Now that's an insightful historical perspective:
From: Mark Johnson
Being from the late 70's MCD crowd . . . I've concluded that people
didn't
like typing EXTRACT with its required but useless trailing parameters.
Thus constantly typing
PRINT EXTRACT(CUST.REC,1,0,0)L#20:
From: Clifton Oliver
Always remember that you should never believe any rule containing
the
words 'always' or 'never'.
Its corollary:
All truth is relative.
and
We will not tolerate intolerance.
taught in all seriousness by Postmodern liberal arts professors.
And exactly opposite
Question (per Mike Henderson's post on this thread):
If a file has been setup for transaction logging,
and you make a copy of a file using OS-level tools (cp, rcp, ftp, tar,
etc.),
how do you remove the TxLg info buried in the file header of the cloned
file?
Answer:
1. Use filepeek to edit
It really matters what flavor you're running.
Which nobody has explained :-)
Basic Ref for MATREAD says:
If the number of elements in array is greater than the number of fields
in the
record, the extra elements in array are assigned empty string values. If
thenumber of fields in the record is
Maybe the Smart People on this list can give a good rule of thumb for
deciding when to READU/WRITE vs. MATREADU/MATWRITE. After all these
years, I still don't have one.
From: Richard Taylor
We are faced with this same situation and the thought behind this
truly baffles me.
I will admit to
Just to pick nits, if you use Transaction Logging, the txlg number is in
the file header, too. Subsequent updates of the cloned file can goof
up your logs.
Is there a similar concern with data replication?
Distributed files also have pathnames buried in them.
-Original Message-
From:
I noticed that they have converted a number of work/scratch files
to dynamic whcih I would think is not such a good idea.
Gerry
Actually, dynamic files are very good for work/scratch files if you use
the MINIMUM.MODULUS parameter. (Use CONFIGURE.FILE ... MINIMUM.MODULUS
command.
I'm
I don't recall ever, ever seeing this before. I'm egotistical enough to
believe that may be true for others, too.
It's in UV 10 User Basic Ref., pg 11-203 (although the bottom of the
page says DELETEU) :
DESCRINFO function
Syntax
DESCRINFO(key, variable)
Description
The DESCRINFO function
From: Stevenson, Charles
From: Jeff Fitzgerald
I'm not running transaction logging but have a few clients who are.
What procedures do you follow when resizing files that are enabled
for
transaction logging? Is RESIZE smart enough to turn logging off and
back on or is this a manual
From: Mark Johnson
How can Basic be limiting. It has everything English (sic)
has and so much more. In fact, there are many reports that
'turn the corner' and cannot be done in English and must be
done in Basic.
If you will stipulate that - generally speaking - the most important
attribute
You know the funny thing is that programmers who use decent editors
don't seem to find the whole indentation thing so challenging. ;^)
Cheers,
Ken
I once declared that UV's FORMAT (as verb or from ED) would be the
standard formatter.
Programmers grumbled.
I said the source is in
I was on a plane TEN years ago and sat beside a Kaiser-Permanente
programming
manager - she said they were leaving Pick.
It's not just MV, it's legacy code. For all it's convolutions and
twists, there is a certain maturity, embedded wisdom if you will, that a
new replacement system will not
I don't get it either. The example looks like a mix of sql mv select
syntaxes (syntaci?)
If you can clarify, I'll bet the answer is yes, there is a good way to
do it with UniQuery.
-Original Message-
From: [EMAIL PROTECTED]
Maybe it's just me, but I read your email multiple times and
a
subroutine to display it nicely, see the list archives
Mike
--
Subject: [UV] SYSTEM(9001) tells you what program you are in where you
came from!
From:Stevenson, Charles ([EMAIL PROTECTED])
Date:03/20/03
After whining about this for, lo, these many years let me
Has anyone written a utility to split out selects for distributed files,
1 phantom for each partfile, then pin the results into 1 select or saved
list?
The purpose would be to decrease total elapsed time for a given select
on the big type30 file.
Was it worth it?
