you can page up with HELP too, if you use the right terminfo for your terminal
emulator.
I like to set my terminal for 132x44, but if my term width is >80 then HELP
starts to mess up drawing it's boxes and placing the text. At 12 columns, it's
an unreadable mess. UVHELP doesn't try to draw boxes
elmarna Ave, Herne Bay, Auckland 1011, New Zealand
> Phone: + 64 9 360 5310 Mobile: + 64 21 400 486
> Email: k...@walstan.com
>
>
> On 31 January 2014 10:34, Ed Clark wrote:
>
>> Curious why you are doing this as a user exit?
>> iirc, there are differences betwee
Curious why you are doing this as a user exit?
iirc, there are differences between using an A and an Itype. In an itype, the
subroutine is called just once for each item in the file. With an A-type, the
subroutine is going to be called once for each value/subvalue in the attribute
referenced by
you only need to do this on universe. unidata is happy to use itypes without
having them compiled first.
universe's CD command insists on only working on the dict. It would be nice if
it accepted the DICT/DATA keywords:
CD DATA VOC IDESC
On Dec 5, 2013, at 3:44 PM, George Gallen wrote:
> Not
iirc, you can't use the <> extraction syntax in unidata itypes. Instead of
@RECORD<15> you need to use something like EXTRACT(@RECORD,15). Or use an
attribute name like you did.
Cool error message though. It's a palindrome.
On Dec 5, 2013, at 4:44 PM, Doug Averch wrote:
> I was playing with EV
>From the dictionary listing, looks like you are on universe (as opposed to
>unidata. Useful information to have, along with the flavor you're using. To
>find the flavor, use:
.L RELLEVEL
RELLEVEL
001 X
002 10.3.7
003 PICK
004 PICK.FORMAT
005 10.3.7
which in my example show I am in pick flavor
My guess would be that it's because the loop was very tight. I've seen lots of
instances where tight loops couldn't be interrupted at all. Maybe in this case
it wasn't in basic code when it interrupted, and couldn't find any basic code
within a few steps, so didn't offer D.
On Oct 2, 2013, at 1
on universe, it looks like only fileinfo(var,0) will let you test.
fileinfo(var,1) etc will abort complaining that var isn't a file variable
On Aug 1, 2013, at 9:52 AM, "Martin Phillips"
wrote:
>> on universe (not sure of unidata), you can use FILEINFO() to see if
>> something is a file varia
>
>
> -Original Message-
> From: u2-users-boun...@listserver.u2ug.org
> [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Ed Clark
> Sent: 01 August 2013 14:41
> To: U2 Users List
> Subject: Re: [U2] What is true
>
> Do people from Wales take
on universe (not sure of unidata), you can use FILEINFO() to see if something
is a file variable:
x=""
crt fileinfo(x,0)
returns 0. Would return 1 for an open file.
On Jul 31, 2013, at 4:47 PM, Wols Lists wrote:
> On 31/07/13 09:06, Martin Phillips wrote:
>> Of course, nothing is ever complet
Do people from Wales take well to being called British? Sadly, almost
everything I know about Wales comes from watching "Torchwood" and "Gavin and
Stacy"
A curious feature of true and false on universe:
0001 x=char(32)
0002 if x=0 then crt 'is zero' else crt 'not zero'
0003 if x then crt 'true'
don't worry about the color of the suit. It's the hat that matters. I think the
choices are black, white, and red.
On Jul 25, 2013, at 8:40 PM, Bob Wyatt wrote:
> We never knew which guys to ask for - the guys in the black suits or the
> white suits?
>
> -Original Message-
> From: u2-u
universe has an AUTOLOGOUT command. I think unidata has something similar.
On Jun 26, 2013, at 5:22 PM, Al DeWitt wrote:
> Tried sending this to sbsolutions but it didn't appear to get there so I'll
> try here in hopes some of you are using SB+/SBClient
>
> We are having an issue where we are
Rang a bell, so I went searching. I'm looking at the may/jun 2001 issue of
Spectrum. Full page ad on page 3, a company named infinetivity that was selling
a number of products including pick fusion "web integration for pick". They
also had a web hosting service. They mention a lot of urls:
pic
Toad the wet sPROCet !!
