Re: Xywin bugs

[Peter Strisik 76556,2651]

Kevin,

I got the XWDFL.DFL file and it works like a charm. I even discovered some new
searching power from the dialog box that I didn't know was there.

The button bar is still clipped. If I change the overall height of the format
bar, it just makes the bar bigger and the buttons within it bigger, but the
buttons remain clipped at the bottom. I fiddled within the XWUIF.UIF file and
managed to get the buttons smaller, larger, offset further down, etc., but
could not prevent the clipping at the bottom of the buttons!

           ........ Peter  - Fri, 8:49pm (in Atlanta)


Hi. Are we getting together for a drink/dinner? I got a few places in mind. Right now isn't so
good, but middle of next month looks great. I'm to London on Tuesday, then Maine for 7-10 days,
then business here, then Texas, then gotta edit a book. But there's a window around Oct.12-19.
What are your plans in that period?

Just picked up SMARTS24.ZIP. Installed it. Playing with it. Random thoughts:

Looks like they reactivated C-frames, aye? Good news. This company is improving.

In {{5,PrsDeFine}} I'd use ®XS01,03,02,04,50¯ in place of ®XS01,03,02,,50¯.
That would leave open the option of the target frame picking up the CMline via IS00 (in case
somebody wanted to take advantage of that, which is perfectly possible). Note that the other Prs
frames avoid resetting or anywise touching 00. BTW, PrsDeFine is indeed a useful variation on the
theme.

Could {{5,Ds}} use different masks for presenting the date|time? The default (only) report looks
quite foreign & unfamiliar to me - is that a standard format somewhere? Like in the law biz?
Didn't you just love those Al Sharpton for Senator posters? "`Cause he's effective!"
`Cause I aint never seen dat floormat befo'!

Weren't "ch" and "sh" original abbreviations for {{5,CallHlist,Chl}} and
{{5,SaveHlist,Shl}}? I think you should keep them (because they've become habitual to at least one
user).

KEYPAD is nuts. How do you think this shit up? LOADEVAR too. Good tool, that one.

In PUTVALUE, I'd dump the " |=| ". It seems to me that if I define the expression
®VA$DAd Mmmm yyy¯ it should just put out "16 September 1994", not " |=| 16
September 1994" -- and not at the end of the line either, but rather where my cursor is. The
only people who will use such a thing are programmers, and they're pithy precise people. They know
they're running PUTVALUE, so don't require the " |=| ".

MEMO is fun... And it even works, too.

I like my {{5pfun}} better than yours!:
{{5pfun}} PFUN
ÿ?Ù(pfun ®PRPFUN-ing...¯®RC¯®RC¯®PR ¯)

BLANK is quite interesting, as to the DOS possibilities. It's also disturbing though, because to me
(or any onlooker) it appears that something is wrong with my machine. Specifically, it looks like a
hung computer. And thus might raise rather than lessen interest (as intended? what is the intent?
to lock unauthorized people out of the computer?) This routine simply looks like genuine benign
inactivity (hit any key to restore the header and get a JMP command that will return you to text;
this can also be setup for a password like "shit"). Regrettably, it doesn't look as
convincing in graphics mode:

{{5end}} Go to EOF and `disappear' the text RJH 5/22/93 Rev.9/17/94Put [optional] password in
ÿAESV01,ÿAF
®SV01,shit¯ÿ?A®SX02,®CP¯¯ÿ?M®SX03,®CP¯¯®LBA¯ÿ?®IF®CP¯<®PV03¯¯ÿ??®GLA¯®EI¯®SX01,@UPR(®IS01¯)¯®SX03,®VN$XM¯¯ÿ?Ù(d xm=*TI)®SV04,¯®LBB¯ÿ?ÿ??ÿ?;®PR ¯®IF®VN|01¯<1¯®RC¯®GLD¯®EI¯®PV04¯®SV04,®PRPassword¯¯®SV05,¯®LBC¯®SX05,®IS05¯+@UPR(®RC¯)¯®IF®VN|05¯<®VN|01¯¯*®PV04¯®GLC¯®EI¯®IF®IS01¯<>®IS05¯¯®GLB¯®LBD¯®EI¯ÿ?Ù(d xm=®PV03¯)ÿ?jmp ®PV02¯ÿ?}

International standard my foot (DS).