In pseudo-code, something like:
From: Dan Fitzgerald
Subject: Re: [U2] [UV] Distributed SELECTs for distributed (Type30)
files
I'm a bit confused by the subject line, though. Type 30 or
distributed?
In UV, type 30 is dynamic.
I'm trying to think of a clever excuse for why I called distributed
files Type30. Can't.
It's
From: Maresh, Mel
Subject: RE: [U2] [UV] Distributed SELECTs for distributed (Type30)
files
^^
Type27
I colleague wrote and used the following code to select from a
Thanks, Mike.
Now I remember why I forgot about it. I'm in Pick-flavor and this thing
seems to just check to see if the routine is in uv/catdir. We have
pick-style cataloguing.
I was thinking !EXIST() would resolve the question following the same
rules as CALL does for finding a subroutine:
Could your ANS variable have contained Char(255) ?
A string containing Char(255) won't passes the (ANS MATCHES '0N') test.
If anyone can show me a string containing characters other than the 10
digits that passes that test, I'd be much obliged.
cds
---
u2-users mailing list
-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of Stevenson,
Charles
Sent: 08 March 2005 23:26
To: u2-users@listserver.u2ug.org
Subject: [U2] [UV] non-numeric error after passing (ANS MATCHES '0N'
ANS0)
4 times in recent months line these have generated the a set of 4
runtime
You may be right. I've no experience with unicode stuff, working
strictly in English.
I'm happy believing Piers Angliss's too-long-string theory.
Thanks for the input,
cds
-Original Message-
From: Mats Carlid
Sent: Thursday, March 10, 2005 4:39 AM
To: u2-users@listserver.u2ug.org
didn't try any combinations/derivatives. I also saw another
post where a very long number passed the test but failed in the array
assignment.
Colin Alfke
-Original Message-
From: Stevenson, Charles
Could your ANS variable have contained Char(255) ?
A string containing Char(255) won't
Bill,
UV same as D3. UD different (from how I read Colin).
UV:
Yes
No
UD:
Yes
Yes
cds
-Original Message-
From: Bill H.
I ran this on D3 (flashed and unflashed).
:bptest $options ext ; ans = str('1',1024) ; if ans matches '0N' then
crt 'Yes' else crt 'No' ; ans =
-Original Message-
From: John Solie
I'm wondering why y'all are using a ...MATCHES 0N...
Usually, I'll use the NUM() function. The code snippet
in one of the first messages in this thread would be:
BEGIN CASE
CASE NUM(ANS) AND ANS 0
more code here
Or is NUM() something
I think we've squeezed enough blood outa this turnip, except for one
niggling little question.
What's with the creeping number of spaces in the subject line in the
various posts?:
Subject:
RE: [U2] [UV] non-numeric error after passing (ANS MATCHES '0N'
ANS0)
RE: [U2] [UV] non-numeric error
In UniData the '.' and '-' will evaluate as numeric.
You might want to check how these would fall through
your logic in UniVerse.
Colin,
But they don't pass MATCHES '0N' test on UD, do they?
In UD (and UV) NUM('.1') is true, as is NUM(-1), but this program does
not test for NUM().
On UD,
Ezhno,
Thank-you for your help. Comments interspersed.
cds
-Original Message-
From: Ezhno Cheveyo
072: BEGIN CASE
073: CASE ANS MATCHES '0N' ANS0
074: IF MITM5,ANS# THEN
075: IF MITM7,ANS# THEN GOSUB 200 ; IF
Anyhow, the LOCATE command on the
-Original Message-
From: Andrew Lakeland
How about writing a mickey mouse program which goes through
the full ASCII char set and see which one passes the logic.
I've tried that several variations on the theme.
For example:
FOR I = 0 TO 255
IF ( 1:CHAR(I) MATCHES '0N') THEN
I wrote:
Ezhno,
Thank-you for your help. Comments interspersed.
cds
I thanked Ezhno but didn't thank Colin and Andrew. That was rude.
Thank-you Colin Andrew, too!
I don't have an answer yet, but thank-you for helping me brainstorm.
