Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!ucsd!cacilj!close
From: close@cacilj.UUCP (Diane Barlow Close)
Newsgroups: comp.sys.atari.st
Subject: Re: WordPerfect revision -- Now clock problem & fix.
Message-ID: <978@cacilj.UUCP>
Date: 23 Apr 89 02:51:09 GMT
References: <0401891959340696@dynasoft.UUCP> <3237@cs.dal.ca>
Reply-To: close@cacilj.UUCP (Diane Barlow Close)
Distribution: na
Organization: CACI, Inc. - Federal, La Jolla, CA
Lines: 27

In article <3237@cs.dal.ca> bill@biomel.UUCP, biomel@cs.dal.CDN writes:
>I appreciate John's information about the clocks, but I remain a bit
>confused about what is happening.  The Paul Close program that his wife
>posted gives me separate readings for three clocks when I boot with the

There are only two clocks.  The MWC routines have two ways of reporting
and interpreting the keyboard clock.  One way returns a value of 1900, and
the other returns a value of somewhere in the year 2000.   One routine
(gettime) ``reads and returns the intelligent keyboard's setting of
the current time.  It returns a 32 bit mask whose bits indicate'' various
things.  The other routine (Kgettime) ``reads the time from the intelligent 
keyboard's clock ... and returns a pointer to the structure tm_t, which
it initializes.''  *That's* the two are different values.  

Kgettime converts between time as encoded in tm_t and the IKBD clock, 
while gettime directly manipulates system time.  It's a slight difference
that had me confused at first, and convinced that the ST had more than two
clocks.  Now I'm set straight, and I hope you are too.

My husband's program won't, by the way, freeze the keyboard clock 'cause
it only *reads* one clock and writes the other--the freezing problem happens
when you write both clocks.
-- 
Diane Barlow Close
      {uunet, crash, ucsd}!cacilj!close
      close@cacilj.uucp
      Free Canada -- Trade Mulroney