A lot of procs get run from the VOC, but for performance reasons it's usually
suggested that the proc in the VOC be only a linkage that calls a proc from a
file. (would you eat it with a proc, would you eat it in a VOC, I do not like
green eggs and ham, I do not like the
I could see a 4gl writing out this kind of code, or a menu system.
When you use indirect calling, the name of the subroutine can contain odd
characters that wouldn't normally be allowed--though on some platforms you can
also quote the call, e.g.:
CALL "my-sub**"(parms)
On Dec 3, 2012, at 7:38
IIRC, universe doesn't let you run *any* query commands, including LIST-ITEM,
on a file if there isn't a dict defined. So at some point someone had edited
the voc f-pointer for LISTU.HISTORY, and put VOC into attribute 3? That's
slightly dangerous. Good thing uv warns.
On Sep 12, 2012, at 6:
The only "decompiler" that I know of is the VLIST command on universe. I don't
think there is an equivalent on unidata.
>CT UVBP TRY
TRY
0001 A=1
0002 A<-1>=2
0003 B=SUM(A)
0004 CRT B'R#10'
0005
>
>VLIST UVBP TRY
Main Program "M:\MV\UVBP.O/TRY"
Compiler Version: 10.3.0.0
Object Level: 5
yep, those 79 programs are for compatibility with prime and are mostly not
documented, because they aren't documented :). the source for all of them is in
APP.PROGS I think.
On Aug 27, 2012, at 4:58 PM, Wjhonson wrote:
>
> Wow!
> We have *SEVENTY NINE* entries in Global.Catdir starting with a
yep, it's there for compatibility with pr1me. the source is in APP.PROGS SLEEP:
0031 subroutine PR1ME(time.in.milliseconds)
0032 ;* if time.in.milliseconds < 1000 then sleep 1
0033 ;* else sleep (time.in.milliseconds / 1000)
0034
0035NAP time.in.milliseconds
0036 return
On Aug 27, 2012, at 4:
So what exactly do you mean by interpolation?
If you google "variable interpolation" you can find a bunch of examples from
other languages, such as:
for($i=1;$i<=15;$i++)
{
${'test_'.$i} = $i;
}
There's nothing like this in any mv implementation that I know of. In mv you
would probably us
I'm pretty sure that's the OS user id that you would see on unix in /etc/passwd
or as the first column in a ps -f listing. If you're root it will be 0. On
windows it's some similar number. unidata returns it in @UID
On Aug 21, 2012, at 1:49 PM, Wjhonson wrote:
>
> Universe 10 on Windows
>
>
There may be a little confusion here too because unidata and some flavors in
universe have vector math on by default, so for example:
001 A=1:@AM:3
002 B=2:@VM:9
003 CRT A+B
on unidata and universe in information flavor this will print 3]9^3, while
universe in ideal or pick flavor will print 0
Have you tried examining the strings?
The equal sign will work if the 2 strings are the same. But they can be equal
as dynamic arrays while not equal as strings. For example, on universe:
0001 A="1":@AM:"2"
0002 B="1":@AM:" 2 "
0003 CRT A=B
0004 CRT EQS(A,B)
prints:
0
1^1
The 2 dynamic arrays
so is this summary correct?
Unidata only accepts the '[]' wildcards in ecltype "P". In ecotype "U" you need
to use the LIKE/MATCH "…" syntax.
However, in ecltype "P", paragraphs always run their commands as ecltype "U".
So in ecltype "P", the "[]" wildcards will work when typed at TCL, but not in
he output, too.
> If you have multiple associated columns, & if your usage gets fancy (that
> (future) option is a reason for this approach), then your EVAL may need
> ASSOC, ASSOC.WITH, MULTIVALUED keywords, besides AS, FMT, COL.HDG.
> Other than that, I love EVAL.
