1989 Computer Message Boards
Msg#8821 To:ROBERT LAMERAND
From:RON LANE Date:04/01/89
Subject:(R)CLIPPER FILE HANDLES P Parent Msg#8763
------------------------------------------------------------------------------
On a network the SERVER needs to have the files set to the maxinum number
that could ever be open IE: if the app can open 10 files and there are 5
workstations then the files= on the server should be 50 +dos & network also don
forget to set CLIPPER=F50 in your autoexec.bat. Make sure your network
software is set to files=255
Msg#8824 To:ALAN GUGGENHEIM
From:STEVE STRALEY Date:04/01/89
Subject:(R)DOSS ARTICLE Parent Msg#8508
------------------------------------------------------------------------------
Yes, I will be writing an article in DOSS on how to write out ANY size
text file within MEMOEDIT(). It's rather simple once you got a few basic
tricks down, but it is a major HASSEL to program. Stay tuned to DOSS...
it will be coming up! And by the way, sorry for the long delay... I've
been out of town, but now that I'm back, I've got to get back to some
tool-breaking work. Take care...
Steve
Msg#8825 To:DIANE LASK (Rcvd)
From:STEVE STRALEY Date:04/01/89
Subject:(R)NANTUCKET Parent Msg#4836
------------------------------------------------------------------------------
Diane, I don't nash teeth over the fact that information coming from the
parent company is slow, I don't nash teeth over the idea that Tools are
being introduced that either compete with my friends or no longer make
their products work, I don't nash teeth when 640K bariers are reached
with very little effort. What I do nash teeth over is non-paying
customers and earthquakes. I guess I'm lucky, I haven't had either. So
I think you've got the wrong impression on me.... but thanks for the
words....
Steve
Msg#8826 To:CONSTANTIN FLORESCU (Rcvd)
From:STEVE STRALEY Date:04/01/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
Dear Constantin,
On behalf of SJS & Associates, I want to personally thank you for taking
a stand. Personally, it is very difficult to see what is happening to
the langauge, but to my friends involved with this issue. Competition is
always a touchy thing. SOmetimes it's good, sometimes it bad. I'd
rahter not publically comment on this. Our official position is as
follows:
SJS & Assoicates is dedicated to the Clipper developer and the core
Clipper language. We will do everything in our power to be bigger,
better, and faster. As per claims that such products "legitimize" our
market, we disagree. Sales and support legitimize our market: nothing
more.
And on that note, I think Dirk Lesko and I have a few interesting
announcements to make within the next 2 weeks. Additionally, if people
are interested, I'll upload a list of new features in ToolkiT 2.0 (Dirk's
seen it so it does exist) with a scheduled release date.
Again, many thanks... on behalf of ALL of the 3rd party guys. Busines is
war and this too shall pass. Again, thanks for the support!
Steve Straley,
Senior Partner - SJS & Associates
Msg#8827 To:DIRK LESKO (Rcvd)
From:STEVE STRALEY Date:04/01/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
Hey Dirk, how about this line for the diddy...
It's a big mistake...
Now you look like Ashton-Tate..
It rhymes... which is rather amazing since I'm still stuck on...
Willy, willy, ain't no more...
For what he thought was H2O,
Was H2SO4.
Steve
Msg#8829 To:DAVID MORGAN
From:STEVE STRALEY Date:04/01/89
Subject:(R)UN-IVING DBFS Parent Msg#8770
------------------------------------------------------------------------------
Dave, do you have Phil Kimble's number.... since he moved out of the Long
Beach operations area, I've since lost it... THanks...
Steve
Msg#8832 To:ALL
From:STEVE STRALEY Date:04/01/89
Subject:I'M BACK Parent Msg#8832
------------------------------------------------------------------------------
Dear Developer,
After a long seminar schedule, it is good to be home. Hope some people
out there haven't forgotton.
If there is interest, and with the permission of Dirk, I've got a file to
up load in a couple of weeks with a couple of new features.
Additionally, I hope at the end of April to have a blow-by-blow release
date on every package (yes, committing myself to a release date).
Again, hope all is well with everyone and stay tuned to this channel,
Dirk and I have some interesting news on the way....
Steve
Msg#8833 To:DIRK LESKO (Rcvd)
From:RON LANDBERG Date:04/02/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
I just received today Nantucket's offer for its Clipper Tools One
package. The letter had the following two sentences, with the first
being most tantalizing:
"Soon, memory problems will be athing of the past - no overlays, just
super fast linking. Then watch - you'll find that future releases from
Nantucket are committed to providing you with a database development
system that meets the requirements of emerging technologies; support for
multi-operating systems, SQL, object oriented-programming and virtual
memory."
I wonder what they mean by `Soon' in the first sentence?
Msg#8834 To:CONSTANTIN FLORESCU (Rcvd)
From:RON LANDBERG Date:04/02/89
Subject:(R)GETBACK Parent Msg#8762
------------------------------------------------------------------------------
You can do what you want. In procedure BY, before looping back to the
GETs, stuff the keyboard with enough CHR(13)'s to get you to the GET you
want to get to. I.E., if you want to get to the third GET, issue the
command as follows:
KEY)+CHR(13))
This will cause the first two GETs to be treated as READ as if the
operator had pressed the return key after each.
Msg#8835 To:JON MELTZER (Rcvd)
From:RON LANDBERG Date:04/02/89
Subject:(R)WE DON'T NEED MACROS! Parent Msg#8796
------------------------------------------------------------------------------
Use that one all the time like this:
oldcolor = SETCOLOR("W/N,")
.
.
.
SET COLOR TO (oldcolor)
Msg#8836 To:CONSTANTIN FLORESCU (Rcvd)
From:RON LANDBERG Date:04/02/89
Subject:(R)GETBACK Parent Msg#8762
------------------------------------------------------------------------------
The command line got garbaged somehow in my last message. It should read:
KEYB(CHR(13)+CHR(13))
Msg#8838 To:STEVE STRALEY (Rcvd)
From:DIRK LESKO Date:04/02/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
Well, the versus are lining up like 'soldiers'...now I have too
many......Seems all year I was afraid I would have nothing to say since
Nantucket had been 'a good little company' all year.....But now I can't
turn the meoldy maker off!..
dLESKO
Msg#8839 To:STEVE STRALEY (Rcvd)
From:DIRK LESKO Date:04/02/89
Subject:(R)D.O.S.S Parent Msg#8828
------------------------------------------------------------------------------
Steve....Put it in a TEHELP FILE!.....
dLESKO
Msg#8841 To:STEVE STRALEY (Rcvd)
From:DIRK LESKO Date:04/02/89
Subject:(R)I'M BACK Parent Msg#8832
------------------------------------------------------------------------------
Go ahead Steve, upload away....Can't wait for the 'blow by blow'
release...
dLESKO
Msg#8842 To:RON LANDBERG (Rcvd)
From:DIRK LESKO Date:04/02/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Soon means they'll have something to say or show at the Developers
conference. The Dynamic Linking Stuff will be RTlink for Clipper....If
you want to catch a glimpse of the future, just get a hold of RTlink
now....Then decide if you like the whole idea.....It will create some new
problems....ANd I don't feel that it will releive the concerns we all now
have...
dLESKO
Msg#8843 To:DIRK LESKO (Rcvd)
From:STEVE STRALEY Date:04/02/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
I know what ya mean... I know Santa ain't gonna fill their stockings this
year cause we now know who's been naughty and who's been nice... hmmm...
another potential line??? Nahhh...
Steve
Msg#8844 To:DIRK LESKO (Rcvd)
From:STEVE STRALEY Date:04/02/89
Subject:(R)D.O.S.S Parent Msg#8828
------------------------------------------------------------------------------
Put what in a TRHelp file??? the Newsletter????
Steve
Msg#8847 To:STEVE STRALEY (Rcvd)
From:DIRK LESKO Date:04/02/89
Subject:(R)D.O.S.S Parent Msg#8828
------------------------------------------------------------------------------
All your back issues...indexed of coursed
dLESKO
Msg#8850 To:JEFF SCOZZAFAVA (Rcvd)
From:DAVE LOBEL Date:04/02/89
Subject:(R)READS Parent Msg#8791
------------------------------------------------------------------------------
Ecellent and to the point.
Judges give 9.5 9.7 10.0 9.7 9.8 2.5 (from guess who???)
YEL-DAVE
Msg#8851 To:DIRK LESKO (Rcvd)
From:DAVE LOBEL Date:04/02/89
Subject:(R)WE DON'T NEED MACROS! Parent Msg#8796
------------------------------------------------------------------------------
Jon Meltzer,
I have put a lot of lines together on a Novel Network, but never had a
problem. Then again, I use ALIAS() instead of SELECT().
Let me know, and vice versa, if you hear of anything else.
YEL-DAVE
Msg#8852 To:ALL
From:DAVE LOBEL Date:04/02/89
Subject:VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
HELP.....
Is there anyone out there that has a SONY Multiscan monitor and a VIDEO
SEVEN (7) FAST WRITE VGA video card?? Or any VGA card???
For some reason my 25 lines only show on part of the screen. All do
show but and inch and a half from the top and bottom.
Does anyone have a solution.
I'm using version 1.09 of Fast Writes Software.
THANKS.......
YEL-DAVE
Msg#8853 To:CONSTANTIN FLORESCU (Rcvd)
From:CLINT BRITT Date:04/02/89
Subject:(R)IDL.LIB Parent Msg#8630
------------------------------------------------------------------------------
Thanks,
Will take a look at it........
Clint.....
Msg#8854 To:STEVE STEINER (Rcvd)
From:CLINT BRITT Date:04/02/89
Subject:(R)IDL.LIB Parent Msg#8630
------------------------------------------------------------------------------
Steve,
Call me a old dog, but I like to stick with things, that have
worked well with me in the past...., always looking to better things, but
hard to sway, thanks, It will be a pleasure to see whats coming up !!!
Clint....
Msg#8855 To:DIRK LESKO (Rcvd)
From:CLINT BRITT Date:04/02/89
Subject:(R)IDL.LIB Parent Msg#8630
------------------------------------------------------------------------------
Dirk,
I have been seeing alot of news on FUNCky, but have not seen a
copy, if there is a way I can I would be glad to see it. You can if you
want count this as a order, If you could, and I know that you sell it,
send out one copy COD. TO: 2713 W. Richardson Place, Phase One Software
Inc., Milwaukee Wi. 53208, Attn: Clinton D. Britt. This sounds
interesting, hummm.
P.S. Leave a message, telling the whole price...
Clint.....
Msg#8856 To:DIRK LESKO (Rcvd)
From:CLINT BRITT Date:04/02/89
Subject:(R)INTERRUPT() Parent Msg#8749
------------------------------------------------------------------------------
Dirk,
Will look for silvercomm. I have done about 50% of the code now,
It has a questionair generater, 2-line access, but need's a major bit of
work to go three, has ansii menu's that have a language of there own to
include data from the databases, it's actually a pretty fun project, will
give you a copy when done......
Clint...
Msg#8857 To:DIRK LESKO (Rcvd)
From:CLINT BRITT Date:04/02/89
Subject:(R)INTERRUPT() Parent Msg#8747
------------------------------------------------------------------------------
Thanks agian dirk, will look for it....
Clint....
Msg#8860 To:DAVE LOBEL (Rcvd)
From:JOHN THORN Date:04/02/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
Dave:
I've a friend with a Video 7 VGA and the darn this is lousy with ROM
bugs! I've a Paradise and have not had ANY problem with compatiblity.
Suggest you give Video 7 people a call and complain....not that it will
do any good (my friend has and only received a new ROM with other
problems)!
Msg#8861 To:DIRK LESKO (Rcvd)
From:RON CALLAHAN Date:04/02/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Maybe it's time to review the situation with Ashton-Tate. At least
they've always been 100% upwardly compatible.
Ron.
Msg#8862 To:CLINT BRITT (Rcvd)
From:NEIL WEICHER Date:04/03/89
Subject:(R)INTERRUPT() Parent Msg#8747
------------------------------------------------------------------------------
The function ONCOM works like this:
N_ONCOM("proc1") calls procedure PROC1 when carrier is lost, detected
or ring is detected on COM1. Similarly N_ONCOM2 for COM2. The procedure
is your code and you can do what you like with it. If you call N_ONCOM()
with no parameters within the procedure it tells you why you got there...
e.g., "C" for carrier detect "c" for carrier lost "R" for ring.
You can Get-It (pun intended) by calling 212-724-0150 and asking for
Jackie. It is $99 plus s/h.
Neil
Oh yes, as in all our software you can return it within 30 days for a
refund.
Msg#8863 To:JEFF SCOZZAFAVA (Rcvd)
From:NEIL WEICHER Date:04/03/89
Subject:(R)NET-LIB
------------------------------------------------------------------------------
Sounds like you need the latest version of NetLib. The N_BOLT function
has been fixed. Also you should get the latest for Get-It. There were
some initial problems with working with Silverware but they have also
been fixed. Call 212-724-0150 and ask for updates.
Neil
Msg#8864 To:JON MELTZER (Rcvd)
From:DAVID MORGAN Date:04/03/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
> "dynamic link libraries" and "virtual memory"... worr[y] me
... is Nantucket aware of the .EXE file size concern John Meltzer
> It will help some but I don't think it is going to do what we
really want. Dirk Lesko
The operating assumption I gather from Clipper users is that they
aren't concerned with EXE file size per se, and that "what we really
want" in real-world terms is eradication of the out-of-memory
condition. Sometimes this memory concern is expressed as a file size
concern, but a concern over memory it truly is and not actually over
file size at all. If the avenue to banishing out-of-memory were to
triple the size of the EXE, Clipper users with big applications would
welcome it.
63K command files in CP/M were considered "too big." But only
because they stood in relation to a a 64K memory space limit. 65K
executables in DOS, even though they are bigger still, are not
regarded as "too big." There was nothing intrinsically "too big"
about 63K of code. Nor is there anything intrinsically too big about
400K, 500K, or 600K of code in an EXE. That's how OS/2 can genuinely
be presented as a solution, even though it facilitates executables
bigger than the ones that are today too big. Executable file sizes
aren't problematic in a vacuum, but only in conjunction with
exogenous memory limits that may apply to them.
The typical Clipper user who feels his application is "too big" might
have an EXE file of 350K. Maybe 325K of that is to-be-loaded code
(not everything in an executable file is executable). He has a
max-configured computer with 640K of memory and 32MB of disk; DOS
occupies 75K. His complaint every time is that the application
leaves him only 240K of memory, rather than that it leaves him only
31.7MB of disk.
Msg#8866 To:CLINT BRITT (Rcvd)
From:DIRK LESKO Date:04/03/89
Subject:(R)INTERRUPT() Parent Msg#8749
------------------------------------------------------------------------------
well Good luck, I think you will find John halvonic at Silverware most
helpful should youdecide to get silvercomm...
PS, I think there is a demo of FUNCky on the board if you download
FUNCky.arc, or FUNCky2.ARC....should give you an idea of what it can do
for you...
dLESKO
Msg#8868 To:RON CALLAHAN (Rcvd)
From:DIRK LESKO Date:04/03/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
What? and jump from the frying pan into the fire?? However, when the
'compiler' ever makes it outwe will be looking at it...
dLESKO
Msg#8869 To:DAVID MORGAN (Rcvd)
From:DIRK LESKO Date:04/03/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
How does that answer the question? If I may pose it directly, how much
memory are we going to gain from a dynamic link library under DOS?
that is the question we would like to see answered. thanks.
dLESKO
Msg#8870 To:DIRK LESKO (Rcvd)
From:CLINT BRITT Date:04/03/89
Subject:(R)IDL.LIB Parent Msg#8870
------------------------------------------------------------------------------
Dirk,
You got it, ship away....... do me one favor, leave a little
note on the super stuff in it, so my mouth can water until it gets here.
(100 words or less), ha..ha....
P.S. so that comes to 154.00 exactly, correct ???
Clint....
Msg#8872 To:STEVE STRALEY (Rcvd)
From:CONSTANTIN FLORESCU Date:04/03/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
Thanks.
Steve, both you and Dirk can count on my support and probably of evryone
loging in on this board. Most of us are here...because you are here...
Personally I don't think Nantucket Tools can ever beat yours or Dirk
Lesko products. After all, how good a dentist can be writing codes?( And
I am serious). First, he has to take care of teeths, then do other
things...But if you try to do everything(dentist+programer) and this
apply to Nantucket also, then you end up doing nothing good...Yoy guys
know what I mean.
<CF>
Msg#8873 To:RON LANDBERG (Rcvd)
From:CONSTANTIN FLORESCU Date:04/03/89
Subject:(R)GETBACK Parent Msg#8762
------------------------------------------------------------------------------
Thanks Ron
<CF>
Msg#8875 To:JOHN THORN (Rcvd)
From:DAVE LOBEL Date:04/03/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
John,
Thanks. I spoke to them today (service), and thay said that I have the
latest release. Tech support is supposed to get back to me.
Thanks Again.
YEL-DAVE
Msg#8877 To:CONSTANTIN FLORESCU (Rcvd)
From:DIRK LESKO Date:04/03/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
I thin he ended working for Nantucket in the UK office.....
dLESKO
Msg#8879 To:ALL
From:RICH WINTER Date:04/03/89
Subject:CLIPPER AND TLINK Parent Msg#8879
------------------------------------------------------------------------------
I've been trying to use the new version of TLINK which came with Turbo-C
version 2.0 to link some routines written in TC to some of my CLIPPER
programs. I am having two problems which may or may not be related. The
first problem is that every time I TLINK a clipper program, the linker
gives me an error message reading:
Fixup overflow in module \CLIPPER\SRC\CRT0.ASM at _TEXT:0021, target =
2BF4:FFFE
The second error is that 99% of my programs go into a hangup forcing a
rebbot at some point. Has anybody else come across these problems using
TLINK 2.0 and come up with any sort of solution?
Msg#8880 To:ALL
From:RICK RAMHAP Date:04/03/89
Subject:JOB OPENING! Parent Msg#8880
------------------------------------------------------------------------------
Position Available:
Long Island-based software development firm currently has openings for
experienced Clipper programmers. Applicants should be familiar with
multi-user
system development. Proficiency in C is a definite advantage.