Christopher Robin tried to teach it to me
Donnie,
Here's a post from a year ago that I saved for my own reference.
cds
-Original Message-
From: Stevenson, Charles
Sent: Wednesday, June 23, 2004 8:32 AM
To: 'u2-users@listserver.u2ug.org'
Subject: [UV] SYSTEM(9001) (was: [U2] SYSTEM(49) on UV)
From: Leroy Dreyfuss
SYSTEM
-Original Message-
From: Jacques G.
There is another possibility. Your non-numeric error could be on
a different line of your code. In Universe, if you have a comment
next to an include
line:
$INCLUDE SOMEINC SOMETHING.EQU ;* Include comment
The Raid debugger will not show the
Vincent,
I am not sure why you get a one character answer one way and a full
word the other. I assume that's either in the nature of how your
SQ_CAL_SUB is written called, or from formatting applied differently
in the 2 methods. I think we need more context. Manu Fernandes seems
to be
The relevant code ( I am VERY sure source matches object):
072: BEGIN CASE
073: CASE ANS MATCHES '0N' ANS0
074: IF MITM5,ANS# THEN
075: IF MITM7,ANS# THEN GOSUB 200 ; IF ERR THEN GOTO 15
287: 200:*Subroutine For Password Check.
291: IF MITM7,ANS[1,1] = # THEN
310: IF
Hear, hear!
I wholeheartedly agree with Ken on code reuse.
Let me add that dictionaries should be under source control. A tool
such as SoftWhere that helps you manage where code is reused is also
valuable. I think shortcomings in these 2 areas are stumbling blocks
that keep people from doing
, March 02, 2005 7:48 AM
To: u2-users@listserver.u2ug.org
Subject: RE: [U2] Using an I descriptor in a UniBasic program
Can you provide any web links for SoftWhere
Thanks
James
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of Stevenson,
Charles
Sent: 02 March 2005
Speaking of Y2K-related programs, I too wrote one, but it searched the
VLIST of the program. That way I didn't care about spacing, etc.. You
might find that method helpful.
For example, these 4 source lines compile to the same PCODE:
VLIST CDS.BP MARK
1: NBR = NBR 5'0'R
1 0 : 09C
Bill,
This is an example from Universe. I *think* it works on UniData, too.
Look at lines 1,2, 8,9,10 of the program.
This is a simple example. I use DICT VOC TYPE, because everyone is
probably familiar with it.
ED CDS.BP BILL
15 lines long.
: P
0001: OPEN 'DICT VOC' TO VOC.DCT ELSE
And, just for the record, you can also stack as many commands as you
wish like this:
COMMANDS = 'SELECT FILE'
COMMANDS-1 = 'PROGRAM2RUN'
EXECUTE COMMANDS [ append optional parameters here ]
[snip]
Generally I prefer this syntax
EXECUTE SELECT FILE,//IN. PROGRAM2RUN
[snip]
EXECUTE SELECT
but with that syntax, wont PROGRAM2RUN execute regardless of whether
or not the SELECT returns results?
Yes. Like I said, it was for the record.
I've deleted the original post, but I think the question was about this
syntax sometimes failing:
CLEARDATA
DATA 'PROGRAM2RUN'
EXECUTE
Generally speaking, you've got it right, John.
A few caveats, notes, gotchas:
- Use EXPLAIN keyword to get RetrieVe to show you what it is going to
do.
- If a field is indexed, and no select list is active when the retrieve
command starts, the index will be applied before any actual select
EXPLAIN is Universe-only, not Unidata. Sorry about that.
It goes anywhere on the RetrieVe command line.
If run from tcl, output is to the screen.
-Original Message-
From: Don Verhagen
Charles,
What is the syntax for the EXPLAIN keyword I don't see it documented
in Unidata (6.X)
I
I've found emperically
kill -4 [pid]
*seems* to nicely kill uv pocesses, but I'm not sure why.
Sometimes I have to repeat the command several times before success.
I kinda assumed MASTER OFF must do a kill -4, but have nothing to back
that up, just that UV processes seem to disappear
and the moral of the story is: a Source Control System is very good
idea.
On a more helpful note, the bottom of the program object file contains,
in ascii, the path to the source (unless compiled with -T option).
