What to do? Consider surrounding the window procedure with your own
>routine to make it a little more friendly.

I've been confronted with this little  "feature"  for about 3 years in
one of my programs.  My strategy was to ask users to employ a command line
parameter (/f for flaky BIOS) if they get a funny effect when using the
part of the  program that has a 1-line window.  My experience is that
all Zeniths more that 1 year old and all IBMS-PC's  till around 1986 
(or perhaps even later) can't handle the one-liner.

If the /f switch is encountered during the program startup, the screen
is cleared and completely recomposed rather than just closing the  window
and recomposing that line. Unfortunately, it is just those  poor  souls
that have to go that route that are most likely to notice the difference
since they have older, slower machines. 



-- 
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
Eric Cabot                             |  elmo@{uhura | db1}.cc.rochester.edu
      "insert your face here"          |  elmo@uordbv.bitnet
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=