A Princton, NJ, base is possible.
Call Brian Young at 516-889-9236 for further information.
Msg#8881 To:ALL
From:HOWARD KAPUSTEIN Date:04/03/89
Subject:TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
I just got Clipper's ad for Clipper Tools One and I have one question
(primarily to you Dirk): What does CTO do that FUNCky does not? From
everything I've heard and seen, FUNCky has twice the features of CTO, and
most of them are actually worth mentioning. In the advertising they say
"...array of functions normally available only in DOS or BIOS. ...check
the DOS switches BREAK and VERIFY. You can even check for the presence of
a math co-processor." Ohthose are trivial interrupts. Here's my question:
Why should I spend $195 for CTO when I can get FUNCky for $195, and get
more than triple the value? Even better, why did Clipper decide to offer
this library?
Msg#8882 To:ALL
From:HOWARD KAPUSTEIN Date:04/03/89
Subject:PC PURSUIT
------------------------------------------------------------------------------
Lately it seems like I can't get a connection here via PCP. I'm calling
direct right now. In the last couple of weeks, any time I want I can get
a 2400 connection via PCP to NJNEW, but it never connects. All I get is
BUSY or NO CARRIER. I just called now and I get a connection. Is this
part of Telenet's wonderful new enhancements to their service?
Msg#8886 To:RICH WINTER (Rcvd)
From:DIRK LESKO Date:04/04/89
Subject:(R)CLIPPER AND TLINK Parent Msg#8879
------------------------------------------------------------------------------
Can't use 2.0, Use 1.0 1.1 or 1.5 only....they goofed it with the OBJ
formats. They did not support the new Microsoft OBJ format in 2.0, it
adheres only to the Intel OBJ format which Microsoft deviates from.....
dLESKO
Msg#8887 To:ALL
From:RON LANDBERG Date:04/04/89
Subject:WK1 FILES Parent Msg#8887
------------------------------------------------------------------------------
Is there a straightforward way to read WK1 (1-2-3) files into a CLIPPER
application ?
Or (I apologize in advance) is there a straightforward way to read those
1-2-3 files into a DBASE 3+ application?
Msg#8889 To:DAVE LOBEL (Rcvd)
From:JOHN THORN Date:04/04/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
Dave: Saying that Video 7's Tech people will get back to you is the
equivalent to saying any one of the following:
The Check's in the Mail
I'll respect you in the morning
The *New* Clipper will be ready, on time, and sans bugs
Msg#8891 To:RON LANDBERG (Rcvd)
From:DIRK LESKO Date:04/04/89
Subject:(R)WK1 FILES Parent Msg#8887
------------------------------------------------------------------------------
call Neil Weicher at Communications Horizons....His product dOneTwoThree
will do the trick for you and elegantly...
dLESKo
Msg#8892 To:RON LANDBERG (Rcvd)
From:RICHARD HORWITZ Date:04/04/89
Subject:(R)WK1 FILES Parent Msg#8887
------------------------------------------------------------------------------
Yes, set up your DBF structure the in the same format as your file, and
then "APPE from <file>.wk1 TYPE wk1". This is documented under APPEND
in the Dbase Manual. You may also want to check out D123 from
communications horizions.
Msg#8894 To:RON LANDBERG (Rcvd)
From:DAVE LOBEL Date:04/04/89
Subject:(R)WK1 FILES Parent Msg#8887
------------------------------------------------------------------------------
Ron,
There are two ways that I know of.
1) There is a book with all file formats (Famous PC File Formats,
WALDEN), ask Dirk for the exact title;
2) Neil Weicher has a software package called dOne-Two Three that will
let you access 1-2-3 files from dBASE and Clipper.
Good Luck......
YEL-DAVE
Msg#8895 To:JOHN THORN (Rcvd)
From:DAVE LOBEL Date:04/04/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
John,
Beleive it or not, they did call back, BUT I wasn't home. Maybe I'll get
them in the morning.
Are you recommending that I return the V7 Fast Write and get something
else???
Keep me informed.
THANX......
YEL-DAVE
Msg#8896 To:DIRK LESKO (Rcvd)
From:DAVE LOBEL Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
FACE IT ALL.........
IT's becomming a FUNCky world out there!!!!!!!!
YEL-DAVE ("FUNC-et") figure it out.......
Msg#8897 To:DIRK LESKO (Rcvd)
From:RON CALLAHAN Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
Dirk:
I have nothing against Nantucket making money. In fact, I hope they
make more money than they can use, if possible. Of course I wish the
same thing for myself. But, I think that they should also take care
of business at hand. I think releasing a below standard set fo
functions is an unwise move at this time. They would be much better
off working closer with those producing lib.'s. (Setting some standards
perhaps, to avoid confussion.) Also the current version of Clipper has
some problems that have not been addressed. I, and others I'm sure, would
be more than happy to pay an update fee to see these problems resolved.
Ron.
Msg#8898 To:DIRK LESKO (Rcvd)
From:DAVID MORGAN Date:04/04/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
> how much memory are we going to gain from a dynamic link library
under DOS?
The answer to that question, I'm afraid, I don't know Dirk.
Implementation details for upcoming products generally, I don't know.
Goals for them, however, as stated on various occasions (e.g., local
user group appearances in So. California) by the implementors, I can
repeat. They are to eliminate memory inadequacy. Under DOS.
Possibly in a machine configured with as little as 256K. Whatever
the "how," that's the "what."
The point I'd like to make is that the much discussed file-size
reduction approach, 1) should be couched first of all in terms of
memory-size rather than file-size, and 2) more importantly, is but
one potential approach among several. If there exist more than one
ways to skin a cat, and your paramount desire is to a have a cat
skinned, who cares how so long as it works good?? If you run out of
gas roadside and stick out your thumb with gas can in hand, do you
refuse a tow from a tow-truck that happens to come along because you
never expected it: "Thanks, but that's not what I had in mind,
actually."? It'd be a gift horse. Aren't ends more important than
mean and isn't that what makes them ends?
I have no crystal ball, certainly, but the public, me among them, has
been told that memory's going to get solved. If it does, genuinely,
by any method, even if not the "best of all possible," the product
will be ahead and so will we. <stepping off soapbox>.
Msg#8899 To:DIRK LESKO (Rcvd)
From:CONSTANTIN FLORESCU Date:04/04/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
I can bet, that is a part-timer dentist...He has to make up for his
salary...how much can he get paid by Nant anyhow...35-40K.. whreas as a
dentist he could do better.
<CF>
Msg#8901 To:RICK RAMHAP
From:CONSTANTIN FLORESCU Date:04/04/89
Subject:(R)JOB OPENING! Parent Msg#8880
------------------------------------------------------------------------------
Quanto dinnero ?
<CF>
Msg#8902 To:HOWARD KAPUSTEIN (Rcvd)
From:CONSTANTIN FLORESCU Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
How right you are Howard. I got it too that offer the other day and I
did not even open the envelope...it ended up same place with Shop Rite
etc flyers. And is not that FUNCky is much superior but also look at the
support we get from Dirk...
I should have taken the time to send that envelope back to Nantucket and
tell them to first solve their big problems MEMORY etc and then compete
with their own supporters. I really believe their business decision is
short of silly...
<CF>
Msg#8903 To:DIRK LESKO (Rcvd)
From:ROB HANNAH Date:04/04/89
Subject:(R)WE DON'T NEED MACROS! Parent Msg#8796
------------------------------------------------------------------------------
I use select() all the time under netware (2.0a & 2.12) without the
slightest problem. As far as select goes, I generally code it like
select select( 'clients' )
That way, if clients doesn't happen to be open, I end up with select 0
instead of a bomb.
ROB
Msg#8905 To:DAVID MORGAN (Rcvd)
From:DIRK LESKO Date:04/04/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Well that's just it. The overall outcome is still left hanging. The
explanation of how it is going to work is still very vague. If a dynamic
link library under DOS was the way to go then we would all have bought
RTlink a long time ago. That hasn't happened so I don't see a rosy future
for that method until it is handled by the operating system....
dLESKO
Msg#8906 To:CONSTANTIN FLORESCU (Rcvd)
From:DIRK LESKO Date:04/04/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
Who knows? In Germany it may be different. IF anybody out there has
Tools, you are welcome to upload your comments....
dLESKO
Msg#8909 To:DIRK LESKO (Rcvd)
From:NEIL WEICHER Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
Dirk:
Which question do you want us to answer first!
Neil
Msg#8910 To:DIRK LESKO (Rcvd)
From:RICH WINTER Date:04/04/89
Subject:(R)CLIPPER AND TLINK Parent Msg#8879
------------------------------------------------------------------------------
<Expletive deleted>! That means I can't use Turbo Debugger to trace
through and debug mixed language programs!
Msg#8911 To:NEIL WEICHER (Rcvd)
From:DIRK LESKO Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
Doesn't matter, ust looking for honest opinions....BY the way..Hi !
dLESKo
Msg#8912 To:RICH WINTER (Rcvd)
From:DIRK LESKO Date:04/04/89
Subject:(R)CLIPPER AND TLINK Parent Msg#8879
------------------------------------------------------------------------------
Yup...Call Borland and ask em why they changed...I thought they would
have opted for closer compatibility...
dLESKO
Msg#8913 To:DIRK LESKO (Rcvd)
From:JON MELTZER Date:04/04/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
That's all? No action on the .EXE size concern? If I wanted RTLink,
I'd buy RTLink ...
Dirk, you and Steve have to give them hell at the conference!
Msg#8914 To:STEVE STRALEY (Rcvd)
From:JON MELTZER Date:04/04/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
So, where do the rest of us poor little elves go to get our
Christmas goodies (smaller .EXE files!!) ?
Msg#8919 To:DIRK LESKO (Rcvd)
From:JON MELTZER Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
Well, I just tried call __setwin ... and then I went out into the hall
and kicked the water fountain ...
can you say anything about __winrest and __winsave? I presume that
they work similar to __scrsave and __scrrest ...
Msg#8920 To:JON MELTZER (Rcvd)
From:DIRK LESKO Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
I beleive all they do is call __winsave and __winrest....You can call
them directly int he following manner:
temp = space(4000) && allocate buffer size of screen
call __winsave with word(0), word(0), word(24), word(79),;
temp
That will save the screen into the buffer....To restore, just call
call __winrest with temp
And that should restore it. There are many 'goodies' like that. however I
would point at that it is a little easier saying:
save screen to TEMP
Than it is to setup the buffer to call __winsave. My point is that alot
of what you are paying for is there already, only if they would have made
it public....But then they would have to support it so it is probably not
a good idea to publish it anyway, you know they will support save screen
to, but __setwin and __winsave will probably bite the dust int he next
release....
dLESKO
Msg#8921 To:HOWARD KAPUSTEIN (Rcvd)
From:STEVE DAVIES Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
Don't forget that with a club discount you can save about $50.00 on
FUNCky also.
Msg#8922 To:RICH WINTER (Rcvd)
From:MARK NEIDORFF Date:04/04/89
Subject:(R)CLIPPER AND TLINK Parent Msg#8879
------------------------------------------------------------------------------
You can't use tlink 2.0 with clipper. There are 2 .obj file formats.
Intels and microsoft's. Tlink2.0 uses intel's, tlink 1.5 AND clipper use
microsoft'
s. Sorry about that.
<< Mark >>
Msg#8923 To:HOWARD KAPUSTEIN (Rcvd)
From:MARK NEIDORFF Date:04/04/89
Subject:(R)TOOLS ONE Parent Msg#8881
------------------------------------------------------------------------------
That's a good set of questions Howard. Watch for replys!!
<< Mark >>
Msg#8924 To:RON LANDBERG (Rcvd)
From:MARK NEIDORFF Date:04/04/89
Subject:(R)WK1 FILES Parent Msg#8887
------------------------------------------------------------------------------
Sure there is. Its an add on product called d123 put out by
Communication Horizions. Call Neil Weicher at 212 724 0150. It sells
for $149.
I've never used it, but I;ve seen it work and it does the job well.
<< Mark >>
Msg#8925 To:DIRK LESKO (Rcvd)
From:DAVID MORGAN Date:04/05/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
> Well that's just it. The overall outcome is still left hanging.
The explanation of how it is going to work is still very vague.
That's right because today it's all future tense. We haven't got the
"how" yet. It's just the nature of the project's current stage. If
the "how" is revealed in detail, it comes later.
While my bias is to stress the positive, that we have the "what" (no
more memory inadequacy), over the negative (that we don't get
blueprints), you might want to reverse the focus instead: emphasize
the absence of "how" rather than the presence of the "what." "Left
hanging" is one way to put it. That's accurate enough, I think, if
you only mean to point out that implementation details are
unprovided, not if you mean unprovided-but-ought-to-be. The
developers have been careful to a fault, inside the company and out,
to avoid commitments to imperatives to reveal details till they feel
it appropriate (as in the DBA interview). They've given us no
"ought-to-be's" and no great detail on internals. You're right in
saying so.
I think (from talking to a lot of them) that Clipper users' real,
main concern is elimination of memory inadequacy. They're largely
indifferent to which solution or how it works if only it does indeed
work. So, will it? Look at the track record. If we're "left
hanging" today about how future Nantucket products will do what they
do, then we were also hanging over the current one back in early
1987; and similarly its predecessor six months before it got
released. Both versions turned out to be great advances, good pieces
of work. So, if the parallel continues, will be the outcome of
current efforts. Vagueness or outright absence of explanation how
it'll all work doesn't have a lot of to do with its actual
workability.
The satisfaction level side of it all boils down to the face-value
acceptance or doubt of the developers' statements: "The end is going
to be elimination of memory inadequacy; means undetailed." You, I,
and the public believe it or we don't.
> If a dynamic link library under DOS was the way to go then we
would all have bought RTlink a long time ago. That hasn't
happened so I don't see a rosy future for that method until it is
handled by the operating system....
Counterexample, I think (I have never used RTlink): RTlink via a
dynamic linking approach that differs from a PLINK static link, ends
up producing a program whose operational RAM requirement at runtime
is the same, doesn't it? No help. A dynamic linking approach,
differently designed, might also just as well produce, instead, an
operation whose RAM requirement was less. Perhaps RTlink wouldn't
help, but that would. People haven't all bought RTlink a long time
ago because it doesn't offer the thing they're after. Something that
did, different story.
My understanding is that dynamic linking, and any other approach for
Nantucket, is not under consideration for its own sake but for that
of eliminating the peculiar and historical memory inadequacy problems
of DOS. If it can't accomplish that, I doubt we'd be hearing about
it. Certainly it'd be ideal if the OS handled it. (Some indeed do,
like OS/2 and Unix, and under those in the potential multi-platform
Nantucket future, it seems likely those capabilities would be
exploited.) But given that under DOS the OS support isn't there, you
have to "roll-your-own" dynamic linking, as you point out. I think
that's what the developers are talking about. See what emerges.
Msg#8926 To:DIRK LESKO (Rcvd)
From:CONSTANTIN FLORESCU Date:04/05/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
I agree that this dentist might be a good programer...I wander if he is
also a good dentist...
But whatever the case, in this situation Nantucket made a very poor
business decision. I agree and I support free competition because after
all that made this country so great but this is not competition...
I don't know but in this case I picture Nantucket like the parent and all
the other developers as children and I don't understand why the the
parent(a wise one at least) should do that(compete or whatever) to its
children...
<CF>
Msg#8927 To:JON MELTZER (Rcvd)
From:CONSTANTIN FLORESCU Date:04/05/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
Ask the dentist...Jon probably he knows better than Nantucket.
<CF>
Msg#8928 To:DAVID MORGAN
From:DIRK LESKO Date:04/05/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Correct on all terms. Dynamic Linking is just a shared overlay that does
not have to be loaded into ram every time it is called. There is no
easy solution to the memory problem....just ask Lotus who had to resort
to PharLaps extender just to get 3.0 to run in a normal 8088. however,
Dynamic Link Libraries were mentioned as the means by which memory
limitations wouuld be conquered in an interim release. This statement
domes from the last rounds that Phill Usher and Basil made just last
month. So the company has stated that yes, memory limitations would be
lifted, but they also have spilled as to the 'how' it will be done. If it
is really what will be, tyhen we will have to wait and judge for
ourselves.....
dLESKO
Msg#8929 To:CONSTANTIN FLORESCU (Rcvd)
From:DIRK LESKO Date:04/05/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
That parallels a statement made to me about 'Parents that eat their
young' by some well known industry Pundit....
dlESKO
Msg#8930 To:ALL
From:JOE BOOTH Date:04/05/89
Subject:TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
I too receive my ad for tools, and would like to pose some thoughts....
Why didn't Nantucket add the functions into EXTEND.LIB, write additional
documentation, and sell the same product as an upgrade ? By offering the
product as third-party add-on, they are competing with some of the best
programmers out there (Steve, Dirk, Niel, etc)...
What should our response be? Should we treat this as another third
party product? If so, there are substantially better libraries available
and I wouldn't lose any sleep over it....
Will this library face the same upgrade headaches other's had to bear if
Nantuckets changes its internals? If I were to use this library for
windowing (as the ad's say it does), then will that be the syntax for the
long-promised windowing??
Hopefully I won't get thrown off the board for this next opinion, but I
don't see anything wrong with Nantucket selling this product. I do,
however, see the potential for a lot of wrong. If Nantucket starts to
use corporate power, to unfairly compete, then I disagree with their
stance. If Nantucket sells this product just like any other vendor
sell his LIB file, then fine. If Nantucket says that their product
should be bought, because of who they are, not on it's merits, then
I disagree.
Before we go off the deep end, lets find out how Nantucket plans to
compete with the third party vendors. If they plan a fair competition,
then we shouldn't complain. If they plan to compete unfairly, selling
the product as the only approved CLIPPER add-on, bombarding all CLIPPER
users because they have the mailing list, or if they forget the main
thrust of their business and work on improving the tools rather than
the compiler, let's complain loudly.....