Cataloging moves that binary file to uv/catdir. So you can reverse
engineer that
Have any of you every seen or heard of a utility that would
take a Pick program and identify what programs were called
by that program and continue to transverse down those
programs until all programs were identified?
SoftWhere, Toledo Assoc. Thorough, excellent product. Cheaply
priced.
Some comments interspersed for those not comfortable with virtual
fields. The primary context here is UD, but I note some differences
between UV UD. Then an example.
cds
From: Anthony W. Youngman
Certainly I would recommend never pointing a virtual field
Really?! Yuck. I never knew that. That's just plain mean.
2nd good candidate today for www.u2ug.org forums enhancements.
I tried the 'EVAL @RECORDn', but can't seem to get it to work.
I've always had to resort to referencing the DICT ID for that loc,
and
we're using UD. Is there an
inefficient compared to what?
And Inefficient measured how?
And is said Efficiency necessarily the significant software quality
attribute when up against competing attributes such as Maintainability,
Reusability, Robustness.
Of course, there's no single right answer. I'm just being curmudgeonly.
Now THIS is getting interesting!
I saw something similar twice. Once a hashed file header was corrupted
with a line that should have been written into a como file. Another
when a file had data from uv/SYS.MESSAGE written in it. Both occurred
in the context of a transaction logging failure. I've
Marco,
I assume you've contacted the vendor(s) by now?
It sounds like a question for IBM support.
Since you have no source code, application support, too.
If when you get an answer, please post it here.
Thanks,
cds
-Original Message-
From: Marco Manyevere
I get the error message
1. T30FILE.
If your MFILES was set to the default, 12, then chances are T30FILE also
needs to be tweaked, given that you have so many dynamic files. I
assume you compared the old DG uvconfig to the new Linux one and can
account for the differences.
2. DISABLE-, BUILD.INDEX.
Try disabling
1. T30FILE.
If your MFILES was set to the default, 12, then chances are T30FILE also
needs to be tweaked, given that you have so many dynamic files. I
assume you compared the old DG uvconfig to the new Linux one and can
account for the differences.
--
My MFILES increase was an 'educated guess'.
-Original Message-
I may soon become a fan of static files if this kind of thing keeps up.
Here is the output that you are referring to.
Dynamic Files:
Slot # Inode Device Ref Count Htype Split Merge Curmod Basemod
Largerec Filesp Selects Nextsplit
01900545 59425
Thank-you, Tom. I was hoping someone would be kind enough to speak up
about Unidata.
Does UD (as does UV) support these alternate syntaxes?:
[]-Syntax. Alternate
Y = X[c,m,n]Y = FIELD( X, c,n,m )
X[c,m,n] = substringX =
TCL verb: RECORD [[DICT] filename [record]]
displays the group the record ID hashed to and indicates whether the
record
was found in the file.
PI had a subroutine called !HASH.ID (if I recall correctly).
I wish we had the same. I used to use it in my own verb I called
PREHASH that would take
I wrote these examples up for our own programmers.
Might as well share them with the group for the newbies' sakes.
__
Suppose X always contains the string
'AB-CD-EF-GH'
for each of the following examples:
Substring extractions:
Y = X[3]-- Y = '-GH'
From: Susan Joslyn
[snip] Gordon, I would think that you could address it with file
triggers... any reason why you can't?
triggers wont work on type 19 files.
Programs have to be type 19 files.
Can PRC work around that one?
cds
---
u2-users mailing list
u2-users@listserver.u2ug.org
To
-Original Message-
From: Ray Wurlod
Subject: RE: [U2] [UD] i-desc mystery
Does UniData have the same MV handling functions as UniVerse?
The OP specified [UD] in the subject line.
I just watched The Fog of War video of Robert MacNamara's (US
Secretary of Defense in 1960s)
Lastly, if you use the REMOVE function within your code,
instead of resetting the pointer, just use a temporary
variable so that the position of the pointer in the passed
in argument is both irrelevant and unchanged. That way
you don't inadvertantly affect a piece of code that's
working
Shawn,
In your mystery, PTD.POSITION.ID references the _e_n_t_i_r_e_ 75th
attribute. It is a dynamic array. It may (or may not) contain value
marks. PTD.POSITION.ID[5,5] does a substring extraction, the 5th thru
9th characters of the array, regardless of where or whether there are
any value
Andrea,
This is a known conversion issue.