>
>
There are 3 ways I can think of offhand to tell the subroutine what date range
to use. I like #3 the best:
1: use common variables, which is what the example function MARKH does. The
downside is that you have to assign the common variables before you run the
query.
2: You could use @SENTENCE i
I think the answer is to use an i-type dictionary.
You can use WHEN instead of WITH on multivalued columns to limit printing to
just the multi values you want, but then you wouldn't get an output line for
the items where there were no calls in the range.
You could do 2 separate reports, one with
gt; just fix the problems when they come up.
>
> Thanks everyone for their input.
>
> Doug
>
>
> -Original Message-
> From: u2-users-boun...@listserver.u2ug.org
> [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Ed Clark
> Sent: Thursday, June 21, 2012 5:2
universe or unidata? I don't have a full answer either way, but I seem to
recall that unidata holds CAPTURING output in a temporary file. Maybe that file
can be checked for.
On Jun 21, 2012, at 10:08 AM, Doug Farmer wrote:
> I just had a situation where a program was printing an error message a
Subject: Re: [U2] Runoff ?
>
>
> Don't be so sure.
> We don't even have a full screen editor built into Universe
>
>
>
>
>
>
> -Original Message-
> From: Ed Clark
> To: U2 Users List
> Sent: Sat, Jun 16, 2012 6:56 am
> Sub
ed Rocket to add the BIG shiny toys
that will make the database competitive with Oracle and SAP.
On Jun 16, 2012, at 3:44 PM, Wjhonson wrote:
>
> Don't be so sure.
> We don't even have a full screen editor built into Universe
>
>
>
>
>
>
>
I'm sure this shiny toy has probably been reproduced many times for universe
and unidata. Maybe someone can open source their version and put it on
pickwiki. Doesn't seem like Rocket Science.
On Jun 15, 2012, at 9:09 PM, Ray Wurlod wrote:
> With you on that!
>
> Bring back RUNOFF!
> __
Cache has a $OPTION that makes identifiers in mvbasic case-insensitive (it has
no effect on string literals or expression evaluation though so "ONE="one" is
still untrue). reserved words are already case-insensitive as they are in
universe and are in unidata with a compile time flag.
I don't th
Just a guess, but I seem to recall that the names of those functions, unlike
most other intrinsic functions, are case-sensitive. Are you casing then
correctly?
On Jun 8, 2012, at 3:28 PM, Boutilier Ron wrote:
> Hi everyone,
>
>
> Always a pleasure to read the useful advice and varying person
I use the extra spacing trick a lot too.
(fyi I work for intersystems) Cache checks the validity of literal format and
conversion codes at compile time. That catches a lot of typos. Cache also
displays an error message at runtime for bad variable codes. It doesn't log bad
codes though. D3 seems
PQN procs are supported in unidata for sure. Just wasn't sure if the
conversions on the MV command work there.
You can use the universe proverb documentation as a starting point for working
with procs in unidata, but there are a lot of differences, including some
fundamental ones. For example,
On May 7, 2012, at 4:45 PM, Kebbon Irwin wrote:
>
> Anyone know/remember how to do this:I have a string in my input buffer
> XX*YYYNNI want to place just YYYNN in my output bufferI feel I should be able
> to do this without writing my own user mode or a program that does a
> procread/procwrit
reen wrote:
> Look at SYSTEM(50)
>
> David A. Green
> (480) 813-1725
> DAG Consulting
>
>
> -Original Message-
> From: u2-users-boun...@listserver.u2ug.org
> [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Ed Clark
> Sent: Monday, March 05, 2012
That will tell you if a particular variable contains a file pointer, which may
be useful if you always use the same name for a file everywhere that it is
used, but it doesn't help if you use different names. For example, if you had a
program that took two file names as command line arguments and
So the guru question is: Does either universe or unidata provide a way to get a
list of opened files from some internal list or pool?
I don't think so. Your application will need to keep track of each file as it
opens them. Somewhere there's a universe technical bulletin that describes a
method
The CAPTURING clause has the disadvantage of capturing what might be a huge
amount of output (not with the COUNT command, but with other commands), so
hushing may be preferable in some circumstances.