I understand the third-party vendors positions, but I am willing to
buy their products, because they appear to be the best LIB's available.
(FUNCKY and the Toolkit are included every time I link). If Nantucket
puts out a product that by far BLOWS away the other libs, I will look
at it. I will also wonder why the company spent manpower and dollars
on this product and not the compiler...
One final question before I sign off.. If this is Tools One, does
this mean other similar products are forthcoming ????
Msg#8932 To:DAVE LOBEL (Rcvd)
From:JOHN THORN Date:04/05/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
Dave: Can't say enough good things about Paradise....you be the judge,
tho....
Msg#8934 To:JOE BOOTH (Rcvd)
From:CLIFF GREEN Date:04/05/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
another question to add to your list: If Nantucket is issuing this set
of Tools _now_, right after "pre-announcing" the pre-release of
not-quite-Pegasus, what does that say about the likelihood of the next
incarnation of Clipper coming out soon? If Nantucket thinks there's a
market for a S87 add-on, doesn't that imply that S89 (or whatever it will
be) is not really as imminent as the pronouncements of the last month
imply?
Msg#8935 To:RON LANDBERG (Rcvd)
From:CHRISTOPHER EDGAR Date:04/05/89
Subject:(R)WK1 FILES Parent Msg#8887
------------------------------------------------------------------------------
Ron,
Check Out Neil's dONEtwoTRHEE library...
--Christopher
Msg#8936 To:JOHN THORN (Rcvd)
From:DAVE LOBEL Date:04/05/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
John,
I finally got a hold of SONY today. They agree that there is a
"HARDWARE" upgrade for my particular monitor. It will cost a bit over
$100.
My question is:
Is it an upgrade of a defect?
We'll see.
YEL-DAVE
Msg#8937 To:CLIFF GREEN (Rcvd)
From:DAVE LOBEL Date:04/05/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
Good point........
Msg#8938 To:DIRK LESKO (Rcvd)
From:CONSTANTIN FLORESCU Date:04/05/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
I like that even better...sounds so close to this case.
<CF>
Msg#8939 To:JOE BOOTH (Rcvd)
From:JOHN THORN Date:04/05/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
Seems to me that, given Nantucket's proven inability to produce a product
on time and without serious bugs, they're looking to improve their
short-range cash flow. With an ultra-thin programming staff, together
with the recent claims made for the next version, they need additional
money to take on more programmers (a.k.a. Basil Homer and the like).
I do not think that Nantucket is stupid enough to think that developers
will flock to a library that is reported to be little more than a
collection of public domain UDFs.
As a Clipper developer since Summer of 85, I've seen the product mature,
become immature, mature again, regress, etc. The Tools issue is another
regression. I concur that they should have updated the EXTEND.LIB (and
fixed up crappy code it contains). Summer of 87's lineage is strewn with
bug reports, anamolies for which there is no work-around, etc. Perhaps
they SHOULD have completely "repaired" their latest release before they
began to market something that's sure to contain >= number of errors.
Msg#8940 To:DAVE LOBEL (Rcvd)
From:JOHN THORN Date:04/05/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
Dave,
It's not a defect! Rather, it is an undocumented feature! Add $100 to
the cost of your Sony and you COULD have had a NEC!
Msg#8942 To:JOHN THORN (Rcvd)
From:DIANE LASK Date:04/05/89
Subject:(R)PROGRAM MAIL Parent Msg#8675
------------------------------------------------------------------------------
John,
Thanks for your comments on Commit. I take a little different
tack when dealing with this situation. I do not hesitate over using
Commit as a rule but I do recommend against using caching systems as we
have had several problems with memory conflicts. My inclination is to
forego the cache and leave the commit. Different point of view I
suppose.
Msg#8943 To:MIGUEL MALDONADO
From:DIANE LASK Date:04/05/89
Subject:(R)COMMIT Parent Msg#8738
------------------------------------------------------------------------------
Hi Miguel. I can assure you that the commit command will flush only to
the dos buffers. DOS 3.2 simply does not have the function call
available for the hard disk write. The DOS COMMIT func6tion was
introduced with the 3.3 version. What you are experiencing is probably
the fact that you have set up your system well and with only a few (ie.
8) DOS buffers. The buffers automaticaly write to the disk when full.
Msg#8944 To:RON LANE (Rcvd)
From:DIANE LASK Date:04/05/89
Subject:(R)DOS4 & CLIPPER Parent Msg#8663
------------------------------------------------------------------------------
Thanks for the tip Ron , I'll keep it in mind.
Msg#8945 To:STEVE STRALEY (Rcvd)
From:DIANE LASK Date:04/05/89
Subject:(R)NANTUCKET Parent Msg#4836
------------------------------------------------------------------------------
Steve, HI!
Guess this thread has gotton so long I'm uncertain as to what
you're referring to when you say 'wrong impression.' Guess thats what
happens when I get to log in only once a week. Anyhow nice hearing from
you.
Msg#8946 To:DIRK LESKO (Rcvd)
From:JAMES CURRAN Date:04/05/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
. Sorry if this is a bit a divergance, but I thought neither
PharLaps extenders, nor 1-2-3 3.0 worked on an 8088, just 80286 + up
Msg#8947 To:JAMES CURRAN (Rcvd)
From:DIRK LESKO Date:04/05/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Whoops....Maybe you're right, I don't seem to recall 100% now that you
mention it....
dLESKO
Msg#8948 To:DIANE LASK
From:DIRK LESKO Date:04/05/89
Subject:HELLO
------------------------------------------------------------------------------
Hi Diane...
Msg#8949 To:JOE BOOTH (Rcvd)
From:RON CALLAHAN Date:04/05/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
My opinion, (for what it's worth), is that anythink Nantucket puts out
should not be add-on lib.s but offcial extensions to the language. In
other words, they would sell them as updates now, but would be included
as official syntax in the next release. If the functions are not to
be included in the next release, then they have no business selling the
product.
Msg#8950 To:JOHN THORN (Rcvd)
From:DAVE LOBEL Date:04/06/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
Just what I wanted to hear at 4:00 in the morning......
YEL-DAVE
Msg#8951 To:ALL
From:STEVE DAHLIN Date:04/06/89
Subject:CORRUPT INDEX? Parent Msg#8951
------------------------------------------------------------------------------
Does anyone know of any quirks in Clipper that would allow you to open
a corrupt index file, only to have the application to give you an
"Open Error" message (DOSERROR #0) when you try to open the very next
index file? When I delete the indexes which are opened prior to the
one that bombs and the system re-indexes them, then it can open the
rest of the index files fine.
Msg#8952 To:JON MELTZER (Rcvd)
From:STEVE STRALEY Date:04/06/89
Subject:(R)THINK TWICE Parent Msg#8685
------------------------------------------------------------------------------
Quick Basic... I think..., but who knows... I'm loosing interest
rapidly...
Msg#8954 To:JOE BOOTH (Rcvd)
From:STEVE STRALEY Date:04/06/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
I get that impression... who knows... maybe Tools Two, UI, a program
generator... but then again, I'm just guessin'
Steve
Msg#8955 To:CLIFF GREEN (Rcvd)
From:JOE BOOTH Date:04/06/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
Nantucket has said that the next release of Clipper, will be compatible
with Tools. Hopefully, that means it will also be compatible with
FUNCKY as well....
Msg#8956 To:STEVE DAHLIN
From:JOE BOOTH Date:04/06/89
Subject:(R)CORRUPT INDEX? Parent Msg#8951
------------------------------------------------------------------------------
CLIPPER does not check the integrity of the index, so what you are
describing can occur routine. (It also doesn't check the integrity
of DBF files for that matter). Dos error 0 is Clipper's way of
saying I give up, so don't look for DOS error 0 anywhere....
Msg#8958 To:JOE BOOTH (Rcvd)
From:CLIFF GREEN Date:04/06/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
a very pertinent point. I hope Clipper2 (Pegasus, whatever) will be
compatible with the various tools purchases I'm planning (and making)
now; if not, I have a real dilemma: stay with the power I'm
accumulating now, or cross over to the promised land (emphasis on the
'promise'). How soon will all this happen? Will there be enough time to
write off/absorb/pass on the expense of the current crop of tools? Will
the transition time (read: time to pinpoint the bugs and problems and
then correct for/work around them) be comparable to that from A86 to S87?
What's the upgrade (presuming it _is_ an upgrade, not a new product with
new product cost buyin) going to cost?
Msg#8959 To:DIRK LESKO (Rcvd)
From:MARK NEIDORFF Date:04/06/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Yep. Best info is that 123 v3.0 is 286+ software.
<< Mark >>
Msg#8962 To:ALL
From:ROGER DONNAY Date:04/07/89
Subject:DCLIP - CLIPPER ZIPPER! Parent Msg#8962
------------------------------------------------------------------------------
** New Product Announcment ** dCLIP - "The CLIPPER ZIPPER!"
For CLIPPER Developers - A complete, interactive development and
runtime system which runs your Clipper applications from a dot
prompt or a runtime engine - without linking! Run applications
at compiled speed or any Clipper command from the dot in
interpretive mode. dCLIP not only speeds up the development cycle
of Clipper applications, but also solves the memory management
problems which cannot be resolved with overlay linking. dCLIP
uses "dynamic linking" technology to run your applications from
your compiled OBJect code or dynamic libraries. Edit your source
code, compile, and run with a "live" Clipper environment, all from
the dot prompt. Includes interactive dBUGGER, report editor,
ASSIST-style database manager, printer manager, and many new
commands and functions which can even be used in your Clipper EXEs.
For more information, download the documentation file ZIPPER1.ZIP
or the complete demo program ZIPPER2.ZIP. If these files cannot
be found on this BBS, call the ClipBoard BBS or our office for a
free demonstration diskette. Our special introductory price for
dCLIP is only $149.00. 30-DAY NO RISK GUARANTEE!!!
DONNAY Software Designs Phone : 714-721-6720
4000 Park Newport, Ste. 401 Fax : 714-721-9495
Newport Beach, CA 92660 ClipBoard BBS : 714-731-3946
Msg#8964 To:MARK NEIDORFF (Rcvd)
From:DIRK LESKO Date:04/07/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Well I stand corrected, they couldn't do it for 8088 period. Another
comment I may make is the Multi-Platform Promise. If I may refer to the
Article in DBA, the goal of Object Clipper is to run unchanged on multi
Platforms, including DOS. Windows, PM, Unix and Macintosh. If I may also
refer to recent articles in PC-Week concerning Mitch Kapor's vision of
his On Technology Company. About 18 months ago his 'Vision' was to
provide a platform for that very purpose. A software layer that would
permit software to run unchanged on the very Platform Nantucket is
eyeing. May I say that Mitch Kapor invested Millions of Dollars nd hired
the top programmers inthe country to make the vision fly. 18 months later
he has abandoned the idea. Why? one very simple reason. Vendor
cooperation, or better said, lack of it. My question is that if after
18 months and millions invested MK is tossing in the towel, are we to
think that Nantucket can pull it off? or should we urge that they really
concentrate on one issue until the technology is proven?
dLESKO
Msg#8966 To:ROGER DONNAY
From:DIRK LESKO Date:04/07/89
Subject:(R)DCLIP - CLIPPER ZIPPER Parent Msg#8962
------------------------------------------------------------------------------
Hi Roger....Glad to see you make it on!
dLESKO
Msg#8967 To:JILL GOHR (Rcvd)
From:DIRK LESKO Date:04/07/89
Subject:(R)SKINNED CATS Parent Msg#8963
------------------------------------------------------------------------------
Hi Jill, was your message sent to the person intended? or did you mean to
send it to Dave Morgan?
dL
Msg#8968 To:ALL
From:RON HOCKEMEIER Date:04/07/89
Subject:GOVT COMPUTER NEWS Parent Msg#8968
------------------------------------------------------------------------------
Remember the recent PC Week issue that reviewed DBMS software... and
didn't mention Clipper.
I just received the April 3, 1989 issue of Government Computer News and
they did the same thing.
The short article includes a lengthy table which list 50 or so DBMS
vendors (Ashton Tate, Wordtech, et al.). Nantucket isn't included.
Interestingly, the short article mentions Clipper/Nantucket...
Ironically, the same issue also contains a quarter page (or so) ad from
Nantucket.
Kinda makes ya wonder...
Be well.
Msg#8969 To:ALL
From:CLIFF GREEN Date:04/07/89
Subject:NETWORK PRGRAMMING Parent Msg#8969
------------------------------------------------------------------------------
Can anyone out there recommend some material for me? I'm looking at one
or two programming projects (I'll do my thing in Clipper) specifically
for running on a lan. I have no lan experience, and don't really know
what to look out for/be careful of (other than the loose notion that
files and records should be locked at appropriate times). I guess my
priorities are:
background info/guidelines
tools
examples
Oh yeah. Anyone want to take a stab at recommending their favorite NOS
or lan configuration?
Msg#8970 To:ALL
From:JOEL SHAPIRO Date:04/07/89
Subject:ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
I don't know if this anomaly has been listed as yet.
In the following code segment the start time line does not execute
until a matching record is found. If there is none, the line will
never execute. This is wrong as the line precedes the loop.
The "GO TOP" seems to be the culprit. Moving GO TOP to just before
the DO WHILE command seems to clear things up.
sf=trim(sf) && search data
* set the filter
popfil = [sf$fcustno.or.sf$fcompany]
set filter to &popfil
go top && sets the filter
*** The following line doesn't execute until a match is found. ***
? time() && start time
do while !eof()
? time() && end time
? fcustno,fcompany
?
skip
enddo
Msg#8971 To:JOEL SHAPIRO (Rcvd)
From:MARK NEIDORFF Date:04/07/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
As I see it, the goto top is finding the
"top of file " with the filter active. In other words, the filtered top
of the dbf. I don't think it is a bug. I think that's the way filters
work.
<< Mark >>
Msg#8972 To:CLIFF GREEN (Rcvd)
From:ROB HANNAH Date:04/07/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#8969
------------------------------------------------------------------------------
I'd say go Netware all the way. And avoid 3COM at all costs. If you
want the smallest size network overhead, go Arcnet. If you want speed go
Ethernet. And if you want neither go Token Ring (they're the least prone
to hardware failiers though...).
As to tools, take a look at Netlib from Neil Weicher.
ROB
Msg#8973 To:MARK NEIDORFF (Rcvd)
From:ROB HANNAH Date:04/07/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
Yeah, but ? time() should still work...
ROB
Msg#8974 To:ROB HANNAH (Rcvd)
From:DIRK LESKO Date:04/07/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
Yes, but it will wait till the filter condition is met....
dLESKO
Msg#8976 To:RON CALLAHAN (Rcvd)
From:GREG MARTIN Date:04/07/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Ron,
I have to stick up for Nantucket on their intentions for their next
release of Clipper. I feel that they are going the right direction
for the future. As I have become more and more of a power user, I
have had to admit that Clipper by itself is just not powerful enough
for many of the things I want to do and because of the way it's built
there is no way to give it that power. I to don't want to have to
rewrite all of my code and I believe that the Nantucket developers
understand that. In my opinion, when the next release of Clipper
comes out (next year?) you'll be able to use your previous code with
(almost) no changes - but you won't be able to take advantage of some
of the sheer power you can have access to. I believe your existing
programs will benefit greatly from the new way that Clipper will work
and that we'll all be blown away by all the new things we can do. I
think that Nantucket is doing the right thing. Maybe they need to be
even more open about what they're doing. IF THEY DO IT RIGHT, I
think we'll all be giving Nantucket congratulations next year.
Greg Martin
Msg#8978 To:DIANE LASK
From:JOHN THORN Date:04/07/89
Subject:(R)PROGRAM MAIL Parent Msg#8675
------------------------------------------------------------------------------
Diane:
Forgoing the cache is OK, provided all the client uses his/her computer
for is code that I write. I've never had difficulty with caching as long
as I set E0 parameter. I view the need to do this as a Clipper bug. A
properly written extended/EMS access will see what's already using the
RAM before trying to occupy the same space. (Hint, hint....next version
should have an expanded parameter setting which lets us specifically
define start/end RAM locations!)
John
Msg#8979 To:RON CALLAHAN (Rcvd)
From:JOHN THORN Date:04/07/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
I concur, wholeheartedly! Tools should be an upgrade, but then, what car
manufacturer sells a vehicle without a jack?
Msg#8980 To:DAVE LOBEL (Rcvd)
From:JOHN THORN Date:04/07/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
Sorry, Dave, couldn't resist the temptation to rib ya!
Msg#8981 To:STEVE STRALEY (Rcvd)
From:HOWARD KAPUSTEIN Date:04/07/89
Subject:(R)TOOLS ONE Parent Msg#8981
------------------------------------------------------------------------------
Greed, one man... Who be he? The president of Clipper? I'm not too
up-to-date regarding personalities in Nantucket. BTW, congratulations, I
saw your book in the bookstore the other day and I thought the name was
familar.
Msg#8982 To:DIRK LESKO (Rcvd)
From:HOWARD KAPUSTEIN Date:04/07/89
Subject:(R)CLIPPER AND TLINK Parent Msg#8879
------------------------------------------------------------------------------
Well, technically they did. TC 2.0 supports the Intel .OBJ specs, I think
they got the idea that the Intel specs are right, and why should they
follow MS, they're just another software company, right? It's good and
bad. They lack the full MS compatibility, but what they're saying is MS'
format is a subset, not the standard. Whether this attitude will work or
not is beyond me.
Msg#8983 To:CLIFF GREEN (Rcvd)
From:RON CALLAHAN Date:04/07/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#8983
------------------------------------------------------------------------------
Cliff:
As far as Lans go on the high end I like 3 Comm's 3 Share system.
It appears to be a bit easier to work with that Novell. It also has
fantastic response time. On the lower end I like to use DNA's system.
It's great for small installations with the proper hardware. (high speed
disks with 286 & 386 systems.) I use it for 5 to 8 user situations who
don't plan to grow much in the near future. It installs in 15 minutes and
is very easy for novice operators to use. You will definitly want Neil
Weicher's Netlib system. It's outstanding for network applications..
