In article <1995Jul3.113622@miavx1>, [log in to unmask] (Dennis S.
Keeler) writes:
> In article <1995Jul3.111628@miavx1>, [log in to unmask] (Kent
Covert) writes:
>> In article <1995Jun29.171055@miavx1>, [log in to unmask]
(M.C.Mastilak) writes:
>>> Not that I'm complainin', but the new NEWS scrolls a LOT more slowly than
did
>>> the old...Is this my problem, or something inherent in the new connection?
>>>
>>> Just curious, and glad I just got onto Netcom...hee hee hee.
>>
>> I think I have this problem fixed now.
>
> It's fixed for me. Thanks a million.
> Just curious, what exactly was wrong and what did you have to do to
fix
> it?
Short Answer
------------
The problem was with the way the program/system dealt with scrolling.
There are 2 ways to scroll the display:
1) Redraw each line of the screen that you want to scroll.
2) Set up a scrolling region within the terminal and let the terminal
do the scrolling.
The system was doing the former rather than the later.
Long Answer
-----------
This was a problem that was (*BLUSH*) caused by me in the national
distribution. The part of NEWS that handles this looks like the following:
if (refresh) c$cks(smg$end_pasteboard_update(&pid));
_c$cks(smg$move_virtual_display(vd,&pid,cp,c$ac(1),0));
if (refresh) c$cks(smg$begin_pasteboard_update(&pid));
This was causing some problems in some code I wrote. After looking at it,
I decided that you couldn't end something before you began it, so I
switched the 1st and 3rd statements. It fixed the error I was getting.
Just to make sure that this was really a problem, I consulted with the 2
people who are in charge of caring for the NEWS code (Charles Bailey at
UPENN and Bob Sloane at UKansas) and they agreed with me.
Anyway, the change made it into the latest version of NEWS (which we're not
running yet, but I had implemented this change in our NNTP version).
Immediately, people began reporting this strange scrolling problem in NEWS.
Someone realized that it was this code change that I made (apparently there
is a call to smg$begin_pasteboard_update earlier in the code execution that
I, Charles, and Bob had missed). But they also realized that changing it
back also caused problems in certain situations, but on Alpha's only. BTW,
the problem manifests itself as an access violation error.
I'm hoping to spend some time in the near future to further diagnose this
other problem and get the code completely correct, but for now, my original
error is gone (I think) and scrolling should be improved.
--
Kent Covert, Software Coordinator
Miami Computing and Information Services
Miami University, Oxford, OH
[log in to unmask]
|
