Re: XyWin & bug vs. feature

["..." adpf@xxxxxxxx]

Jim--Here's some speculation based on an empirical, decidedly crude
understanding of how xyW works: I think that v4's improved memory
management as regards big files amounts to more fluid swapping.
I suspect that anytime xyW must suddenly display text not in the active
buffer--e.g., when it reformats from TF at a save *or* when it fails
to find a search string in the active buffer--some internal imperative
forces it into the view mode your default has instructed it to use
to display a newly opened file.

Let me guess: Robert--who writes that he almost always works in XP, has
experienced an unwanted switch to page-line view "about 3 times in the
last 4 years," and calls complaints about it nitpicking--starts up with
DT set at 0?

If so, for anyone who starts up with DT >0 and does not find
the view change trivial, one "solution" is obvious. Since the partition
where my v4 is installed has turned to Swiss cheese I can't test
my theory or variations of Robert's SAV.PM, but perhaps Jim or
someone equally displeased can adapt SAV.PM for search commands
(and other situations that produce the effect--TF/BF?).

Your description brought it all back to me, Jim. When I was trying to
explore xyDos 4's big textual system files and the uninvited switch to
page-line view was driving me to distraction, it never occurred to me
that the phenomenon was anything but a bug that I assumed must have been
fixed in maintenance upgrades more recent than my release--otherwise,
how could all these folks tolerate v4? I might add that in my system one
keystroke sets up a table. I use it liberally to format normal 8.5x11
text and files I never print like printer drivers and .kbd files (not
for 10-column spreadsheets).

I've stated my own solution. I took to editing v4 printer drivers and
.kbd files with v3, which seemed so loony I had to wonder why I was using
v4 at all. I use the standard .kb3 workaround to suppress TF display
at save--DX/DOing the XC (works even with graphics cards that choke on
DX)--bracketed with calls to routines I described recently (effectively,
&FBC = BX--grabs CMline, CMline cp, CMline/text cursor location, and TI
state--and XC$7 = Q2; $7 restores all). So my save key reads:
	31=&fBC,s,a,DXXCDO$7
And tiresome as v3's reaction is when it can't find a search $ in the
active buffer--"[Blah, blah] ... Do not remove Drive C"--I prefer it to
DT defaulting to 0. Let's hope twpfkaxw has a better way. 	--a

============================= adpFisher  nyc