Ron
Msg#8984 To:ROB HANNAH (Rcvd)
From:RON CALLAHAN Date:04/07/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#8969
------------------------------------------------------------------------------
Rob.
I'm curious. What problems have you had with 3COM. I've used it and
Netware for the past couple of years. Performance wise I haven't seen
much difference. Is there something I should be looking out for?
Msg#8985 To:GREG MARTIN
From:RON CALLAHAN Date:04/07/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
Greg.
If your right, I couldn't agree with you more. I'm all for progress.
It's just that I've been burned a few times. Let hope for the best.
Ron.
Msg#8986 To:DAVID MORGAN (Rcvd)
From:HOWARD KAPUSTEIN Date:04/07/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
From a bit of observation, I would just like to drop a couple of remarks.
They are of a purely personaly basis, but I'm sure others out there have
had some of these thoughts too.
Nantucket is promising us "Dynamic Link Libraries" to reduce the memory
overhead, a critical bottleneck for Clipper applications. No more
overlays, no more 1 line program requiring 200K+ to run. All is good.
But... to me, it sounds like an OS/2 version of Clipper is coming soon,
and Nantucket is trying to implement DLLs under DOS (as opposed to the
operating system). What I'm concerned with is Nantucket's commitment to
DOS vs. OS/2. Yes, OS/2 is the future operating system. But due to high
hardware requirements (and, thus, $$$) its acceptance is still a ways
off. DOS is around now, it works, and it will still be here a few years
from now. It's good to see Nantucket looking to the future, but I am much
more concerned with solving the current problems. What concerns me is:
will Nantucket be "passing the buck" by promoting OS/2, with DOS support
as secondary?
This may sound anti-OS/2, but it's rather realistic. How many clients
want to purchase an OS/2 system just to run a (comparatively) simple
inventory system, or sales, or... Face it, many applicaitons don't NEED
all the bells and whistles. Interapplication communications, etc. are
nice, but not really required when an old-fashioned PC with 512K and a
hard drive can do just as well. Hell, how many people run Clipper
applications in color? OS/2 doesn't support anything less than EGA.
Please make any noises you feel appropriate.
Msg#8987 To:CLIFF GREEN (Rcvd)
From:HOWARD KAPUSTEIN Date:04/08/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
Or doesn't it imply that the next Clipper version won't be fully
functional under all operating systems (DOS).
Msg#8988 To:JILL GOHR
From:HOWARD KAPUSTEIN Date:04/08/89
Subject:(R)SKINNED CATS Parent Msg#8963
------------------------------------------------------------------------------
"Any given program will expand to fill all available memory."
- Laws of Computer Programming
Msg#8989 To:ALL
From:HOWARD KAPUSTEIN Date:04/08/89
Subject:DBASE HEADER FORMAT Parent Msg#8989
------------------------------------------------------------------------------
I've got a quick, but important question. I'm trying to implement some
functions in C to access dbase files. The header of a dBASE III+ file is
supposed to have unsigned char Length, which is the length of a field.
Well, that gives dBASE C fields a length of 255 max. According to the
Clipper specs, Clipper can handle C fields up to 64K. My question is,
how? Here's part of the structure for a field:
struct FIELD {
...
unsigned char length;
unsigned char decimal;
unsigned char mutli_user[2];
...
};
Does Clipper assume that a C field has no decimal point, and the length
and decimal variables can be used as a single unsigned int (2 bytes,
0-65535)? I would guess this is how Clipper does it, but would like some
verification if anyone actually knows. Thanks.
Msg#8991 To:CLIFF GREEN (Rcvd)
From:STEVE STRALEY Date:04/08/89
Subject:(R)TOOLS ONE Parent Msg#8930
------------------------------------------------------------------------------
One of the things I hope is that my ToolkiT will be compatable, cause if
it ain't, then we all will be waiting for the great nothing.
Steve
Msg#8992 To:GREG MARTIN
From:STEVE STRALEY Date:04/08/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
The words "right" direction has been tossed about without any
consideration to the present direction. Additionally, if the future
direction ties us into a language that 3 developers are working on
without ANY assurance of 100% compatability in a time frame which is, for
lack of a better term, lightning makes me highly skepticle. I don't mean
to take anything away from them and I hope they can pull it off, but
Dirk's point is well taken. Additionally, I know there are plenty of
talented and knowledgable people in the east and collectively, they say
the Pegasus project is impossible in the given time frame with the
limited resources.
Greg, I believe in Clipper and as a believe, logic can't apply, but my
"believes" are no longer on solid ground... I just don't think it all can
be done in the time frame given, let along being done right....
Steve
Msg#8993 To:HOWARD KAPUSTEIN (Rcvd)
From:STEVE STRALEY Date:04/08/89
Subject:(R)TOOLS ONE Parent Msg#8981
------------------------------------------------------------------------------
I'm not sure... I just don't see the reason for introducing a set of
tools that is not already present in the industry. Again, I like to look
at things logically and if I see no reason other than cash-flow, then
I'll call it that way. Taking a look at some of the other comments here,
I'd have to think most people agree.
As per the book, ya... you should be seeing a NEW book out there as well
soon. Thanks...
Steve
Msg#8994 To:STEVE STRALEY (Rcvd)
From:RON CALLAHAN Date:04/08/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
The point is well taken. Consider Mr. Homers estimates about UI2.
What was worse, every time Wallsoft released a version, it wasn't
compatible with previous versions.
Msg#8995 To:DIRK LESKO (Rcvd)
From:JILL GOHR Date:04/08/89
Subject:(R)SKINNED CATS Parent Msg#8963
------------------------------------------------------------------------------
No and yes! Thanks for pointing this out. If there is a Walt Morgan,
I apologize for not paying closer attention. I was a "little" hot
under the collar ... and still am. Will kill msg and resend.
Msg#8998 To:JOHN THORN (Rcvd)
From:DIRK LESKO Date:04/08/89
Subject:(R)THINK TWICE Parent Msg#8998
------------------------------------------------------------------------------
Good stuff!
dLESKO
Msg#8999 To:HOWARD KAPUSTEIN (Rcvd)
From:DIRK LESKO Date:04/08/89
Subject:(R)DBASE HEADER FORMAT Parent Msg#8989
------------------------------------------------------------------------------
Clipper Sticks the overflow into the decimals portion of the field
description...Don't have the exact calc but I cam get it if you need
it...
dLESKO
Msg#9000 To:JILL GOHR (Rcvd)
From:DIRK LESKO Date:04/08/89
Subject:(R)SKINNED CATS Parent Msg#8963
------------------------------------------------------------------------------
No Problem....I was just wondering if he would figure it out.....You
replied to a message sent to him by someone else.....I don't think you
need to kill the message, just let him know it was intended for someone
else...
dLESKO
Msg#9002 To:ALL
From:MARC SCHNAPP Date:04/08/89
Subject:APPEND VS. LOW-LEVEL I O Parent Msg#9002
------------------------------------------------------------------------------
What's faster APPEND FROM <> SDF or using the low-level functions within
a loop (FOPEN(), FREAD(), FWRITE())?
I'm appending about 7-9K recs and the performance is disappointing-- has
anyone tried to do that? Samples of code and comments would be
appreciated.
Msg#9003 To:MARC SCHNAPP (Rcvd)
From:NEIL WEICHER Date:04/08/89
Subject:(R)APPEND VS. LOW-LEVEL I Parent Msg#9002
------------------------------------------------------------------------------
Mark:
The APPEND FROM SDF will be just as fast as anything. You can use DB3C
to make it go faster, it just depends whether its worth it to you.
Also: if you can have the target file (the one you're appending into)
open in exclusive mode it will go faster.
Neil
Msg#9005 To:JOHN THORN (Rcvd)
From:DAVE LOBEL Date:04/08/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
John,
No problem, I'll just owe ya one. Anyway, I brought my machine into
computer factory today to hook it up to an NEC 3D monitor. It looks
great bu I think there is something wrong with my card. I am going to
return it, but should I down grade to the Vega VGA card, try another
Fastwrite, or since I'm spending the money, go for the gusto and get the
VRAM VGA? Or, just say the heck with it and go Paradise??????
What do you think (or anyone else)??????
Thanks for any input.....
YEL-DAVE
Msg#9006 To:MARC SCHNAPP (Rcvd)
From:DAVE LOBEL Date:04/08/89
Subject:(R)APPEND VS. LOW-LEVEL I Parent Msg#9002
------------------------------------------------------------------------------
Marc,
If I remember correctly, in one of the steering committe meetings awhile
back, that was brought up. I think with Summer '87 it didn't make a
difference, except for easier programming (APPEND FROM <> SDF).
Haven't spoke in a while, give me a ring when you have some free time.
good luck......
YEL-DAVE
Msg#9007 To:ROB HANNAH (Rcvd)
From:MARK NEIDORFF Date:04/08/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
Not until the file pointer has found the top of the database according to
the filter condition. See what I mean?
<< Mark >>
Msg#9008 To:HOWARD KAPUSTEIN (Rcvd)
From:MARK NEIDORFF Date:04/08/89
Subject:(R)CLIPPER AND TLINK Parent Msg#8879
------------------------------------------------------------------------------
No. I think that Borland considers MS the competition and so will go to
lengths to avoid supporting their products when possible. (This is in
regard to the .obj file format debate brought about by tlink 2.0)
<< Mark >>
Msg#9012 To:DIRK LESKO (Rcvd)
From:RON LANDBERG Date:04/09/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
Why?
Msg#9013 To:RON CALLAHAN (Rcvd)
From:CLIFF GREEN Date:04/09/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#8983
------------------------------------------------------------------------------
actually, I'm thinking of roughly three phases: 1) file server and two
(maybe a third) nodes as a pilot, 2) add about 5 more nodes in the same
building, but not all on the same floor, and 3) add three or four more
nodes in different buildings. Phase 1 looks like it's going to happen,
2 depends on its success, and 3 on that of 1 & 2. There's been some talk
about adding a bridge to another, larger network, but I'm not going to
worry about that yet. Have you run into any problems with distance
between nodes? Have you had any experience in starting with one LAN,
then switching to another after the system "grew up"?
Msg#9016 To:ALL
From:JOEL SHAPIRO Date:04/09/89
Subject:ANOMALY
------------------------------------------------------------------------------
Thanks all for your comments. Anomaly or not, I think it's strange the DO
WHILE loop should appear to be self-executing because a filter condition
isn't met; especially when lines before the ENDDO don't attempt to use
database information.
Oh well, we learned something new.
Joel
Msg#9018 To:ALL
From:RON LANE Date:04/09/89
Subject:DOS4 TIP2
------------------------------------------------------------------------------
Well another one for the list of things that want work on
MS-DOS 4.01 and CLIPPER
clea
fiinfo = "Hello World"
s_dri = "A:"
fi = FCREATE("&s_dri.test",0)
FWRITE(fi,fiinfo,11)
FCLOSE(fi)
QUIT
Gives this error msg: If the disk has a write protect tab.
Exception # 13 at 02E6:7321
Error: 0000
Do you want to T)erminate the program, R)eboot, or try to C)ontinue?
What happened to: Write protect error writing drive A
Abort,Retry,Fail ?
Also this happens before any ERRORLEVEL check works. IE:
fi = FCREATE("&s_dri.test",0)
IF fi > 0
FWRITE(fi,fiinfo,11)
FCLOSE(fi)
ELSE
RETU
ENDI
Same error msg.
Dos 3.3 is looking better all the time.
Ron Lane
Msg#9019 To:DIRK LESKO (Rcvd)
From:JOHN THORN Date:04/09/89
Subject:(R)THINK TWICE Parent Msg#8998
------------------------------------------------------------------------------
Just tryin to do my "bit" for the cause.....
John
Msg#9020 To:DAVE LOBEL (Rcvd)
From:JOHN THORN Date:04/09/89
Subject:(R)VGA .AND. SONY Parent Msg#8852
------------------------------------------------------------------------------
Dave:
For MY! money, Paradise has it all over Video 7 folks....sorry, found
some code that accessed IBM specific registers and Paradise was able to
handle it - the Video7 VGA crapped out. I've also a friend with a NEC
and his Video7 VGA displays a nice crisp line down the screen. Video7
techies said it was his monitor....oh, yea, then why doesn't appear when
using a Paradise, ATI....???
John
Msg#9021 To:CLIFF GREEN (Rcvd)
From:RON CALLAHAN Date:04/09/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#9021
------------------------------------------------------------------------------
Cliff:
The only thing I'm doing over any distance at the moment is to use
a program such as PC Anywhere to dial up a station at the remote site.
This is not very efficiant because it takes an unused station at the
remote site to dial up to. Unfortunately this is the only way to go
with most networks.(It's also slow depending upon the baud rate of
your modems.) One advantage of PC Anywhere compareed to other programs
such as Carbon Copy, is that you can call in with a dumb terminal instead
of another PC running another copy of Carbon Copy.
You might want to check with the makers of the 10 Net Network system.
At one time they advertised the ability to dial up and connect directly
to the server without going through another station.
If the above sounds confusing let me know and I'll try to restate it.
Ron.
Msg#9024 To:RON LANDBERG (Rcvd)
From:DIRK LESKO Date:04/10/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
Because the TOP of the file with a filter is whenever a record is found
that matches the filter. That could mean a run through the entire file
before a condition is found, or not found...
dLESKO
Msg#9026 To:HOWARD KAPUSTEIN (Rcvd)
From:JOHN NEWTON Date:04/10/89
Subject:(R)CLIPPER AND TLINK Parent Msg#8879
------------------------------------------------------------------------------
Who has the bigger stick??
Msg#9027 To:HOWARD KAPUSTEIN (Rcvd)
From:JOHN NEWTON Date:04/10/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
I still don't understand why everyone is so hoped up over OS/2. I don't
need an operating system thats going to spend 30% to 40% of its time
drawing icons. It's no problem reading text, in fact I don't have to
figuer out what some little picture is suppose to be. The bottom line is
that the majority of programs in use today - and from what I can see for
tomorrow - will be text based/driven. For this reason I hope Nantucket
goes with the DOS version first then OS/2.
Of course I hope they continue their development in the multi-platform
area, particularly UNIX. The Army is moving steadly towards UNIX, and
with SCO and Interactive both releasing System V along with the DOS shell
system from Interative, you can now run multiple DOS applications on a
UNIX machine anlong with you UNIX applications. While my reason are a
little selfish, I don hope Nantucket gets a UNIXversion out in a year or
two.
-John Newton
Msg#9029 To:RON CALLAHAN (Rcvd)
From:ROB HANNAH Date:04/10/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#8969
------------------------------------------------------------------------------
We've been programming primarily in Netware - there's alot of stuff we do
that is kinda hard to replicate in 3COM. Like finding out the physical
and logical station number, who has what files open / locked. Not to
mention the superior security features under Netware. We've also seen as
much as a 15% performance degradation under 3COM (which isn't really
saying much as the hardware environments weren't the same).
Probably the worst thing with 3COM is the tremendous amount of memory
overhead the drivers take.
ROB
Msg#9030 To:MARK NEIDORFF (Rcvd)
From:ROB HANNAH Date:04/10/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
That sucks...
ROB
Msg#9031 To:DIRK LESKO (Rcvd)
From:ROB HANNAH Date:04/10/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
Are u saying that the 'goto top' is simply pausing the program execution
until the filter condition is met, then allowing the first '? time()' to
execute?
ROB
Msg#9032 To:MARC SCHNAPP
From:RON HOCKEMEIER Date:04/10/89
Subject:(R)APPEND VS. LOW-LEVEL I Parent Msg#9032
------------------------------------------------------------------------------
Hi Marc. One of my applications does the equivalent of "appe from" all
the time. (Source records come from an Optical Mark Reader [fill-in the
circle with a number two pencil type of device.])
There might be one record or 600 records.
I use low level file open and pop the data into an array. I wrote the
code just after S'87 came out so I no longer remember the time
differences. I "feel" there was a 30 to 50% time savings. A good part
of the savings is due to use of an array (rather than a disc file) to
temporarily hold the data.
I uploaded that portion of the code. The file is named "markfrom.ron."
The code was written in a hurry to solev a CRISIS in early '87. Since
then, I've added the low level/array business and IDL library. (I
haven't had a chance to smmoth the code and upgrade to Funcky.)
BTW the variable "timer" is the processing time per record.
(I time a task so the software can give a user the est. time a job will
be done.) With dBxl on a bare bones PC. it took 4.1 seconds to process
each record. With Clipper, low level/array, IDL.lib on a 386, etc. it takes
0.02 seconds to process a record.
Be well.
Msg#9033 To:RON CALLAHAN (Rcvd)
From:CLIFF GREEN Date:04/10/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#9021
------------------------------------------------------------------------------
no, that's not confusing. In fact, one of the things I wanted to explore
was calling in from another site via modem to a modem on the server.
Hopefully, that would put any activity through the modem under
application control, which something like pcAnywhere (which I would want
for different, support reasons) would not. (muddied syntax?)
Msg#9034 To:ROB HANNAH (Rcvd)
From:CLIFF GREEN Date:04/10/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#8969
------------------------------------------------------------------------------
I believe Netware only provides you with the NOS; what kind of hardware
(cards, cabling, anything else?) are you using for your lan(s)?
Msg#9036 To:ROB HANNAH (Rcvd)
From:DIRK LESKO Date:04/10/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
That's what should be happening. If the file is rather large it will scan
the entire file before realizing that the filter was not met. That could
take minutes depending on machine speed and DBF size....
dLESKO
Msg#9037 To:ROB HANNAH (Rcvd)
From:MARK NEIDORFF Date:04/10/89
Subject:(R)ANOMALY Parent Msg#8970
------------------------------------------------------------------------------
Not really. Remember that you are moving the file pointer when you issue
the go top. If you don't need the filter yet, why move the pointer?
Yes, filters are slow and need to be redone. I gripe about that too.
<< Mark >>
Msg#9038 To:RON HOCKEMEIER (Rcvd)
From:MARK NEIDORFF Date:04/10/89
Subject:(R)APPEND VS. LOW-LEVEL I Parent Msg#9032
------------------------------------------------------------------------------
Your PC to 386 comparison is an apples to oranges (er diamonds to boats
that is. So I don't get sued by you know who). Try the test on the same
machine and the difference isn't so big.