Why execute a HUSH? There's a basic HUSH statement that would be more efficient.
What's the purpo
I don' know about Yar's Revenge, but there was a new version of Dig Dug
released last year in the App store that sold a lot of copies.
On Feb 20, 2012, at 9:12 PM, Allen E. Elwood wrote:
>
> I think the proof of the pudding is that you're not going to find ONE vendor
> designing *and advertisi
I know several older programmers who use ED exclusively. It's incredible how
quickly they can jump into code and make complex changes with just a few
keystrokes.
In the course of the day I use ED, a few full-screen non-gui editors
(jed/vi/maybe JET or SED sometimes) and some gui editors (WED, Ca
honestly I don't do a lot of dictionary maintenance. I have each attribute
defined, and translates to define inter-file associations, but If I need a
different width/format/heading or unique expression for a query I use the
FMT/CONV/EVAL keywords and create them on the fly.
On Feb 18, 2012, at
actually your example is pretty easy on current windows versions. Type *.txt
into the windows explorer search box, then select and delete. On the upside,
when you delete the files, you know for sure what directory you are in because
it's in the title bar. On the down side, your search retrieved
ronization. So that's why that
flag is still there.
On Feb 16, 2012, at 9:43 AM, Bob Rasmussen wrote:
> On Thu, 16 Feb 2012, Ed Clark wrote:
>
>> If the universe server is on windows XP (or windows 2000/nt or I think
>> windows server 2003) you can use services.msc to all
sk Manager and there are a
> lot
> f processes with my name on them from failed attempts where some input was
> equested.
> Which also tells me why files weren't closing and updating, and why I was
> etting errors that files were still in use and so on.
>
>
>
>
&
yeah, they left that out. If you look at the docs for the SH. CSH, and VI
commands, they suggest that you look at the unix documentation on those
commands, but they don't mention it in the manual for DOS. Maybe they assumed
you would figure that out by looking at the VOC entry:
DOS
001 V
00
I think it's already been mentioned, but I'll repeat. Both universe and unidata
have the ability to create a cross-reference listing of a program. unidata's is
particularly nice and shows every variable, including what lines the variable
is defined and used on.
but it would still be nicer to ha
Essentially, adding the () or +0 or :"" creates an expression. CALL then passes
the result of the expression instead of the address of the variable. You get
the same result by passing a dynamic array reference like IVAR<1>, though not a
dimensioned array reference like A(5), because each array e
the parenthesis is part of the syntax, normally surrounding the fill part of
the fmt expression and separating it from the justification and descale parts,
for example:
CRT A "R(#20)"
prints A right justified in a field of 20 spaces. It's confusing because that
example will also work wit
You should always work with dates/times/numbers in internal format and only
convert them to external format for display.
Wherever ST.DATE came from, it should have been converted using ICONV:
ST.DATE=ICONV(ST.DATE,"D")
and have a value of 16091 (assuming that the "12" means "2012" and not "1912"
You could try it in DosBox. It's also pretty likely to work in vmware or
virtualBox, but that's not really trying :)
On Nov 21, 2011, at 4:45 PM, George Gallen wrote:
> I just came across my old AP DOS diskettes , and wondered if they would still
> load on Windows 7?
> Thought it might be inter
Accuterm is pretty good at things like transferring a file from the host to the
client and then opening the file in a viewer application like notepad or excel.
Or opening a web browser on the client that connects to a web page on the host.
If the display is mostly a very wide report then you cou
could be someone who learned on unidata too. unidata supports proc, but there
doesn't appear to be a manual for it (If anyone has one, can I get a copy?). If
you were doing new development on unidata, you may easily never have seen a
proc.
On Sep 13, 2011, at 12:56 PM, Wols Lists wrote:
> On 0
yes Include Cache. And Stratus too please, though I think I may have been
one of only 5 or 6 people who actually used it.
On Sep 9, 2011, at 5:06 PM, Dawn Wolthuis wrote:
> Right, it does not have Sequoia under it when it has that name, it flows
> into TigerLogic. There was a ton of informat
mvEnterprise doesn't require any support :) It just runs and runs.