I don't think there is a workaround other than to write your own TRANS
function replacement. (If you do, be sure to mimic some of the
efficient caching of already-read records that TRANS does.) If IBM
wanted to, a new UVCONFIG option could be a way
From: Mats Carlid
What's wrong with
FUNCTION ATRANS( F,K,P,X )
RETURN (RAISE(TRANS(F,K,P,X))
END
?
--
looks good, except how about something like:
FUNCTION ATRANS( F,K,P,X )
REMOVE DUM FROM K SETTING MORE ; check for any delimiters
K = K ; * reset rmv ptr in case
YES! You do need the K=K.
Well, either that or
SETREM 0 ON K
That's better. More explicit.
OK, I admit it: I do K=K cuz of 10 yr old PI habits that won't die. PI
didn't have SETREM.
It is important to reset the remove pointer just in case some future
calling program, over which the
Wendy,
(can't believe no response posted yet.)
Cn,Prompt Text will take nth word from command line. Similar to
GET(ARG.) in basic.
In,Prompt Text will take nth word from command line, but if none
supplied, will prompt user.
[control,]prompt text[,options] is general syntax.
P,R,A, s are
Richard,
Besides the 7 character name limit, here's another scenario that would
cause grief. Don't know if it applies to you. It has nothing
particularly to do with VB.
Consider a single session that LOGTOs different accounts:
1. VBNOTEPAGE3 successfully runs in account A. common is
TRUE = (1 EQ 1) ;* If not otherwise defined
FALSE = NOT(TRUE) ;* If not otherwise defined
Just use @TRUE and @FALSE variables U2 already gives us.
VLIST of example in Universe:
1: IF @TRUE THEN CRT 'YES'
1 0 : 2DE testfw 1 E:
1 8 : 046 crtcrlf
From: Dave Schexnayder
Sent: Friday, December 10, 2004 9:13 AM
1) Resizing. When we asked IBM about resizing, their answer was
FAST. However, I am not impressed with the product nor the
expense, are there any alternatives?
1st, I hope you understand and use dynamic hashed files (type-30
New uvconfig parameters can control optional behaviour if a proposed
enhancement would affect backward compatibility. That leaves it to user
to decide which compatibility he wants, PI or older UV.
Thanks, Uffe, for the QUIT paragraph tip. Makes sense. That
workaround/resolution is not
From: Womack, Adrian
Phantoms under PI/Open would execute the ON.EXIT paragraph when
exiting.
what does UD do?
It's pretty annoying that UniVerse doesn't do the same thing (IMO it
should).
Agreed.
Knowledgebase G21418 mentioned by Clif Oliver has been open since
9/4/97, UV 9.3.
For more
Bill,
You're forgiven - but just cuz it's Thanksgiving tomorrow.
Chuck Stevenson, old primate
-Original Message-
From: Bill H.
Goes to show ya...coming from a Pick background I see nothing wrong with
the results. :-)
---
u2-users mailing list
[EMAIL PROTECTED]
To unsubscribe
Verbing weirds nouns. (William Safire quote, I think??)
CDS
P.S. Nouning weirds verbs, too.
-
I think it was AT T who architected unix.
(I checked at dictionary.com, No entry found for architected)
Yes, but according to this link there are about 209,000 references
, November 10, 2004 12:31 PM
To: [EMAIL PROTECTED]
Subject: Re: [U2] UV slow after reboot
Stevenson, Charles wrote:
For several hours after reboot users weep and gnash their teeth over
the slowness . By the end of a normal workday things seem to be back
to normal. I *think* the performance
it was probably the programs themselves being cached.
-Original Message-
From: Stevenson, Charles
For the first day after a reboot, our Universe application seems to be
extraordinarily slow;
[snip]
Has anyone else seen this sort of thing? I'm fishing for ideas here.
The only thing I can
Rod,
Try SET.INDEX.