<< Mark >>
Msg#9039 To:DIRK LESKO (Rcvd)
From:HOWARD KAPUSTEIN Date:04/10/89
Subject:(R)DBASE HEADER FORMAT Parent Msg#8989
------------------------------------------------------------------------------
Are the bytes in low:high order? So 0xF0AA would be stored as AAF0 ? I
think that's how words (2-byte values) are stored by DOS.
Msg#9041 To:JOHN NEWTON (Rcvd)
From:HOWARD KAPUSTEIN Date:04/10/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
I agree, partly. I've worked and programmed with the Mac and Sun
workstations in addition to DOS, DESQview and Windows (no OS/2, yet) and
I can honestly say that the GUI (Graphic User Interface) makes things
more intuitive, and hence easier to use. Compare Quicken on the PC and
the Mac. There is no comparison. The problem is because you only get 25
lines of display in text mode with DOS (without resorting to unsual
modes, 43, 50, etc.) and when deal with pixels instead of lines you get
more control. It's a bigger pain in the a** to line things up where you
want them, but the finished product is better.
I agree that OS/2 is a nice idea, maybe, but practically speaking most
Clipper programmers are more concerned with DOS support in the short to
medium term future (1-3 years) and would rather wait for the OS/2 version
than the DOS version. I just hope Nantucket has the same emphasis.
Msg#9043 To:MARK NEIDORFF (Rcvd)
From:STEVE STRALEY Date:04/11/89
Subject:(R)TOOLS ONE
------------------------------------------------------------------------------
No, this one is one Advanced Programming with introduction of C with some
enhanced library routines... like, a TRUE and WORKING dbedit()... The
networking book has taken a side-line since I haven't gotten together
with Neil to work out the details... I guess both of us are swamped
right now....
Steve
Msg#9044 To:HOWARD KAPUSTEIN (Rcvd)
From:DAVID MORGAN Date:04/11/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#8761
------------------------------------------------------------------------------
> to me, it sounds like ... Nantucket is trying to implement DLLs
under DOS .... What I'm concerned with is Nantucket's commitment
to DOS vs. OS/2. ...will Nantucket be "passing the buck" by
promoting OS/2, with DOS support as secondary?
An intention to move toward dynamic linking doesn't necessarily
amount to an intention to move toward OS/2, right? DOS (unlike OS/2)
doesn't have it built-in. So under DOS you have to "roll-your-own"
dynamic linking, if you want to have it. I think that's what the
developers are talking about doing. Nantucket published a press
release about "future stuff" about a month ago. It can be downloaded
from Nanforum on CompuServe. In the "multi-platform" discussion I
think it says something about platform prioritization and my
impression is that DOS got first emphasis.
Msg#9046 To:HOWARD KAPUSTEIN (Rcvd)
From:DIRK LESKO Date:04/11/89
Subject:(R)DBASE HEADER FORMAT Parent Msg#8989
------------------------------------------------------------------------------
Yes, that is correct. However, DOS doesn't store it in that format...C
does. A LONG is also in the same order...
dLESKO
Msg#9048 To:MARK NEIDORFF (Rcvd)
From:RON HOCKEMEIER Date:04/11/89
Subject:(R)APPEND VS. LOW-LEVEL I Parent Msg#9032
------------------------------------------------------------------------------
Mark: Of course. I didn't log all the changes in the timing as the
software evolved. I just wanted to show the differences in incremental
improvements. Moving from dBxl vers 1.1 to Clipper A '86 meant a change
from 4.1 secs to 1.1 secs on the same machine. After that, the changes
were very fast and I no longer have my notes on timing. That's why I
qualified my perceptions and indicated the low level/array probably
resulted in a 30% to 50% time savings in my particular application.
Be well.
Msg#9052 To:ALL
From:MIGUEL MALDONADO Date:04/12/89
Subject:LIBRARIES Parent Msg#9052
------------------------------------------------------------------------------
Does anyone know if the functions in the Clipper Libraries can be access
from a C program?
Msg#9053 To:JOHN NEWTON (Rcvd)
From:MIGUEL MALDONADO Date:04/12/89
Subject:NEW CLIPPER RUMORS Parent Msg#9053
------------------------------------------------------------------------------
Hi John,
I too do not understand why everyone is hoped up over OS/2. I was hoping
Nantucket would come out with a version for UNIX, but the last meeting of
the NY Clipper User Group, Nantucket said it would not be coming out with
it, because they said that there is no proof that a UNIX standard exist!
.
Wordtech Systems announced that they have release QUICKSILVER, their
dBASE compiler, for UNIX/386!
Doesn't Wordtech know that there is no UNIX standard!? I guess they want
to waste their money.
.
Nantucket wants facts that there is a standard and that it is going to be
the market choice. I would like to send them some. If you have any ideas,
let me know see if we could get them some. I like Clipper, creating a
multi-platform system is good, but to leave out UNIX, that's a sin.
Msg#9054 To:MIGUEL MALDONADO
From:RON CALLAHAN Date:04/12/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#9053
------------------------------------------------------------------------------
I really don't see what the problem is. The folks at Progess release
a version of their software for nearly every version of Unix in
existance. Believe me, PROGRESS is much larger and complicated system
then Clipper.
Msg#9057 To:ALL
From:SCOTT VANOVER Date:04/12/89
Subject:DBF'S WITH MEMO FIELDS
------------------------------------------------------------------------------
Does anyone have an easy way to update dbf file structures when the dbf
contains a memo field? Currently I use modify structure but I now have
half dozen installations and everytime that I do an update I must
manually update the file structure for each client. Appending the old
data to a new database works great if there is no memo field? Any tricks
or suggestions will be appreciated! Thanks -SPV-
Msg#9058 To:RON CALLAHAN (Rcvd)
From:MARK NEIDORFF Date:04/12/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#9053
------------------------------------------------------------------------------
I think part of the problem is that clipper is written very DOS
specific and would have to be rewritten to be ported.
<< Mark >>
Msg#9059 To:DIRK LESKO (Rcvd)
From:HOWARD KAPUSTEIN Date:04/12/89
Subject:(R)DBASE HEADER FORMAT Parent Msg#8989
------------------------------------------------------------------------------
Ah, that's right. My oversight. Is a long reversed by words or bytes
(FF112233 would be 11FF3322 or 2233FF11 ?)
Msg#9060 To:MIGUEL MALDONADO
From:HOWARD KAPUSTEIN Date:04/12/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#9053
------------------------------------------------------------------------------
Perhaps Wordtech is willing, since they've got ties to Oracle...
Msg#9062 To:MIGUEL MALDONADO (Rcvd)
From:JOHN NEWTON Date:04/13/89
Subject:(R)NEW CLIPPER RUMORS Parent Msg#9053
------------------------------------------------------------------------------
From what I've seen (in my limited research) it looks like both SCO and
Interactive have a fairly "standard" UNIX implementation that conforms to
System V guidelines.
.
Granted UNIX does not have a single source like DOS, but that is what
makes it nice (or a real headache - depending). The purpose of UNIX was
to allow several different Value Added Feature to be provided on top of a
"standard". All Namtucket has to do is address the "Stnadard" unix
system calls. Perhaps make use of the new X-Windows "standard" - Sounds
simple enough - just glad I'm not a systems programmer trying to
implement it -grin.
.
You may have noticed the quoted "STANDARD". Anyone who has worked on
several differnt manufactures mainframes in COBOL will know that there is
IBM "Standard" COBOL, UNIVAC "Standard" COBOL, and so on and so on
and....
.
I do howvere hope that Nantucket continues towards the UNIX environment
as I think that it will open a rather large market to them plus make my
life a whole lot easier.
-John
Msg#9063 To:ALL
From:RON HOCKEMEIER Date:04/13/89
Subject:INFOWORLD Parent Msg#9063
------------------------------------------------------------------------------
The latest issue of Infoworld:
* Mentions the dBase IV compiler is now scheduled to ship in second half
of 1989. (Surprise, surprise.)
* Reviews several multi-users DBMS including dBase IV [but omitting
Clipper .and. FoxBase]. In essence, dBase IV is not recommended for a
multi-user application. (Surprise, surprise.)
Msg#9067 To:ALL
From:CHRISTOPHER EDGAR Date:04/14/89
Subject:DOS UPGRADE
------------------------------------------------------------------------------
I need some help & information upgrading a clients WYSE AT compatable
from DOS 3.1 to version 3.3. I think I need to do this for a CLIPPER
application I am writing (network app) because I remember reading about
the difference in the way th two versions differ handeling buffers.
Could someone please tell me what exactly Äis the difference between the
two versions of DOS and if i should upgrade this client to 3.3?
If so, does this require re-formating the hard disk (30 ßmeg), or is a
"SYS C:" and a replacement of DOS files sufficient? Any Help is
appreciatedß.
--Christopher
Oh yeah...How much more memory does 3.3 eat?
Msg#9070 To:RON HOCKEMEIER (Rcvd)
From:JOHN THORN Date:04/15/89
Subject:(R)INFOWORLD Parent Msg#9063
------------------------------------------------------------------------------
Yea, InfoWorld recommends Paradox 3.0 - then DBMS Magazine turns around
and tells you that there's a bug in Paradox that will cause you to lose
your data!
Msg#9071 To:ALL
From:ROBERT HEYMAN Date:04/15/89
Subject:DATA COMM.
------------------------------------------------------------------------------
Can Anyone out there reccommend a book that explains data communications
well enough to allow me to use SilverComm.
Thanks
Rob
Msg#9072 To:CLIFF GREEN (Rcvd)
From:JIM KILLEEN Date:04/15/89
Subject:CLIPOVLS Parent Msg#9072
------------------------------------------------------------------------------
Cliff,
Let me thank you for the great amount of help I got from the clipovls
file I came across. It was just what I needed when I needed it! I've
been experimenting with a lot of PCL stuff (I don't have the Tech Ref.
yet) and a lot of it has been trial and error. You saved me a lot of
time with your explanation of the macro commands. I was buliding forms
for an accountant to use, but without the overlay macro I would have to
have them load a file of escape sequences before each page was printed,
not a very eloquent solution. Now I'm setting up a menu of forms to be
loaded before he does a job. Once again, thanks.
Jim K.
Msg#9073 To:JIM KILLEEN (Rcvd)
From:CLIFF GREEN Date:04/15/89
Subject:(R)CLIPOVLS Parent Msg#9072
------------------------------------------------------------------------------
my pleasure.
Msg#9077 To:MARK NEIDORFF (Rcvd)
From:TERRY DILLARD Date:04/15/89
Subject:(R)WHY AN OPEN ERROR?
------------------------------------------------------------------------------
Thanks to everyone who offered suggestions to my "open files" problem. I
had files and buffers set correctly in my config.sys, and counted exactly
15 files open including the .EXE file. Still no luck. Fortunately, one
of the databases was small enough to change to a CASE statement and kill
the .DBF and associated .NTX file, so know I have no more problem.
Thanks for your help.
Msg#9078 To:ALL
From:RICK RAMHAP Date:04/15/89
Subject:JOB OPENINGS!
------------------------------------------------------------------------------
Positions Available: (new as of 4/13/89!)
Long Island-based software development firm currently has openings
for experienced Clipper programmers. Applicants should be familiar
with multi-user system development. Proficiency in C is advantagious.
Both permanent and temporary positions are available.
Call Brian Young at 516-889-9236 for further information.
Msg#9080 To:HOWARD KAPUSTEIN (Rcvd)
From:DIRK LESKO Date:04/15/89
Subject:(R)DBASE HEADER FORMAT Parent Msg#8989
------------------------------------------------------------------------------
A long that is FF112233 would be stored out as 332211FF. The low
byte/high/byte works on the 4 bytes in 2 byte chuncks, and then on those
2 byte chuncks as well....
dLESKO
Msg#9081 To:ALL
From:ALAN GUGGENHEIM Date:04/15/89
Subject:MULTI-USER CLIPPER Parent Msg#9081
------------------------------------------------------------------------------
QUESTION: How do people run large clipper applications on a
network, when the networking software takes up 50 - 150k of overhead?
Is there any way to break the 640k barrier on DOS when using a
network? (assume that I have overlayed and optimized the code of
the application). It seems that networks take away more of the
precious memory than some applications can afford.
Has anyone heard of 10-net? Is it compatible with clipper?
What is its overhead in kilobytes (K) for a station? Is it compatible
with NETLIB (if so, what library)?
Thanks in advance...
Msg#9084 To:MIGUEL MALDONADO (Rcvd)
From:DIRK LESKO Date:04/15/89
Subject:(R)LIBRARIES Parent Msg#9052
------------------------------------------------------------------------------
Yes, if you know how....
dLESKO
Msg#9086 To:ALAN GUGGENHEIM (Rcvd)
From:DIRK LESKO Date:04/16/89
Subject:(R)MULTI-USER CLIPPER Parent Msg#9081
------------------------------------------------------------------------------
Well, if you have a 386 you can use Quarterdecks 386 to the max to get an
extra 100-200k by putting all the network software up in expanded memory.
dlEKSO
Msg#9088 To:ALL
From:JOSEPH CATTANO Date:04/16/89
Subject:CLIP VS DB3 DBFS
------------------------------------------------------------------------------
Would anyone know if there is an incompatability between a .DBF
file created under dBASE III Plus and one created under Clipper?
I have an application that i wrote in dBASE and compiled with
Clipper. The Clipper version works, except that SEEKs through an index
(.NTX) file always come up "not found" after a record is added or
changed. This problem disappears when the .NTX files are deleted and
re-created, but returns if a change is made to a record. The problem
disappears completely if the .DBF file is created with the Clipper DBU
utility and the records are copied in from the dBASE created file.
The only difference i can find is that if you creat two .DBF
files (one with dBASE and one with Clipper) with the same format and
no records, the Clipper .DBF will be one byte longer and one byte will
be different.
The dBASE .DBF ends in 0D1A, whereas the Clipper .DBF ends in
OD001A, 00 being the extra byte. The differing byte is at offset 08.
In dBASE it is C2, in Clipper it is C1.
I've noticed other differences when you start to add records to
the files, but since i don't know the format of the file header i'm
not sure if these matter.
I'm new at dBASE & Clipper, so any ideas or comments will be
greatly apperciated.
Thanks in advance,
joseph
Msg#9089 To:ALL
From:NEIL WEICHER Date:04/16/89
Subject:NETLIB NOTICES
------------------------------------------------------------------------------
NetLib Notice:
There is a fixed version of NetLib available if you are having
problems with any of the following:
N_FCOPY, N_REDIRECT, N_SCATTER, N_GATHER, N_BOLT
Please contact us at 212-724-0150 for update info.
Neil
Msg#9090 To:ALL
From:NEIL WEICHER Date:04/16/89
Subject:C CLASS
------------------------------------------------------------------------------
I am looking for a small number of people to participate in an
intensive 5-day C language class to tenatively begin on April 24,
in New York City. It will include the C/Assembler interface and
the Clipper/C interface.
It will be taught by a top instructor and will cost approx. $500
per person. Of course, the more people, the less per person.
If interested please contact me immediately at 212-724-0150.
Neil Weicher
Communication Horizons
Msg#9091 To:ALL
From:TIM CREAGH Date:04/16/89
Subject:CONFIG.SYS Parent Msg#9091
------------------------------------------------------------------------------
Has anyone written code to check the config.SYS for the files =
statement? Since my program opens 30 files I should have the thing check
for the statement in the installation. People just don't understant how
to change the config.sys themselves.
thanks
-tim
Msg#9092 To:DIRK LESKO (Rcvd)
From:RICHARD HORWITZ Date:04/16/89
Subject:(R)MULTI-USER CLIPPER Parent Msg#9081
------------------------------------------------------------------------------
Quarterdeck has QEMM-386, which is different than 386-to-the-max.
Msg#9093 To:RICHARD HORWITZ (Rcvd)
From:DIRK LESKO Date:04/17/89
Subject:(R)MULTI-USER CLIPPER Parent Msg#9081
------------------------------------------------------------------------------
Whatever.....
Msg#9094 To:DIRK LESKO (Rcvd)
From:NOEL MANZANILLA Date:04/17/89
Subject:INUSE()
------------------------------------------------------------------------------
dirk, I have been using your Funcky.lib for a while and have recently had
the need to find out what indices are open in a work area. in your
manual, you mentioned a Clipper inuse() function. I can't find it
anywhere. am i missing something ? how about a work around ? thanks
Msg#9100 To:NEIL WEICHER (Rcvd)
From:MARC SCHNAPP Date:04/17/89
Subject:(R)APPEND VS. LOW-LEVEL I Parent Msg#9002
------------------------------------------------------------------------------
Thanks for the suggestion to SET EXCLU ON (it's a single user app
anyhow).
Msg#9101 To:DAVE LOBEL (Rcvd)
From:MARC SCHNAPP Date:04/17/89
Subject:(R)APPEND VS. LOW-LEVEL I Parent Msg#9002
------------------------------------------------------------------------------
Thanks for the input re: APPEND. Neil says set exclu on will speed it up
somewhat.
Msg#9102 To:CONSTANTIN FLORESCU (Rcvd)
From:CLIFF GREEN Date:04/17/89
Subject:(R)CLIPOVLS Parent Msg#9102
------------------------------------------------------------------------------
sure. just download the file from here. Any questions after that? --
just post 'em here.
(filename should be CLIPOVLS.ARC)
Msg#9107 To:DIRK LESKO (Rcvd)
From:MARK NEIDORFF Date:04/17/89
Subject:INFOWORLD Parent Msg#9107
------------------------------------------------------------------------------
Hi Dirk:
Not believing anything (much) that I read, I thought that I'd tell you
that Nantucket just had its third-party (gee what happened to the first
party and the second party?) developers out to discuss Tools one. Were
yoy invited? Did you choose to attend? Did you crash the party (just
kidding)? Can you shed any light on what happened there?