On Sep 9, 2011, at 4:31 PM, Tony Gravagno wrote:
>> From: Wjhonson
>> I mean there's no reason to try to "distinguish"
>> mvEnterprise from mvBase until you're all folded into
>> a single company right?
>
> mvEnterprise is a Un
Both Sequoia and Stratus licensed Pick OA to sell on their fault-tolerant
systems and ran with it in different directions. Sequoia ran a little farther
:) Sequoia ran their pick on top of their proprietary TOPIX operating system
and Stratus ran theirs on top of their VOS os. Sequoia eventually s
Sorry, I'm coming to this late and haven't read the previous emails. Are we
talking about universe or unidata?
universe has a $OPTIONS STATIC.DIM which you can use in any flavor. It is set
by default in pick/in2/reality flavors. With $OPTIONS STATIC.DIM you get
pick-style arrays:
1: array size
doing some hack so that it can deal with wildcards
on converted data, but the result is inconsistent queries.
Bottom line, if you want to do case-insensitive selects, you need to either
move to D3, or use a calculated attribute. For example:
A1
0001 A
0002 1
0003 A1
0004
0005
0006
0007
0008 M
universe or unidata? what flavor?
I don't see how this would have worked in the first place? When you use a query
with literal comparisons, the literal is compared against the pre-converted
value. If the attribute has a reversible conversion then your literal will be
reverse-converted first, fo
Pretty much what Tony said. I could argue some points but this isn't a Cache
list :)
To answer the original post, any questions about U2 mumps should be directed to
quadramed, or maybe to mumps or cache-related google groups (there are
several). You can download Cache manuals from Intersystems.
The behavior varies somewhat by platform. Are you using universe or unidata?
I'm guessing you're on unidata because you are using the "MD" conversion code
(I haven't seen that used much on universe)
The 2 lines:
> X.B.ORIG.APPROVED.AMT = OCONV(V.B.ORIG.APPROVED.AMT, "MD2")
> X.B.ORIG.APPROVED.AM
yeah, you can add $OPTIONS STATIC.DIM to all the programs that you compile in
information flavor, or $OPTIONS -STATIC.DIM to the ones in pick flavor. I would
put the COMMON definition in an include file, along with whichever static.dim
directive that you choose. The choice would be based on WHY
I'm guessing that you have an array dimensioned in the common?
Pick flavor default to $OPTIONS STATIC.DIM, while Information flavor defaults
to $OPTIONS -STATIC.DIM. With STATIC.DIM, each array A(n) takes up n slots in
common, but with -STATIC.DIM, the entire array takes one slot.
On Jun 24, 201
I could be wrong since I've never used NLS on U2, but I think that somewhere
there is a table identifying upper/lowercase pairs.
In the default locale, char(253) is the value mark and doesn't have an
associated case pairing, but in some other locales it is "latin small letter y
with acute" and
I was only suggesting for this one particular case where you were already using
readseq, know that you have the quoting, and have non-line-break carriage
returns defeating readseq :) In general, I use READBLK and parse my own line
breaks--legacy code that goes back to reading fixed blocks of dat
just an idea I haven't thought about too deeply:
Use readseq to read a line, then use the COUNT() function to count the quotes.
If there are an odd number of quotes (mod(2)=1) then add a value mark and read
and append another line. Loop until you have an even number of quotes (because
there migh
be I knew this once, but no longer recall,
> so if anyone knows, I'm curious. Were there any multiVerse production users?
> Was Profile the name of the company that wrote multiVerse? --dawn
>
> On Wed, Apr 27, 2011 at 3:53 PM, Ed Clark wrote:
>
>> sorry, there's al
sorry, there's already a multiVerse. Look at the logo page in International
Spectrum Magazine. It came from the people who brought us pcVerse. I used
pcVerse (a sample version of it came in the box with OS/2 Warp 4) but I think
multiVerse may have been just vapor.