The path to the index dir is buried in the header of the data file.
SET.INDEX will let you change that.
cds
-Original Message-
From: Rod Hills
Is their a way to copy a data/index files and not have to rebuild the
index.
---
u2-users mailing list
[EMAIL PROTECTED]
One can also create an I-descriptor that returns a true/false (then
SELECT ... WITH XYZ.FLAG ... ). Buried in the I-descriptor can be
explicit control (via IF,AND,OR, parenthases) of whether or not you do
certain expensive tests depending on results of cheaper tests.
You can also apply
Yeah, and get rid of Type1 for dirs in uvhome while you're at it
(catdir, APP.PROGS, BP, INCLUDE, etc.).
And default file types for SAVEDLISTS, PH, COMO, HOLD should be
Type19 when they get created, too.
The rabble has been 'roused.
cds
-Original Message-
From: [EMAIL PROTECTED]
For the first day after a reboot, our Universe application seems to be
extraordinarily slow; dramatically slower than what I've seen on other
systems, and maybe worse than how this system used to be after reboots
(subjective assessment).
For several hours after reboot users weep and gnash their
Barry,
On UV10.0.16, HPUX11i, I get a reasonable distribution:
0001: DIM DISTRIB(10) ; MAT DISTRIB = 0
0002: FOR I = 1 TO 99
0003: THIS.NUM = RND(370570)
0004: DISTRIB( 1+ INT( THIS.NUM/37057 ) )+= 1
0005: NEXT
0006: FOR I = 1 TO 10
0007: CRT
From: Doyen Klein
I found the manual for uv on IBM site
(http://publibfi.boulder.ibm.com/epubs/pdf/25119270.pdf)
Any where I can find ...
1. a more detailed internals description of processor mode and
dispatch type?
Not that I know of. Just the whole VOC section (chapter 3) in the
System
If I understand your problem, you can use INSERT exactly as you would
with dynamic arrays:
INS newValue1 BEFORE FLD(1)1,VMC
INS newValue2 BEFORE FLD(2)1,VMC
INS newValue3 BEFORE FLD(3)1,VMC
etc.
Heck, it is *possible* to INS *attributes*, not just values or
subvalues:
INS
I've been out sick for a month and am jumping into the middle of this
thread.
I gather that someone wants to maintain an index on a TRANS() result,
but is frustrated by UV or UD's failure to maintain the index.
Here is a good workaround using my own verb, XSELECT, first implimented
when we were a
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Stevenson,
Charles
Sent: Thursday, November 04, 2004 12:44 PM
To: [EMAIL PROTECTED]
Subject: RE: [U2] Index problem
[snip]
Then 2 steps at TCL or pararaph:
SELECT ORD.HDR WITH CUST.NAME = XYZ ABC etc
I posted the following on the U2Ug.org Enhancements Forum:
__
[UV] Let UniAdmin's Performance Monitor capture to a file
(UniData may already have this feature?)
Rather than just catch realtime data on a screen, I'd like to be able to
capture it to a file
It was a UD question, but for those interested in UV try SYSTEM(9001):
OBJPATH = SYSTEM(9001)1,2
PGMNAME = OBJPATH[ '/', DCOUNT(OBJPATH,'/'), 1 ]
or alternate syntax:
PGMNAME = FIELD( OBJPATH, '/', DCOUNT(OBJPATH,'/'), 1 )
Forward shash assumes Unix. It also assumes you don't have
Thank-you! I couldn't remember or find it for some reason.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of David T. Meeks
Try using the RECORDLOCKED function...
Locked = RECORDLOCKED( file.variable, record.ID )
It will return you a value indicating
On UV, I would compile it with the -X option and then look at the
cross-reference map.
BASIC BP PGM -X
ED BP.L PGM
---: F Xspace
That would show all lines where variable X is referenced, assigned,
etc.
I think you have something similar in UD.
cds
-Original Message-
From: [EMAIL
LeRoy,
Are we going to be able to get it at the IBM DB2 Information Management
Technical Conference?
Karjala
Isn't there a pre-conference seminar on the topic, too ? ? ?