Thanks,
<< Mark >>
Msg#9108 To:ALL
From:JON MELTZER Date:04/17/89
Subject:UP DOWN ARROW KEYS & GET Parent Msg#9108
------------------------------------------------------------------------------
I need to have the up and down arrow keys work in a Clipper GET the same
way as in a dBase GET : the cursor column position is not lost in
character input. I've tried UDFs and have found that Clipper sets
COL() to the last space of the GET before the VALID UDF is executed.
Can anyone tell me how to keep the column position as I move up or
down the GETs?
Msg#9109 To:MARK NEIDORFF (Rcvd)
From:DIRK LESKO Date:04/18/89
Subject:(R)INFOWORLD Parent Msg#9107
------------------------------------------------------------------------------
Can;t say anyhting till I know what is going on....
dLESKO
Msg#9110 To:JON MELTZER (Rcvd)
From:RICHARD HORWITZ Date:04/18/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
If you have FUNCky (and you should) You could setup an ONKEY() function
which checks for the up/down arrow keys. If it is an arrow, get the
current cursor position, activate() the previous/next get, and csrput()
the cursor at the appropriate column. Should work.
Msg#9111 To:DIRK LESKO (Rcvd)
From:MIGUEL MALDONADO Date:04/18/89
Subject:(R)LIBRARIES Parent Msg#9052
------------------------------------------------------------------------------
Well if it's possible, I'm going to give it a try. Let you know how
things go.
Msg#9112 To:JON MELTZER (Rcvd)
From:DAVE LOBEL Date:04/18/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
Jon,
If you have FUNCky, there is an answer.
There is a FUNCtion called ONKEY( "func()" ). ONKEY() will execute that
"func()" on every key stroke. Therefore, you create a UDF that will
store the current COL() and ROW() after each keystroke. Just substitute
this UDF in place of "func()".
This should answer your problem.
Good luck......
YEL-DAVE
Msg#9114 To:ALL
From:CONSTANTIN FLORESCU Date:04/18/89
Subject:TEXT FILE Parent Msg#9114
------------------------------------------------------------------------------
Anybody could tell me how to dump the result of an application to a .TXT
file e.g.
@10,10 say "Lname"
@11,10 say "Fname"
I want the output of this(Lname &Fname) to be written to a disk file
under a TXT file. I know that's a very common command but just don't
remember it.Thanks
<CF>
Msg#9117 To:ALAN GUGGENHEIM
From:MIKE DRAPER Date:04/18/89
Subject:(R)MULTI-USER CLIPPER Parent Msg#9081
------------------------------------------------------------------------------
How about Lantastic? Each workstation uses only about 20K for the
network drivers. While it is very cheap with the Lantastic twisted pair
cards, it will also run with Ethernet or Arcnet. It provides peer
sharing (everyone can be a server) and no special disk formating. While
it is much slower than Novell or 3Com, it is fast enough for 3-6 active
nodes and works with standard Clipper file locking.
Msg#9118 To:TIM CREAGH
From:JOEL SHAPIRO Date:04/18/89
Subject:(R)CONFIG.SYS Parent Msg#9091
------------------------------------------------------------------------------
Code is available as part of Steve Straley's Toolkit 1.2
Msg#9119 To:CONSTANTIN FLORESCU (Rcvd)
From:JOEL SHAPIRO Date:04/18/89
Subject:(R)TEXT FILE Parent Msg#9114
------------------------------------------------------------------------------
SET ALTE TO and SET ALTE ON
Msg#9120 To:ALL
From:LARRY GOTTLIEB Date:04/18/89
Subject:CLIPPER DEV. CONFERENCE Parent Msg#9120
------------------------------------------------------------------------------
I've just received registration information for the 1989 Clipper
Developers Conference. Dirk Lesko is scheduled to hold a workshop on
assembler routines for Clipper; lots of other workshops, of course.
I plan to go; does anyone else? Note that for the low early registration
fee of $595, you get an official conference t-shirt -- what a bargain!
Msg#9122 To:ALL
From:MENACHEM BAZIAN Date:04/18/89
Subject:UNIX Parent Msg#9122
------------------------------------------------------------------------------
There has been some discussion of clipper running on other platforms
(such as UNIX). Is there anyone out there who is more than familiar with
UNIX?? I have a few technical questions on UNIX and hardware
configurations.
Specifically. Can a 386 25mhz (or 33mhz) with as much RAM as needed (say
8-16 Meg) running with intelligent controllers run decently with 40
terminals??
Thanks,
Menachem
Msg#9123 To:RICHARD HORWITZ (Rcvd)
From:JON MELTZER Date:04/18/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
Sounds good ... Dirk, can I do this with FUNCky?
Msg#9124 To:CONSTANTIN FLORESCU (Rcvd)
From:CLIFF GREEN Date:04/18/89
Subject:(R)CLIPOVLS Parent Msg#9102
------------------------------------------------------------------------------
Constantin,
If, after going through the material in CLIPOVLS, you've got any
suggestions/requests for commonly used/needed functions, please fire
away. I'd like to expand the stuff I started there into a broader
library of functions; I respond quicker to prompts, though the result
may not be as clean.
Msg#9125 To:CONSTANTIN FLORESCU (Rcvd)
From:RICHARD HORWITZ Date:04/18/89
Subject:(R)TEXT FILE Parent Msg#9114
------------------------------------------------------------------------------
I think what you're looking for is SET ALTERNATE.
Ex. SET ALTE TO <FILE>
SET ALTE ON
......print stuff.....
SET ALTE OFF
CLOS ALTE
I think that should work.
Msg#9127 To:CONSTANTIN FLORESCU (Rcvd)
From:MARK NEIDORFF Date:04/18/89
Subject:(R)TEXT FILE Parent Msg#9114
------------------------------------------------------------------------------
Something like:
Set Alternate To
is what you mean, I think.
<< Mark >>
Msg#9128 To:MENACHEM BAZIAN (Rcvd)
From:MARK NEIDORFF Date:04/18/89
Subject:(R)UNIX Parent Msg#9122
------------------------------------------------------------------------------
There are all sorts of issues with unix. You can't install it on aby 386
and expect everything to work . Drivers need to be written for just
about everything! and even then you may not be home free.
The 386 has the p[ower to support 40 terminals, but it is a lot of work.
If you want to know more, or details, just ask!
<< Mark >>
Msg#9129 To:RICHARD HORWITZ (Rcvd)
From:DIRK LESKO Date:04/18/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
That will move the cursor but it will return to the first character of
the GET as soon as he hits a key, he will have to stuff cursor key
movement to get the logical cursor, as well as the physical cursor tpo
the correct position...
dLESKO
Msg#9131 To:MENACHEM BAZIAN
From:DIRK LESKO Date:04/18/89
Subject:(R)UNIX Parent Msg#9122
------------------------------------------------------------------------------
Yes....
Msg#9132 To:JON MELTZER (Rcvd)
From:DIRK LESKO Date:04/18/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
I think I left a message to Rich that explained why not....You will have
to stuff cursor keys into the next GET to move the logical cursor to the
correct position. If you just move the physical (hardware) cursor to the
correct spot, Clipper's internal cursor locations will not be updated
(that's the logical)...I did that trick in memoedit() with the mouse
being able to move the cursor, it's pretty simple. You should be able to
figure out how many cursor key values to stuff in order to get it into
the position you want....
dLESKO
Msg#9133 To:DIRK LESKO (Rcvd)
From:RICHARD HORWITZ Date:04/18/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
Physical cursor meaning CSRPUT? Doesn't stuffing the arrow keys move the
cursor to the correct position?
Msg#9134 To:MARK NEIDORFF (Rcvd)
From:CONSTANTIN FLORESCU Date:04/19/89
Subject:(R)TEXT FILE Parent Msg#9114
------------------------------------------------------------------------------
Thanks. Yes that I was looking for.
<CF>
Msg#9135 To:RICHARD HORWITZ (Rcvd)
From:CONSTANTIN FLORESCU Date:04/19/89
Subject:(R)TEXT FILE Parent Msg#9114
------------------------------------------------------------------------------
It does and thank you.
<CF>
Msg#9137 To:ALL
From:CONSTANTIN FLORESCU Date:04/19/89
Subject:NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
Is there possible to enter a small calculation in a Dbase/Clipper numeric
field ? I noticed that in Dataflex they can enter small calculations like
4+2/2 and get the result in the same field. Seems that their numeric
field accepts + - / * operators which sometimes is pretty handy. As you
know Dbase/Clipper numeric field doesn't accept nothing else but a
number. Is there any way that that can be done ?
<CF>
Msg#9138 To:RICHARD HORWITZ (Rcvd)
From:DIRK LESKO Date:04/19/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
Did you mention that he would have to stuff cursor keys?...than I
mis-interpreted...
dLESKO
Msg#9139 To:DIRK LESKO (Rcvd)
From:JON MELTZER Date:04/19/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
This is still not clear. I can, of course, have a VALID call an assembler
function that will move the cursor --- but, I have no way of deducing
what the cursor column is! All I know is that I am SOMEWHERE within
the character GET. Now, I assume that Clipper must know, during the
READ, where the cursor column is - but that information is, it appears,
lost as soon as a GET termination occurs. I have to know the COL()
position before Clipper "fixes" it for the next GET.
Msg#9140 To:JON MELTZER (Rcvd)
From:DIRK LESKO Date:04/19/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
Can't you use onkey() to keep track of the last cursor position?
Msg#9141 To:DIRK LESKO (Rcvd)
From:JON MELTZER Date:04/19/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
OK, Dirk, Dave, and Richard ... is this what I should do?
PUBLIC stored_row, stored_col
ONKEY("storecol")
@ 1,1 get string1 valid fixcol(1,1)
@ 2,1 get string2 valid fixcol(1,1)
*etc.
READ
FUNCTION storecol
stored_row = ROW()
stored_col = COL()
RETURN t.
FUNCTION fixcol
PARAM start_row,start_col
DO CASE
CASE LASTKEY() = down_arrow && and not last GET...
KEYBOARD REPLICATE(CHR(right_arrow),stored_col-start_col)
csr_put(start_row+1,stored_col)
* <etc.>
ENDCASE
RETURN .t.
Msg#9142 To:ALL
From:CONSTANTIN FLORESCU Date:04/19/89
Subject:SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
I was trying to dump the "SAYS" of a program file to a TXT file and I was
told to use "SET ALTE TO FILE.TXT"
E.G.
set alte to cf.txt
@10,10 say "constantin"
@11,10 say "florescu"
set alte off
I thought that this will write the above SAYS to a TXT file CF.TXT but it
doesn't. All it does is to display. Seems that CF.TXT is created but is
nothing in it. Anybody can clerify this for me. Thanks.
<CF>
Msg#9143 To:CONSTANTIN FLORESCU (Rcvd)
From:JOEL SHAPIRO Date:04/19/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
After SET ALTE TO CF.TXT you should have
SET ALTE ON
and
after SET ALTE OFF
You should have SET ALTE TO to close the file
Joel
Msg#9144 To:DIRK LESKO (Rcvd)
From:JON MELTZER Date:04/19/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
I have solved the problem - thanks to all for assistance. I replaced
the keyboard interrupt with one that saved the cursor position before
reading the keyboard.
Msg#9147 To:CONSTANTIN FLORESCU (Rcvd)
From:CLIFF GREEN Date:04/19/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
unless I'm mistaken, SET ALTE works with ?/?? printing, while @SAY will
require you to use SET PRINTER TO <filename>. You would, of course,
reset the default printer destination after writing to the file with SET
PRINTER TO, or the equivalent.
Any particular reason why you're writing to the file in this manner?
Msg#9149 To:CLIFF GREEN (Rcvd)
From:CONSTANTIN FLORESCU Date:04/19/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
Yes. I have to create some records(text) on same line at a certain column
position...You're are right about ?? and SAY stuff(I just realized)..but
here's the message I prepared for you or anybody who can give me any idea
of how can be done....
What I was trying to do is this
set alte to textfile
set alte on
@10,01 say "Clipper"
@10,12 say "is"
@10,25 say "also a DataBase"
set alte off
set alte to
According to this above code must have written a text file like this:
Clipper is also a DataBase.
(on the same line)
But seems that doesn't do it. First one must store "Clipper", "is" etc
to a variables then issue the command "set talk off" before "set alte
to textfile" in order to get something. Otherwise the "textfile" has
only 1(one) byte !
More over (and that is the most important for me) is that above are
stored in the "textfile" not on the same line but like this:
Clipper
is
also a DataBase
IF anybody could tell me how I could have those variables on the same
line and at the column position I mentioned in my Code I would
appreciate.
Thank you.
<CF>
PS I will try with the ?? stuff but will take me forever to put the
variables at certain column position.
Msg#9150 To:JON MELTZER (Rcvd)
From:DIRK LESKO Date:04/19/89
Subject:(R)UP DOWN ARROW KEYS & G Parent Msg#9108
------------------------------------------------------------------------------
You have to check for negative values, then convert them to positive with
the ABS() function...
dLESKO
Msg#9151 To:CONSTANTIN FLORESCU (Rcvd)
From:DIRK LESKO Date:04/19/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
Aren't you supposed to use SET DEVICE TO <filename> in order to print the
@ .. says?
dLESKO
Msg#9152 To:DIRK LESKO (Rcvd)
From:MENACHEM BAZIAN Date:04/19/89
Subject:(R)GETBACK Parent Msg#8762
------------------------------------------------------------------------------
Dirk,
I don't recognize those functions...Are they from Funcky or something
like that???
Menachem
Msg#9154 To:CONSTANTIN FLORESCU (Rcvd)
From:ROB HANNAH Date:04/19/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
Constantin, you need to add the commands:
SET DEVICE TO PRINT
SET PRINTER TO <filename>
In order for @ says to work correctly.
It's actually in the Clipper manual, too!
Msg#9155 To:MENACHEM BAZIAN
From:DIRK LESKO Date:04/19/89
Subject:(R)GETBACK Parent Msg#8762
------------------------------------------------------------------------------
Yes they are. FUNCky also has those Hebrew GETS I mentioned awhile
back...complete with fonts.....That's part of the reason why it's named
FUNCky.....Becase it is in some ways....
dLESKO
Msg#9156 To:CONSTANTIN FLORESCU (Rcvd)
From:RICHARD HORWITZ Date:04/19/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
You must SET ALTE TO *then* SET ALTE ON
Msg#9157 To:MENACHEM BAZIAN
From:RON CALLAHAN Date:04/20/89
Subject:(R)UNIX Parent Msg#9122
------------------------------------------------------------------------------
Menachem:
I believe that it would be possible depending on the architecture of
the hardware. (ie. type of diskdrive, etc.) I've had some experience
with the older Convergent Megaframes using a 68010 chip running at
12 mghz. The system did a fair job running 20 users. Again I must
stress the hardware design of the system.
Ron.
Msg#9160 To:ROB HANNAH (Rcvd)
From:CONSTANTIN FLORESCU Date:04/20/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
Yes I think almost everything is in the CLipper,Dbase,Funcky,Tollkit etc
etc but do have problems handling all these information. I do read
them(or at least most of them) once or twice but after a while I just
don't remember all of them(even if they appear to be simple) especially
if I don't use these command every day. But believe me I still use and
read the manuals. Sorry for trouble and thanks for help.
<CF>
Msg#9161 To:RICHARD HORWITZ (Rcvd)
From:CONSTANTIN FLORESCU Date:04/20/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
Thanks. I also just found out(and realized) that set alte to doesn't work
with @ say
<CF>
Msg#9162 To:ALL
From:CONSTANTIN FLORESCU Date:04/20/89
Subject:SET ALTE ON
------------------------------------------------------------------------------
Just wanted to thank everyone you helped to clear this problem. For your
information the right commands and sequence to send the results of
@..says to a textfile is as follows:
set printer to Textfile.txt
set device to print
set print on
@10,10 say "This goes in a text file"
set device to screen
set print off
set printer to
It works in Clipper only. Not Dbase
<CF>
Msg#9163 To:ALL
From:DONALD WISHNEK Date:04/20/89
Subject:NETWORK FUNCTIONS
------------------------------------------------------------------------------
I'M PROGRAMMING FOR A 3COM NETWORK RIGHT NOW AND I WAS WONDERING IF
ANYONE CAN HELP ME WITH THE FUNCTION LIBRARY NETFN WHICH I DOWNLOADED TO
MY PC FROM THIS BOARD. IT REQUIRES THE LATTICE C LIBRARY LC.LIB I WENT
OUT AND PURCHASED THE LIBRARY AND I GOT SEVERAL LINK ERRORS. DOES ANYONE
HAVE EXPERIENCE WITH THE NETFN ROUTINES THAT CAN HELP ME OUT. I'D REALLY
APPRECIATE IT.
THANKS
DON
Msg#9165 To:CONSTANTIN FLORESCU (Rcvd)
From:MARK NEIDORFF Date:04/20/89
Subject:(R)NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
Sure you can do calcs in a numeric field:
Instead of getting a numeric, get a string type. Then parse the string
looking for operators and use a case statement to do the operations.
(Its easier to buy a hand help calculator though (grin))
<< Mark >>
Msg#9167 To:CONSTANTIN FLORESCU (Rcvd)
From:MARK NEIDORFF Date:04/20/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
I think that the @say command is sinking you on this. It sends a CR/LF
by itself each time it executes. You don't have control of that. That
is what the ?? is for. Too bad it is so weak!
<< Mark >>
Msg#9170 To:ALL
From:MARK NEIDORFF Date:04/20/89
Subject:I SPY A BUG?
------------------------------------------------------------------------------
Well, the question is: Is it a bug or am I misreading what is not
happening here?
The following command does not generate an error message where I expect
it to. Can anyone explain why (if there is a legimate reason)?
Temp_var = if(which="F",GetFilter(fld_name[choice,@choice),GetIndex(...
missing closing bracket---------^
the compiler does not return an error, nor does the linker. Runtime...
returns an unexpected value in Temp_var.