On Apr 27, 2011, at 4:44 PM, M
15 years ago I had a web server written mostly in universe running on unix. A
small piece of C code ran as a demon on port 80, set up some environment
variables, and forked a process that executed uv. The uv process served the
httpd request using mostly basic. I had some GCI code linked into uv
fyi, the "MX" conversion code is only valid in basictype "P". In default
basictype "U" you need to use "MX0C"
On Apr 15, 2011, at 3:38 PM, Rex Gozar wrote:
> EQU ALGORITHM$SHA1 TO "SHA1"
> EQU DATALOC$STRING TO 1
> EQU DATALOC$FILE TO 2
> RESULT = ""
> ERRCODE = DIGEST(ALGORITHM$SHA1, TEXT, DATA
I don't remember all exact details. The biggest problem had to do with universe
proc treating backslash as a special character. They used one or the other as a
delimiter in data that they passed around, and proc ate it. It was something
line this:
0001 PQ
0002 RI
0003 IHA\B\C
0004 F
0005 IH1\2\
A lot of people used to code their applications to "R83" or least common
denominator. The company I worked for 15 years ago had a mix of Adds Mentor,
Stratus OA, Power95 and a small universe system at a satellite office. We
bought a new accounting system that was written to R83. The vendor was a
> view the PDF (via wIntegrate file launch) instead of using the email print
> choice.
>
> -Original Message-
> From: u2-users-boun...@listserver.u2ug.org
> [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Ed Clark
> Sent: Tuesday, April 12, 2011 7:45 AM
> To
Swiftview is nice and has a lot of features. There is also pcl2pdf from
www.visual.co.uk which I know works on AIX.
On Apr 12, 2011, at 10:27 AM, Horacio Pellegrino wrote:
> We use a product called SwiftView converter, that works in Windows and
> Linux.
>
> hp
>
> On Tue, Apr 12, 2011 at 9:39
t is to stay
> credible and open then we should all at least state if we have an influence.
> to post on a U2 list and omit to say that you work for a competitor is not,
> IMHO, acceptable behaviour.
>
> George Land
> APT Solutions Ltd
> U2 UK Distributor
> U2UG Board Member
'm trying to be informative from personal experience.
On Apr 11, 2011, at 1:45 PM, George Land wrote:
> This is a classic example of the issue that came up a few days ago about
> people being open about who they are.
>
> Ed Clark posts from u...@edclark.net, there is also an Ed
Cache gives you a lot of flexibility here. Data is stored in databases. Each
database is a single OS file. But you use the data through a Namespace (called
an "account" in mv). By default all the mv files for a namespace/account are in
a single database. But you can map individual files or even
Yes, Intersystems has developers, support engineers, and salesmen who worked
for other mv vendors. But Cache isn't jbase-centric. It has emulations for all
the MV variants.
Mumps is still around, though most of the large vendors were consolidated by
Intersystems. After that, Intersystems develo
yes, if your application is tied to SB, then your platform options are limited.
But it sounds like a lot of developers want to move away from SB into more
mainstream web-oriented applications. You want to compare building web
applications on Cache vs U2 (and vs non-mv platforms as well, because
Intersystems has a presence in the UK as well. Probably one that will grow some
since NHS Scotland is going to be running their Patient Management System on
Cache.
On Apr 11, 2011, at 4:23 AM, Symeon Breen wrote:
> Tony RE why Unidata instead of Pick
>
>
>
> Well Tony I don’t live in rainy C
security by obscurity has a nice ring to it :) Unfortunately it's the same
security you get by putting an un-pickable lock on a windowed door. The lazy
thief goes on to the next house, but if it's the place you want to get into,
you just do.
On Apr 10, 2011, at 11:03 PM, Dawn Wolthuis wrote:
>
u have F or A correlatives, you may
> wish to consider something other than Unidata as the target to convert to.
> Universe supports them just fine.
>
> -Original Message-
> From: u2-users-boun...@listserver.u2ug.org
> [mailto:u2-users-boun...@listserver.u2ug.org] On Be
a BASIC has the -I option to allow reserved words to be case
> insensitive - "print" would work, although this wasn't available in the older
> versions.