Your description of XSTREET is great. Especially the last line. Like if they travel, where else?

In SAVESESS, I think you should expand this sentence: "Write a little Help frame that lists
these files on demand (e.g., ÿ?«dir d:\path\*.sesÿ?), and use your RUNCODE key to
switch quickly to the desired session." Well, I don't get how RUNCODE is going to help me in a
DIRectory display, and neither will anybody else (I think).

SAVESESS also has a bug in it. If I run without args (and without any configuration either, just
straight factory-issue), it stops working after it asks the "Close files?" question.
EXits (apparently) without comment. Haven't investigated why.

In FMARK, do you need four markers "", "", "" "" --
can't you just use one marker and write ÿAEVN@014ÿAF? Curious...

I like the FMARK demos -- very effective. That's the best way to get people to grasp programming
power. FMARK works real well, too.

In LOG I'm getting entries like "09/19/94 12:18 AM: 09/19/94 12:18 AM: " which are due to
- what? lack of a carriage return? Maybe LOG ought to look for a terminating carriage return as the
last char in the file and, if not found, supply one.... Good idea, though.

You ought to mention how to turn TICKER off! They probably don't know.

How about an SFIND in addition to SDIR? That would be quite handy.

GOFILE is great. But I think it should have a different framename than that of a native Xy command.
 Begs confusion.

I'd prefer if FINDDIFF didn't go to ÿ?ñbut just accepted current cursor positions. Almost
always, I use it to compare small sections of two texts, not to compare whole files. If I want the
latter, I can just set both cursors to ÿ?ñ.

EG is great. Here's something odd about it, though. If I EG in ÿ?­mode, I get an
"NP" in my style, whereas if in ÿ?¯mode, I get an "AP". No, wait,
that's inacccurate. The *first* time I ask for an EG in *either* mode (with no style previously
issued in the doc), I get "NP"; subsequently I always get "AP". Isn't that a
printer pause command from way-back XyWrite II+ or so? Wierd.

One problem with these maintenance upgrades is that I lose my customization of things like FISHOUT,
because I forget that I've already customized them. Shit. Too bad. Neither did I maintain a list
of what's customized and what isn't, because... well, it's extra work. Wish I could think of a way
of flagging those customizations automatically.

Shit you've done so much programming its just amazing.

REPEAT is interesting programming, but I still think the flaw in it is that its report presents you
with a *lot* of reading, mostly unnecessary -- because in real writing 99% of computer-flagged
"redundancies" are benign. I'd like a much pithier report, no more than 1 line per
occurrence.

I was going to say that I think READ would be more useful if it had the capability of toggling
various attributes, not just read-only -- but then I looked at your routine, and don't understand
what you're doing there. Why not just use the ATTRIB command? Don't think that you need to
ÿ??anything.

An immense amount of programming!

I gotta sleep. To England tomorrow night. I'll post a public msg when I get back mentioning my
favorite routines. Bye for now.


Date: 22-Sep-94 01:18 EDT
From: Harry Binswanger [71571,3052]
Reply to: XyWrite

Robert:
 Good to hear from you again. I, too, dislike XyWin. My
problems with XY4 for DOS have to do with most of the ways
it is different from III+. E.g., the help menuing system
and the Wysiwyg screen.
 But more specifically: the old "DO" command doesn't seem
to work (though "DOS /C" does); ctrl-alt-del doesn't always
get a response; neither ≪PI≫ nor ≪EF≫ send anything at
all to my printer (judging by what I see in FO.TMP).
  There's more, but I use it so infrequently that I can't
remember all of it.
  Is there still a user BBS somewhere?
  And where is the list of what the error messages mean (e.g.,
I get #214 a lot, which I think means "file not found" during
startup.int).

  Some of the stuff is great--esp. having wildcards in search
and replace! That is a dream and works flawlessly.
  Regards,
  Harry


Date: 22-Sep-94 01:21 EDT
From: Harry Binswanger [71571,3052]
Subj: Additional Query

  Robert:
  Forgot to ask: do you know which is the latest version of
XyDOS? And does Tech. Group supply drivers for the HP4+
(just got one and it is incredibly fast).
  Harry