Thanks,
<< Mark >>
Msg#9173 To:DAVID MINTER (Rcvd)
From:NEIL WEICHER Date:04/20/89
Subject:(R)NETLIB Parent Msg#9173
------------------------------------------------------------------------------
Dave:
Thanks for the feed-back. Yes we do have a Norton version for NetLib.
If you call the office at 212-724-0150 Jackie can send it out to you
gratis.
Neil
PS - I prefer the Rettig over the Norton guide myself.
Msg#9174 To:CONSTANTIN FLORESCU (Rcvd)
From:JOE BOOTH Date:04/20/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
Add SET ALTERNATE ON after the SET ALTE TO cf.txt command
Msg#9176 To:CONSTANTIN FLORESCU (Rcvd)
From:DAVE LOBEL Date:04/20/89
Subject:(R)NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
Constantin,
Using FUNCky's onkey() and timeout() FUNCtions, you can devise a sheme
that will let you use both a character and a numeric field in the same
get area.
What you do is this:
First, find out as soon as you enter the numeric get.
Second, once you do, you to a savegets() and activate a single character
get.
Third. When you are in this character get, the user can enter in an
equation, a numeric, and up arrow, down arrow, etc...
After the user is finished with this get, you evaluate it to see
what was done. If the result was an equation, then you can stuff
the value in to the numeric field, otherwise something else.
Give it a shot, I'm sure it will work.
YEL-DAVE
Msg#9179 To:ALL
From:DAVE LOBEL Date:04/20/89
Subject:NETWORK EXPERT.....
------------------------------------------------------------------------------
To ALL:
I am looking for a NOVELL expert that has
experience installing NOVELL software and
Network hardware. There is an existing
Network, but the client wants to upgrade
his hard disk.
If you are qualified and interested, please
leave your name and number (day and night)
as a private message and I will get back to
you as soon as possible.
Oh, by the way, there is good $$$$ in this
one.....!!!!!!!
Thanks.
David E. Lobel
YEL-DAVE
Msg#9181 To:MARK NEIDORFF (Rcvd)
From:CONSTANTIN FLORESCU Date:04/20/89
Subject:(R)NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
Thanks. But seriously, you think is not worth to write such a code ? ANd
if written can estimate how many lines will be there ? Probably there is
a function out there(Funcky) which could do it ?
<CF>
Msg#9182 To:MARK NEIDORFF (Rcvd)
From:CONSTANTIN FLORESCU Date:04/20/89
Subject:(R)SET ALTE ON Parent Msg#9142
------------------------------------------------------------------------------
It doesn't sent a CR/LF but instead when you say @01,01 say "Mark" that
will go actualy not at col 1 row 1 but at col 2 row 2 (of course if you
consider it as a text file and not as a screen display)
<CF>
Msg#9183 To:DAVE LOBEL (Rcvd)
From:CONSTANTIN FLORESCU Date:04/20/89
Subject:(R)NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
Thanks Dave. I will try and let you know.
<CF>
Msg#9185 To:CONSTANTIN FLORESCU (Rcvd)
From:MARK NEIDORFF Date:04/20/89
Subject:(R)NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
It's definitely worth it to write an expression (numeric) evaluator if
you need it. Dave Lobel left one idea on how to do it. I would :
before a get stack evaluate the type of each variable and find which are
numeric.
(2) Substitute a char variable for the numeric one.
(3) Get the string for these vars.
(4) Evaluate it char by char (use substr or whatever in a loop) storing a
number in a numeric variable.
(5) when I see an operator, store it and
(6) evaluate the next number (use substr again)
(7) at the next operator, or the end of the string, evaluate the result
of the operation.
(8) loop until the string is finished.
I don't know how much code it would wind up taking, but its not hard.
Just moderately long.
<< Mark >>
Msg#9186 To:TIM CREAGH
From:JOHN THORN Date:04/20/89
Subject:(R)CONFIG.SYS Parent Msg#9091
------------------------------------------------------------------------------
Tim:
I've got one someplace (in with a bunch of other .prg(s)). Let me dig it
out and I'll upload it.
Msg#9187 To:LARRY GOTTLIEB (Rcvd)
From:JOHN THORN Date:04/20/89
Subject:(R)CLIPPER DEV. CONFERENC Parent Msg#9120
------------------------------------------------------------------------------
Please take good notes and remember us po' folks who can't attend....
(and get that super deal on a T-shirt!)
Msg#9188 To:CONSTANTIN FLORESCU (Rcvd)
From:JOHN THORN Date:04/20/89
Subject:(R)NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
Enter it as text, call a UDF with the VALID, break the text down into an
equation, return the result, write over the old get with the ans as a
numeric.
Msg#9189 To:JOHN THORN (Rcvd)
From:LARRY GOTTLIEB Date:04/21/89
Subject:(R)CLIPPER DEV. CONFERENC Parent Msg#9120
------------------------------------------------------------------------------
Will do!
Msg#9190 To:NEIL WEICHER (Rcvd)
From:DAVID MINTER Date:04/21/89
Subject:(R)NETLIB Parent Msg#9173
------------------------------------------------------------------------------
Actually, I am starting to like the Rettig version also. The problem is
that I Just purchased DBase-Online for all of the dBase stuff. I
downloaded Dirk's Rettig version for funcky. But your engine doesn't
talk to dirk's and his doesn't talk to yours. I can't really justify
buying Rettig's when I just got nortons.
Oh well, serves me right I guess. I'll give Jackie a call right away.
Dave
Msg#9191 To:DAVID MINTER (Rcvd)
From:DIRK LESKO Date:04/21/89
Subject:(R)NETLIB Parent Msg#9173
------------------------------------------------------------------------------
Dave, Norton Engines are available for FUNCky as well....Basically I
think the online help thing is going to get really messy because all the
newer products I have seen coming down the pipe all have some kind of
built in help....That means you'll have a different kind of help system
for every language product you'll be using.....yuk!
dLESKO
Msg#9192 To:MARK NEIDORFF (Rcvd)
From:CONSTANTIN FLORESCU Date:04/21/89
Subject:(R)NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
I might try to write a generic function do that. But probably will take
me a while, since now I am very very busy. I will let you know. Thanks.
<CF>
Msg#9193 To:JOHN THORN (Rcvd)
From:CONSTANTIN FLORESCU Date:04/21/89
Subject:(R)NUMERIC FIELD Parent Msg#9137
------------------------------------------------------------------------------
Thanks.
<CF>
Msg#9195 To:RON CALLAHAN (Rcvd)
From:JOHN KASTER Date:04/21/89
Subject:(R)ERROR TRAPPING Parent Msg#6284
------------------------------------------------------------------------------
Ron,
Not only that - many times you will get memory errors when the Clipper
datafile is corrupted. This is not the type of error message I would
expect to see if a data file were corrupted, but it makes sense in a
perverse sort of way, considering how Clipper probably opens the files.
John Kaster
Msg#9196 To:MARK NEIDORFF (Rcvd)
From:JOHN KASTER Date:04/21/89
Subject:(R)INSERT Parent Msg#6282
------------------------------------------------------------------------------
Not in dbedit(). Somebody (a vendor) has a function called vertical that
I've seen demonstrated that is pretty flexible, though.
John
Msg#9197 To:ROBERT HEYMAN (Rcvd)
From:JOHN KASTER Date:04/21/89
Subject:(R)INSERT Parent Msg#6282
------------------------------------------------------------------------------
The way we do that is to have a field in the database called list order.
It may seem like a kludge, but it works great - make it a real number
with 4 decimal places. Any time you want to insert a record, just take
the two boundary records and average them, putting this value in listord.
If your index is on listord, viola! (or is that cello?) you have just
inserted a blank record between two others. In dbedit(), might I add.
Pardon the pun, please.
John
Msg#9200 To:DIRK LESKO (Rcvd)
From:DAVID WEST Date:04/21/89
Subject:CONFUSION... Parent Msg#9200
------------------------------------------------------------------------------
How the heck do I reset my 'last message' read pointer? I've got all the
stuff for March dloaded, but every time I log in I end up a 03/29.
Msg#9201 To:CONSTANTIN FLORESCU (Rcvd)
From:JOHN DUKE Date:04/22/89
Subject:(R)SET ALTE ON Parent Msg#9201
------------------------------------------------------------------------------
Constantin, In the example you give, you've left out the SET ALTE ON,
goes after SET ALTE TO CF.TXT. Also, this only puts the data 'CONSTATIN
and FLORESU' (your exp) in the CF.TXT file it does not put '@10,10 say
"constatin" in. Further be aware, that once you issue the SET ALTE OFF
command, should you later in the program decide to write to this same
file again thinking you would be appending data, you won't. What you will
do is over-write the original file with the new data. Hope this helps.
Msg#9202 To:CONSTANTIN FLORESCU (Rcvd)
From:JOHN DUKE Date:04/22/89
Subject:(R)SET ALTE ON Parent Msg#9202
------------------------------------------------------------------------------
For you info, it does work in dbase also.
Msg#9203 To:DAVID WEST (Rcvd)
From:MARK NEIDORFF Date:04/22/89
Subject:(R)CONFUSION... Parent Msg#9200
------------------------------------------------------------------------------
Sorry to butt in:
The easiest way to get the messges is to download the monthly file of
clipper messages. The filename is of the form CLIPmmyy.ARC where mm are
the numbers (03 for march) representing the month and yy is the year.
To reset your message pointer, try:
first [Q] to quit reading at the last message pointer.
type [R] to start a read
when prompted for the messgae number, type the last number followed by a
minus sign.
This should do it.
<< Mark >>
Msg#9204 To:DAVID MORGAN (Rcvd)
From:JILL GOHR Date:04/22/89
Subject:SKINNED CATS Parent Msg#9204
------------------------------------------------------------------------------
> The attitude is that of the Clipper users I've talked with.
Paramount is a key word. To the extent that you qualify your
desire to skin the cat with reservations about how, that desire
is something less than paramount. The apparent intensity of an
apparent majority of Clipper users in expressing a desire to
eliminate memory inadqeuacy, almost always unqualified by particular
opinions about how, is my reason for concluding that their desire to
see the problem solved is as strong as they express it to be: namely,
paramount.
Out of curiosity, why didn't you reply to my message publicly? If you
had, maybe you would have heard from others that feel the same as I do.
Yes, I'm sure you hear from the Russells, McConnells, and the Ushers,
but are they truly the majority?
My only point, was that all too often, because no constraints are given,
either in regards to memory OR disk space, we end up with memory AND
disk hogs for software. If in fact the software development teams were
given tighter boundaries to work within, I truly believe we would see an
improvement in software efficiencies.
It also seems to me that the problem is, at this point in time, so
deep-rooted that there's no going back. The easy way out is NOT always
the BEST way. And how many times have you seen or heard "I'll get it
working now, and fix it later"? This statement is a classic as far as
I'm concerned, because 9 times out of ten, the end result is sloppy and
inefficient code that is NEVER fixed, but just begets more of its kind.
I will not belabor the point any further, because I only allow myself 1
windmill a day. I will however, congratulate you on your command of the
English language ... not a car dealer or politician alive could have
stated what you did in a more succint and decisive manner.
Sincerely,
Jill Gohr
Msg#9205 To:DAVID WEST (Rcvd)
From:DIRK LESKO Date:04/22/89
Subject:(R)CONFUSION... Parent Msg#9200
------------------------------------------------------------------------------
Try reading from the end first, you can go backwards by appending a minus
sign to the message number, like so:
r 9999- <return>
That will start reading at the last message (or 9999 if there are more
than 9999 messages) and will work it's way backwards....
dLESKO
Msg#9207 To:RON CALLAHAN (Rcvd)
From:HOWARD MARKS Date:04/23/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#9021
------------------------------------------------------------------------------
Avoid at all costs any network solutin that uses a modem to connect a
workstation directly to the server. They are ALWAYS much too slow since
they extend the net over the phone line. If you do a SEEK then part of
the .NTX file comes over the network to your workstation which examines
it to find the right record and then the data comes over the net.
Doing this at even 9600bps is REALLY slow. Not to mention that with the
usual 35% network overhead loading a 100K .EXE file will take
2-3MINUTES..
PC Anywhere Etc. only send the screen and keyboard data across the net
which can be anywhere from 10-1000000 times less than the disk I/O data.
God forbid you need to do a reindex with a remote bridge to the server;
it could take forever.
Msg#9209 To:CONSTANTIN FLORESCU (Rcvd)
From:MARK NEIDORFF Date:04/23/89
Subject:PRINTING
------------------------------------------------------------------------------
I recall a few messages ago you had a printing problem where you wanted
to get several things on the same line on the printer from different @
say statements in a prg file. Does the prow() function help? It forces
printing on the current printer row.
<< Mark >>
Msg#9210 To:HOWARD MARKS (Rcvd)
From:RON CALLAHAN Date:04/23/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#9021
------------------------------------------------------------------------------
Howard:
Good point. I've never really thought about it that way. I'll bring that
up to my clients the next time I have a problem.
Ron.
Msg#9211 To:DIRK LESKO (Rcvd)
From:DAVID MINTER Date:04/24/89
Subject:(R)NETLIB Parent Msg#9173
------------------------------------------------------------------------------
I got the FUNCky norton database a while back, and that was a real time
saver. I do like TR's better though, so after I went and spent $99 bucks
on dBase On-Line a month ago I decided what the heck, let's get TR's too!
(I snuck past a budget of $9000 for software purchases for this year, and
my boss wants me to use it up!)
Thanks for the info anyway,
Dave
Msg#9213 To:JOHN DUKE
From:CONSTANTIN FLORESCU Date:04/24/89
Subject:(R)SET ALTE ON Parent Msg#9201
------------------------------------------------------------------------------
Sure it helps. ANd thanks.
<CF>
Msg#9214 To:JOHN DUKE
From:CONSTANTIN FLORESCU Date:04/24/89
Subject:(R)SET ALTE ON Parent Msg#9202
------------------------------------------------------------------------------
I tried several times and it didn't but I will try again...
Thanks
<CF>
Msg#9216 To:JILL GOHR
From:DAVE LOBEL Date:04/24/89
Subject:(R)SKINNED CATS Parent Msg#9204
------------------------------------------------------------------------------
Go Jill, Go!!!!!
YEL-DAVE
Msg#9218 To:ALL
From:DONALD WISHNEK Date:04/24/89
Subject:LATTICE C LIBRARY
------------------------------------------------------------------------------
Can anyone tell me if i can link a lattice C library with clipper '87?
Msg#9220 To:HOWARD MARKS (Rcvd)
From:CLIFF GREEN Date:04/24/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#9021
------------------------------------------------------------------------------
pardon my ignorance, but let me get this straight. There's roughly two
ways of getting information from a network over a modem: 1) make the net
think the modem is a legitimate node, and have it send _everything_ down
the line, or 2) use something like pcAnywhere and just send the video and
console i/o up and down the line, right? What is the difference in
implementation, and is the setup for this kind of thing straightforward
(i.e., laid out in the documentation for the application [eg, pcAnywhere]
and/or the NOS)?
Msg#9221 To:MARK NEIDORFF (Rcvd)
From:DAVID WEST Date:04/24/89
Subject:(R)CONFUSION...
------------------------------------------------------------------------------
Thanx, I think that did work. BTW, I usually call in and get the ARC
files every month, but I still like to call in periodically to check out
what's going on. I always hate reading a 3-week old message that I could
have replied to if I'd read it in 'real-time'.
Msg#9223 To:CLIFF GREEN (Rcvd)
From:RON CALLAHAN Date:04/25/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#9021
------------------------------------------------------------------------------
Actually it's all somewhatcademic. The only Network I've heard of
that allows direct connection via modem is 10 net.
Ron.
Msg#9225 To:DIRK LESKO (Rcvd)
From:NOEL MANZANILLA Date:04/25/89
Subject:FUNCKY DEMO Parent Msg#9225
------------------------------------------------------------------------------
dirk, I just gave a friend of mine a compiled version of your demo
program that came with the update of the FUNCky library. He tried to run
it on a PS/2 with a color VGA card and a mono VGA monitor. For some
strange reason, he says that the text (@ Say's) don't print to the
screen. The boxes and fill characters come out though. Any thoughts ?
He is planning to purchase your library and is going to write a multiuser
application but his network has a mix of different video cards. Will
appreciate any help/suggestions.
thanks,
Noel
Msg#9226 To:NOEL MANZANILLA (Rcvd)
From:DIRK LESKO Date:04/25/89
Subject:(R)FUNCKY DEMO Parent Msg#9225
------------------------------------------------------------------------------
Yes, I know...A problem that has been long since corrected. If he wants I
can send him a demo from here. The video checking routine assumed that a
VGA card was always referenceing the color segment when in fact, the IBM
VGA remaps the video buffer to monochrome sometimes. You might have him
try issuing a MODE CO80 at the DOS prompt to verify that. Some programs
invoke a monochrome mode which puts the VGA into a monochrome emulation.
I was not aware that the VGA would actually remap video memory inthat
circumstance, and since the demo's n this board were uploaded before I
found that out, they display that problem. You can have him call us qt
201.435.8401 and we'll ship him a a demo disk that is compiled with the
current version of the library. Thanks for your help....
dLESKO
Msg#9227 To:RON CALLAHAN (Rcvd)
From:CLIFF GREEN Date:04/25/89
Subject:(R)NETWORK PRGRAMMING Parent Msg#9021
------------------------------------------------------------------------------
10-net> Oh. Thanks.
Msg#9229 To:ALL
From:DONALD WISHNEK Date:04/25/89
Subject:DOS ERROR TRAPPING Parent Msg#9229
------------------------------------------------------------------------------
IS THERE ANY WAY IN CLIPPER I CAN TRAP THE ERROR "Error reading Drive x
Abort, Retry, Ignore)?" ? I'm doing file copies on off hours and i need
to make sure the program ends insted of just hanging there. I'd like to
be able to redirect it to a error routine.
Thanks,
don
Msg#9231 To:ALL
From:CHRISTIAN BAUDE Date:04/26/89
Subject:CLIPPER DECOMPILER Parent Msg#9231
------------------------------------------------------------------------------
Hi.....
.
Does anyone know if there exists such a thing as a CLIPPER decompiler??