>
> -Original Message-
> From: u2-users-boun...@listserver.u2ug.org
> [mailto:u2-users-boun...@listserver.u
3 because I think it's a real
> pain in the neck.
> Sounded great to start with but when you see the first printouts (i.e.
> invoices) you curse yourself if you didn't turn it off.:-(
>
>
> On 07/04/2011 15:44, Ed Clark wrote:
>> I'm guessing that you are c
Trying to change the title, because this isn't really about conversion in
general anymore.
There are different aspects of case sensitivity of course, and different
platforms have different levels of support. If you haven't actually used D3,
you may not realize how far it carries case insens
I'm guessing that you are converting from d3 to unidata because you are most
familiar with unidata? (and less so with d3). Or is there some other benefit of
moving the application to unidata specifically? If you aren't tied to unidata,
consider Intersystems Cache. D3 migrations to Cache go prett
the universe shell is trying to save the command stack. It puts the stack by
default into &SAVEDLISTS& in an item that concatenates the user's login id and
port number so that when the user logs back on they will get their stack
back--not particularly useful unless the user somehow always gets t
Only in part. Intersystems has many developers, all developing away. I'm trying
not to sound like an advertisement here--just making the point that innovation
is actively happening on at least 1 mv platform.
On Feb 8, 2011, at 10:26 AM, Wols Lists wrote:
> On 08/02/11 13:38, Ed Cla
>> I'll go on a limb and state my belief emphatically that we will
>> never see another new language implemented within the DBMS
>> itself.
Not from universe/unidata/d3 maybe (and sadly), but this is what Intersystems
did with Cache. They added mv basic as a language in the dbms (beside the
origi
I guess Ian was partially right? On jbase, all basic code is precompiled into c
code (which is then compiled into os native object and executables), and if you
wanted to you could just skip the precompiler and write the c code directly.
They provide object code libraries with functions for all t
I've often wished for some sort of block syntax in basic as well. Usually I use
LOOP with an exit, like you do, or a gosub, but when I migrate code to Cache
mvbasic there is a nice construct--the TRY/CATCH block:
TRY
block code
block code
if condition then THROW exception
block code
if
I've never used either, but is there any similarity between UPDATE and universe
REVISE ?
On Feb 3, 2011, at 5:39 PM, fft2...@aol.com wrote:
>
>
>
> Let's say, you happen to stumble across a site that is using Ultimate's
> UPDATE processor. Did any other vendor ever have a migration path off
were developed by different
> companies. cheers! --dawn
>
> On Mon, Jan 31, 2011 at 11:21 AM, Ed Clark wrote:
>> naw. PL/I was an IBM creation. See wikipedia
>> http://en.wikipedia.org/wiki/PL/I
>>
>> On Jan 31, 2011, at 11:47 AM, charles_shaf...@ntn-bower.com wrot
naw. PL/I was an IBM creation. See wikipedia http://en.wikipedia.org/wiki/PL/I
On Jan 31, 2011, at 11:47 AM, charles_shaf...@ntn-bower.com wrote:
> Cleaning out the old room where everything computer related gets sent to.
> Came across a book on a programming language called PL/I. Just taking a
These aren't bugs. They are features :)
Going all the way back to the beginnings with pick and microdata/reality, math
in the F and A processing codes has been integer only. This was on purpose, for
speed I think. And it works fine if you follow the fundamental rule of not
using decimal points i
You can run anything in a virtual machine :)
I run ADDS PC/OS in a virtual machine on mac.
I've run PCVerse under OS/2 in a virtual machine on the mac. (Anyone remember
PCVerse? IBM shipped it on the samples application disk with OS/2 Warp 4).
It looks like the only *native* pick implementations
Intersystems Cache runs on the mac.
On Mar 29, 2010, at 3:38 PM, fft2...@aol.com wrote:
> Are there are native implementations of a Pick system on Macs? Or are all
> Mac connections just terminal emulators into a Windows/Unix system?
>
> I know there is or used to be an emulator by Carnation t
1 - 100 of 163 matches
Mail list logo