---
u2-users mailing list
[EMAIL PROTECTED]
To unsubscribe please visit http://listserver.u2ug.org/
Time to move the acronym discussion to u2-community?
To be an Acronym it must be pronounceable as a word,
Otherwise it's an abbreviation.
TLA, pronounced tee-ell-eh is a Three Letter Abbreviation.
pronounced 'tla ( hard for English speakers - do any
Indo-European languages start words
The way I really hoped would be implemented was to have the index
files exist within the file subdirectory, i.e. all files with
indexes
would be similar in structure to Type30, but have a .index
subdirectory within them. So, from then on, the indexes would be
tied to the main data
UV's standard pdf doc set has translog.pdf.
Chapter 8 is Universe commands.
Userref.pdf has descriptions of TCL commands, including txlg cmds.
Look in UV home directory at the motif menus and follow the paths to
see the commands that tx logging submenus execute.
Look at the verbs in the VOC and
IBM could help us here by writing a backward link in the I_INDEX.MAP
that says where the primary file is that the index supposedly
references. If there is a mismatch, error out.
Is there a situation where the current state, where more than one
primary files could point to the same index is a
(Lurking Newbies:
I hope you are reading this thread. This can sting you really HARD!, as
Martin Phillips cites at the bottom. (meaning the bottom of this post
being where MP's comments are, not where you'll get stung. Well,
maybe...)
The issue is this: the path to the index is
fair 'nuff
I once wrote a general purpose LOCATESUB that made LOCATE functionality
available to I-descriptors so I wouldn't have to write a separate
subroutine each time I needed it.
In retrospect, it was overkill. Altho I generally like the technique of
reusable general utility subroutines, I
UV10 STILL DOES NOT UPDATE INDEXES if the index is part of the ID.
From: Norman, David (SAAS)
In UV9.4 CNAME didn't update secondary indexes, which left a bit of
a mess behind. This has now been fixed, probably from 9.6.
Example. A 2-part id: [internal-date]*[something] indexed on date (
@DATA.PENDING is attribute-mark-delimited string holding the data stack.
---
u2-users mailing list
[EMAIL PROTECTED]
To unsubscribe please visit http://listserver.u2ug.org/
Karl,
I'm confused about your removing the hidden .Type30 files:
Here's some detail on the files as they currently exist (I
removed .Type30 from each file to save space):
BOHDR:
-rwxrwxrwx 1 vsifax staff 761856 Jul 19 12:21 DATA.30
-rwxrwxrwx 1 vsifax staff 468992 Jul 19
From: Karl L Pearson
I really didn't remove the .Type30 files from disk, only the
email. That would have been pretty daft of me...
speaking of daft, the joke's on me!
I thought maybe we were swinging at a moving target.
cds
---
u2-users mailing list
[EMAIL PROTECTED]
To unsubscribe please
What Brian Leach said. I wrote a MAKE.LIST verb that creates an active
select list from the contents of the command line. (I think some
versions of Pick have that, maybe by a different name. Can't remember.)
The program loads a dynamic array from Brian's Get(ARG.)s, then does
FORMLIST. The
Oh, joy. It's a stupidity that Prime gave us, Vmark preserved.
(don't know about UD.)
A line that begins with a * is a comment - if it also has a space in
it somwhere after the *..
A line that begins with a * and has no spaces after the * is treated
like a globally catalogued program.
Note,
From: Mark Johnson
Not for nothin' but there are some pretty ugly I-type dict
items with all the SUBS, EQS, REUSE etc stuff.
Before I disagree with some points, let me concur that I-types'
MV-handling can be ugly.
Although Prime traded the ugly correlative syntax for the inherent
simplicity
OK. Can't figure out how to execute the change function
inside an I-desc.
I'm trying to do: CHANGE(@ID,X,(X))
I wrote catalogued:
UTIL.BP CHANGE
01: FUNCTION CHANGE( EXPRESSION, SUBSTRING, REPLACEMENT, OCCURRENCE,
BEGIN )
02: RETURN ( CHANGE( EXPRESSION, SUBSTRING, REPLACEMENT,
301 - 400 of 417 matches
Mail list logo