.Thanks for any replies..... CHRIS
Msg#9232 To:ALL
From:FRANCOIS DESROSIERS Date:04/27/89
Subject:BROWSE Parent Msg#9232
------------------------------------------------------------------------------
I am doing an voluntary application for the summer games in Quebec.
I import the data from a file of athlete and disciplines in a
temporary file that the user complete using the browse module
from the extendP file in summer 87.
.
The only problem I have is whit the way that BROWSE works in
Clipper. The user must press ENTER before editing a field. Does
there is a way or a library that exist that can simulate a
BROWSE the same way that in DBASE. I mean without the obligation
to press ENTER before editing a field.
.
Also is-it possible to simulate the FREEZE option of DBase.
.
I should appreciate to know if FUNCKY or any other library has those
option
.
Many thanks for your help.
.
. Francois Desrosiers
Msg#9234 To:FRANCOIS DESROSIERS (Rcvd)
From:DIRK LESKO Date:04/27/89
Subject:(R)BROWSE Parent Msg#9232
------------------------------------------------------------------------------
Hi, you have the source code to browse inthe examplep.prg file. I am
pretty sure you can simulate what you want to do with the reads()
function in FUNCky. It can be used to highlight the field and edit at the
same time. You can set up the CTRL/ARROW keys to scroll the fields. Just
remember to convert the fields to string type before editing, and then
convert back to whatever it was....Other than that what you want to do
will require some pretty messy programming...
dLESKO
Msg#9235 To:DIRK LESKO (Rcvd)
From:DAVID BAKER Date:04/27/89
Subject:MAIN MOD DO'S Parent Msg#9235
------------------------------------------------------------------------------
Hello Dirk and All:
I am new to the Confab. Want to say hello and ask a "?" about
something you disscused waaaay back when. I am now in the need of
overlays, and way back you said to get EXE's smaller to DO all PRG's in
the Main Mod of the PRG. ??--> Does this include those in an Overlay
thanx D.M.B (running out of time) Bye
Msg#9238 To:FRANCOIS DESROSIERS (Rcvd)
From:MARK NEIDORFF Date:04/27/89
Subject:(R)BROWSE Parent Msg#9232
------------------------------------------------------------------------------
Hi:
Haven't heard from you for a long time! I think that the source code for
browse() is provided with the distribution disks. If not, I would use
[and I do!] FUNCky to do what you want to.
To stick with clipper, you need to play with the udf in dbedit so that
when the user presses a lettewr or a number input mode is selected and
the key is sent to the keyboard command. (lastkey() will get you there)
I think funcky is better.
<< Mark >>
Msg#9239 To:DAVID BAKER (Rcvd)
From:MARK NEIDORFF Date:04/27/89
Subject:(R)MAIN MOD DO'S Parent Msg#9235
------------------------------------------------------------------------------
Overlays are a separate matter. There you need the separate compliation.
<< Mark >>
Msg#9240 To:DAVID BAKER (Rcvd)
From:DIRK LESKO Date:04/27/89
Subject:(R)MAIN MOD DO'S Parent Msg#9235
------------------------------------------------------------------------------
Well, }ÞYkÑA¼o{ 7if xw¾Ûöou are using overlays you will probably
still have to use them. But if you klump all the PRG files contained in
each overlay into one LARGE PRG file you should gain some extra K of ram.
do the samne for your main PRG files in the root. Savings should be in
the neighborhood of 9-15K, depending on the size of your application. In
general I find that the better your code is written to begin with, the
smaller the savings....
dLESKO
Msg#9241 To:DONALD WISHNEK
From:DIRK LESKO Date:04/27/89
Subject:(R)LATTICE C LIBRARY
------------------------------------------------------------------------------
ANytime....Have FUN!(Cky)
dLESKO
Msg#9242 To:DONALD WISHNEK
From:STEVE STRALEY Date:04/27/89
Subject:(R)DOS ERROR TRAPPING Parent Msg#9229
------------------------------------------------------------------------------
Yes, use the low level file functions to FIRST see if the drive exists.
These functions are outside of Clipper's internal error-trapping
routines. If you are able to FCREATE a dummy file to a drive (or a
subdirectory), you know that the drive is ready and in place. I have a
couple of functions VALID_DR() and VALID_PATH() in ToolkiT 2.0 that takes
care of this, but you can write it to...
Steve
Msg#9244 To:DIRK LESKO (Rcvd)
From:DAVID MINTER Date:04/28/89
Subject:TIMEOUT() AND N_SETTIME() Parent Msg#9244
------------------------------------------------------------------------------
Dirk and Neal,
I have both Funcky and NetLib. Both lib's have a function that seem to
do similar things, ie. start a background function (Funcky's timeout()
and Netlib's n_settime()). My question is, could you use both to start
TWO background processes, or would they conflict with each other.
I haven't tried it yet, but it could help solve a few problems.
Thanx,
Dave
Msg#9245 To:MARK NEIDORFF (Rcvd)
From:DAVID BAKER Date:04/28/89
Subject:(R)MAIN MOD DO'S Parent Msg#9235
------------------------------------------------------------------------------
Thanx Mark, I know I need seperate compilation, just though I could save
a little space.
Msg#9246 To:DIRK LESKO (Rcvd)
From:DAVID BAKER Date:04/28/89
Subject:(R)MAIN MOD DO'S Parent Msg#9235
------------------------------------------------------------------------------
Thanx for the reply (some of these boards outwest you never get an answer
unless you are in the clique). This is what I thought, I was cunfued
when reading the old thread if the technique would not work for overlays
too.
D.M.B.
Msg#9247 To:ALL
From:DAVID BAKER Date:04/28/89
Subject:RELATIVE RECORD 'S Parent Msg#9247
------------------------------------------------------------------------------
Does anyone have or has anyone seen (FUNKy...) or know of a way to return
a relative record number from an indexed data base. In a Geologic Data
Manager I an writing I need to display, sometimes with Achoice() and
othertimes Dbedit(), consecutive records in the DBF. I can SEEK the
first matching record and SOFTSEEK or DECEND() to find the last, but
there is no way short of COUNTing between these two to get a number of
records between. The count is needed for initialising an array or for an
end condition in DBEDIT() (e.g. to display only those with the same
serial #). On a SORTed DBF one could just take the difference in RECNO()
of the boundery records, obviously this will not work on a random but
INDEXed DBF. I have read the last few mounth confabs and see others
would put this on a wish list too. There has got to be a way, Counting
is far to slow especilay when the match numbers get large.
Any help would be great.
D.M.B.
Msg#9248 To:DAVID BAKER
From:RICHARD HORWITZ Date:04/28/89
Subject:(R)RELATIVE RECORD 'S Parent Msg#9247
------------------------------------------------------------------------------
If you have funcky, I belive Jeff Springer Uploaded a version of
mx_dbfview which allows the passing of a condition to restrict the
scrolling. This would let you popup a lookup list, which would only show
records which matched your condition (like while this number is equal to
whatever). It beats the hell out of setting a filter in dbedit!!!
Msg#9249 To:ALL
From:DALE THORPE Date:04/28/89
Subject:COM1 & COM2 HANDLING Parent Msg#9249
------------------------------------------------------------------------------
I need a procedure to allow my Clipper program to read data from COM1 or
COM2 and to write data to the same. I am installing a file management
system that is connected to a Cannon camera that stores documents on
film. The camera is connected to COM1, but the client wants the ability
to choose between COM1 and COM2. If anyone has any suggestions I would
greatly appreciate it.
Thanks
Dale Thorpe
Msg#9250 To:DONALD WISHNEK
From:DAVE LOBEL Date:04/28/89
Subject:(R)DOS ERROR TRAPPING Parent Msg#9229
------------------------------------------------------------------------------
Donald,
YES!!! It's a pretty FUNCky idea!!!!!!!!
YEL-DAVE
Msg#9252 To:CLIFF GREEN
From:HOWARD MARKS Date:04/28/89
Subject:(R)NETWORK PRGRAMMING
------------------------------------------------------------------------------
Right You can either extend the net to the user's PC (compute in the user
PC) software for this is typically only available from teh net vendor
(Novell includes it with Netware) or use a remote package like anywhere.
To use anywhere you set up a PC on the net and stick a modem in it then
treat it like any other anywhere station. (documentation with anywhere)
Msg#9254 To:DAVID BAKER
From:MARK NEIDORFF Date:04/28/89
Subject:(R)RELATIVE RECORD 'S Parent Msg#9247
------------------------------------------------------------------------------
I think that you will have to know the index file structure to know how
to do that. If you can figure out how the index file is put together and
you know the index key length (that's easy) then finding the distance
between the records could be "easily" calculated.
Steve Straley is taking an index file apart in the current issue of his
newsletter DOSS. Maybe he can help here.
<< Mark >>
Msg#9255 To:DONALD WISHNEK
From:JON MELTZER Date:04/28/89
Subject:(R)DOS ERROR TRAPPING Parent Msg#9229
------------------------------------------------------------------------------
Yes. If you have DOS 3.3 you can use Interrupt 44 (IOCTL), subfunction
0F. You'll have to write an assembler routine. See Ray Duncan's book,
_Advanced MS-DOS_, 2nd edition, for details. I know of no way to trap
the error in earlier DOS versions.
Msg#9256 To:CHRISTIAN BAUDE
From:JON MELTZER Date:04/28/89
Subject:(R)CLIPPER DECOMPILER Parent Msg#9231
------------------------------------------------------------------------------
Dirk's next project? :-)
Msg#9257 To:DAVID MINTER
From:DIRK LESKO Date:04/29/89
Subject:(R)TIMEOUT() AND N_SETTIM Parent Msg#9244
------------------------------------------------------------------------------
I beleive you should be able to run 2 background processes without any
problems. Just remember that if you overload the background process it
can affect the speed of the foreground process. My recomendation is to
use one or the other and control the 2 processes from one initial
starting procedure....
dLESKO
Msg#9258 To:DAVID BAKER
From:DIRK LESKO Date:04/29/89
Subject:(R)RELATIVE RECORD 'S Parent Msg#9247
------------------------------------------------------------------------------
There was a significant discussion on that very issue at one time on
NanForum on COmpuserve. Right now there is no way to get a relative
record number out of the index. It is just not in there, and on top of
that, when the index is updated, ALL the relative record numbering would
change creating an unbelievably time consuming update of EVERY node in
the Tree...
dLESKO
Msg#9259 To:DALE THORPE (Rcvd)
From:DIRK LESKO Date:04/29/89
Subject:(R)COM1 & COM2 HANDLING Parent Msg#9249
------------------------------------------------------------------------------
Com Stuff? Suggestions? - SilverComm, SilverComm or SilverComm....take
your pick...
dLESKO
Msg#9260 To:JON MELTZER
From:DIRK LESKO Date:04/29/89
Subject:(R)CLIPPER DECOMPILER Parent Msg#9231
------------------------------------------------------------------------------
Don;t think I haven't thought about it....I know how to do it, but by the
time I'd get it done they would change the dam,n thing on me.....
dLESKO
Msg#9261 To:DIRK LESKO (Rcvd)
From:FRANCOIS DESROSIERS Date:04/29/89
Subject:(R)BROWSE Parent Msg#9232
------------------------------------------------------------------------------
Thanks, I will order FUNCKY by the organisation of the Games.
Francois Desrosiers
Msg#9262 To:MARK NEIDORFF (Rcvd)
From:FRANCOIS DESROSIERS Date:04/29/89
Subject:(R)BROWSE Parent Msg#9232
------------------------------------------------------------------------------
I am here at each week but with the cost of the communication from
Quebec, I restrain my envy to communicate!.
thanks for your help.
Francois Desrosiers
Msg#9263 To:CHRISTIAN BAUDE
From:HARRY GILBERT Date:04/29/89
Subject:(R)CLIPPER DECOMPILER Parent Msg#9231
------------------------------------------------------------------------------
I believe Hilco Software is working on this -- I'll look in my archives
and get back to you. (Hilco sells the programs to salvage corrupted dBASE
files, and to "un-tokenize" dBASE II, dBASE III and Foxbase "compiled"
code.
Msg#9264 To:CHRISTIAN BAUDE
From:HARRY GILBERT Date:04/29/89
Subject:DECOMPILER Parent Msg#9264
------------------------------------------------------------------------------
Hilco Software, 11266 Barnett Valley Road, Sebastopol CA 95472-9555 (707)
829-5011 is working on a Clipper decompiler. Give Lee Hilliard a call,
and while you're at it, ask for literature on his other products. They
work well!
Msg#9265 To:ALL
From:MARK KASS Date:04/29/89
Subject:LIST BOXES
------------------------------------------------------------------------------
Help !!!
I don't remember where I saw it, but i remember seeing either a
message on this board, or a file or an article in DOSS that gave an
example on how to do built a list box that the user could list and check
all the reports that they wanted to produce - I think it used achoise().
Well, I can't remember where I saw this idea and I now could use
it for an application that I'm building. Does anyone remember the
concept and where I might have seen it ?
Thanks
Mark
Msg#9266 To:MARK KASS (Rcvd)
From:RICHARD HORWITZ Date:04/29/89
Subject:(R)LIST BOXES Parent Msg#9266
------------------------------------------------------------------------------
I uploaded a file called MULTARRY.ARC in the FUNCKY section Oto do that.
(you of course need FUNCky!)³. It allows you to toggle multiple array
elements, and Óhen returns those yÍou selected.
Msg#9270 To:DIRK LESKO (Rcvd)
From:MARK NEIDORFF Date:04/29/89
Subject:(R)RELATIVE RECORD 'S Parent Msg#9247
------------------------------------------------------------------------------
Ah, now I see what I suggested about reading the index file would not
give relative record numbers.
<< Mark >>
Msg#9272 To:RICHARD HORWITZ (Rcvd)
From:MARK KASS Date:04/30/89
Subject:(R)LIST BOXES Parent Msg#9266
------------------------------------------------------------------------------
Richard
Thank You !!!
I knew I saw how to do what I wanted to do, but I forgot where I
saw it. I've already downloaded Multarry.arc and of course I use FUNCky
- I forget how I used to write code before FUNCky
Thanks again
Mark
Msg#9275 To:JILL GOHR
From:STEVE STRALEY Date:04/30/89
Subject:(R)DOSS
------------------------------------------------------------------------------
Thanks for the message Jill...
To make sure about your subscription, please call the offices of Four
Seasons Publishing at 212-599-2141 and ask for Joe King. He'll make sure
it is set up for you in the next year. As far as driving goes, I used my
last 100$ to move out to the east (a one way ticket on People's Express)
and with my PC in my lap, I moved to the Big Apple. I sold my car to pay
off the last of my stuff in L.A. and being in NY, I don't need a car.
But now, with work and all, I wanted to get away from the city this
summer and head for the beach. A car is the solution. Also, when my
license expired 3 years ago, I didn't fly to L.A. to renew it... I jsut
continued to work and live in New York without giving it much thought.
Now with a car, I've got to get a learner's permit and then take the
driving test. Ironic ain't it! Especially considering that I am a
native (3rd generation) Los Angeles person, I was born to drive... and
yet, I still have to have a learner's permit. Oh well... enough of this
rambling... have a great week...
Steve
Msg#9276 To:ALL
From:HOWARD KAPUSTEIN Date:04/30/89
Subject:DBASE .NDX FILES
------------------------------------------------------------------------------
T]his is going to get very technical, so if you get left behind, don't
feel bad. I got lost many a time too, and I know Äthe material
semi-intimately by now.
I've been working on a host of functions to access .NDX files from C. I
just finished the IndexOn()function, and I noticed a peculiarity. My
functions are 100% compatible³ with dBASE III+. Quick brief: .NDX files
are B+trees, where each node has a maximum number of Key Entries as
determined by information in the Anchor Node (node 0 is the Anchor, nodes
1-end are data). I was using an index file with a max of 9 key entries
per node. Each node is 512 bytes.
ÍWhen I had 9 entries in the index, all in one node, and tried to insert
another, the node was spdit into 2 blocks, and a new root was created, No
sweat.] When I split the node, I take 1/2 the entries in the orignal node
+önd move them to a new node. Then I insert the new key en³try into
whichever node it should go. So after splitting a full node, I would have
2 nodes, the first with 5 entries, the second with 4 entries. After
inserting the new key entry, I had 6 and 4, respectively.
Here's the puzzling part. The dBASE generated index had 2 noÖdes just
like me, but with 9 keys in the first and 1 key in the second. The new
entry inserted had a value of " " so it would come before everything
else in the index. Worked fine, for me and dBASE. Both of our indexes are
compatible with each other.
Why doesn't dBASE share the key entries evenlyÖ between both nodes when
splitting? I always thought in B+trees, ³if a node has X entries and you
split it, the old and new n½·de should each have X/2 nodes. ÛWith dBASE's
method, you could (potentially) have a very ineffeicient tree. Create a
large tree (i.e. 1000 records) and delete some records. Potentially, you
could pare the tree down so you never removed any of the nodes (say 1000
records = 100 nodes), with only 100 records indexed, and every node has
only 1 entry. ßThis would have the same number of disk a]ccesses (and
thus or performance) regardless of whether you had 100 records indexed or
1000b.
Does anyone know why A-T does it this way? Is it just laziness? Their
metho$ is slightly easier to code, but ]isn't as high performance.Öß The
actual coding difference is marginal too.
Msg#9277 To:ALL
From:STEVE STRALEY Date:04/30/89
Subject:DEAD HORSE
------------------------------------------------------------------------------
I order to spark some additional activity on this board, should the dead
horse be ressurected or should a new issue be tackled... Since Dirk lives
only 3 blocks away, or for this weekend, he's more like 3,000 miles away,
maybe he could bring up a new topic....
Always in gest....
Steve
Msg#9280 To:HARRY GILBERT
From:RON CALLAHAN Date:04/30/89
Subject:(R)DECOMPILER Parent Msg#9264
------------------------------------------------------------------------------
Can anyone give me a valid reason for a decompiler for Clipper.
This is the reason I don't sell software without a signed trade secret
agreement.
Comments
Post a Comment