Maybe passing a matrix in parentheses DOES pass by value - it passes the
value of the matrix reference! Which still contains pointers to the
matrix elements ... :-)
With respect to the docu, I suspect it's meaningless when talking about
matrices because it's not talking about what you think it
Mike Rajkowski [EMAIL PROTECTED] wrote...
0005: CALL P2(X(100), MAT X)
On a similar vein there's the mistake of using common block variables as
arguments - at least on UD and Pick (can't comment on UV).
Simon
--
u2-users mailing list
[EMAIL PROTECTED]
Well, Ray has spoken, so I guess that makes it gospel. Top posting is
wrong-posting, and only newbies who don't know any better do it. Speak for
yourself, Ray - I, and many others, would rather not have to scroll down in each post
to see the new material. I want to see it at the top, and scroll
Hmmm...
I downloaded the complete Unidata 6.0 Personal Edition to my PC, uninstalled the old
Unidata ODBC driver, installed the NEW Unidata ODBC driver, and retried. Still
getting the same message when I try to pull the data into Excel 2000 (as 'Get External
Data'): Non-numeric was found
I have to say, I just loaded Outlook 2003, and sort EVERYTHING by
conversation... I'm kinda sorry people are now making efforts to complying
with 'netiquette', because 'top posts' are SO much easier to use in this
venue - it is awesome to be able to run through the whole set of topics in
the
Except - the most important thing is COURTESY. To OTHER PEOPLE.
I'm amazed I've stayed out of this as long as I have :-)
The problem is that top-posting encourages people to not trim. That
leads to, typically, maybe 90% of the average post being junk that's
been seen before?
Add to that, it is
As much as I hate to defend Microsoft...and I do hate defending Microsoft...they have
patents on their implementation of XML, not XML in general. This has no effect on
XML. It's akin to someone patenting a method of casting a die out of aluminum. That
patent doesn't affect aluminum, nor does
Heh - I notice you top posted, and didn't trim. There must be a deeper
meaning there... =:-o
DW
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Anthony Youngman
Sent: Wednesday, February 18, 2004 8:31 AM
To: U2 Users Discussion List
Subject: RE:
Good question. I'm not exactly sure where I got this version - I can't
remember if I've reinstalled the drivers or not. I did get a .dll from IBM a
while back (when I had similar problems). That could be where I got it.
I keep thinking there was something else that might have caused that error
Top Posting vs. Bottom Posting - didn't I see this in Gulliver's
Travels with Big Endians and Little Endians? Where is our technical
moxie boys and girls? With this group, I figured this would morph into a
discussion on which of us could whip out a better e-mail client for
parsing the
Hi,
I've tried the demo account, and get the same error.
I'm trying a different avenue. A freshly set up PC, where I'll only install the
Unidata ODBC driver and Microsoft Excel. I'm just making sure that it isn't some
other program that I may have on my PC having overwritten a file
I am using Visual Schema Generator to SQLize some UniData files so that
I can access them through an OLEDB connection to Unidata. I can enter
the file name in without any problems and pull up the File/Table Action
screen. I click on Map Attributes and the File Attributes screen comes
up. The
You have to go the VAR route now. We've found that IBM won't even help
license U2 products when the end user calls.
I could send you the UCI.DLL that IBM sent me. However, I'll have to check
that it's one of the files in the 'distribution' package and it's OK for me
to send it. Ibm has already
Hi,
That would be great, if it's ok to do it. You can reach me at [EMAIL PROTECTED]
Thanks again for the help.
- Dave
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of [EMAIL PROTECTED]
Sent: Wednesday, February 18, 2004 11:01 AM
To: [EMAIL PROTECTED]
I have a client with a printing problem. Running UniVerse 10.x on HP-UX.
The client is able to print except if he is logged in as root. When
logged in as root the print jobs appear to go off to data hell and don't
show up in the spooler anywhere.
Sounds like it might be a permissions problem or
Ian Stuart wrote:
Firstly, the performance of OOo is really good and we have not been able
to match it with MS Office on a thin-client platform. At a customer site
we have successfully migrated 45 users onto a single Intel-based server
running Linux and OpenOffice for thin-client based users.
I haven't seen the problem specifically, but my first thought would be to
look at the differences in the PATH environment variable between root and a
normal user.
-Original Message-
I have a client with a printing problem. Running UniVerse 10.x on HP-UX.
The client is able to print
I have a set of small UniData files that I would like folks to be able to
receive via e-mail and then install. The requirements are:
1) OS can be most flavors of Windows or UNIX
2) DB can be UniData or Universe
3) I would like to have a single zipped up file for distribution, but if
two or more
1) Make your source files Universe type 19s (eliminate the whole fnuxi
thing)
2) Gzip them all up along with a paragraph to be placed in the voc file to
resize them to whatever filetype you prefer (I'd go with dynamic for
anything other than source code files, but thats just me).
Installation
Gzip is gnu licensed and available for both Unix and Windows platforms. You
can get them here:
http://www.gzip.org/
-Original Message-
Thanks -- I zip all the time, but never have I gzip'd, so I'm green. I'll
have to figure out what tools are used for that since it appears that my
O bother ... I clicked before asking my other question ...
If I create the files in UniData instead of Universe, what should the
CREATE-FILE statement look like or is it important to make them type 19
Universe files? Can the same paragraph to alter them to dynamic work in
both UV and UDT or not
Dawn,
The UniData DIR file is the functional equiv. to UniVerse Type 19.
Dawn M. Wolthuis wrote:
O bother ... I clicked before asking my other question ...
If I create the files in UniData instead of Universe, what should the
CREATE-FILE statement look like or is it important to make them type
Is there any way to directly access information stored in the file header? I
use the information returned by the FILE.USAGE command to determine if a
static file can be archived or if it needs to be checked to see if it needs
to be resized.
Vance
-Original Message-
From: [EMAIL
Colin, you are great!
The updated uci.dll file worked great, and I can import a Unidata file into Excel
now!
(Now back to VB.NET for the program.)
Thanks so much! :) :) :)
- Dave
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of Dave Tabor
Sent:
I believe there is. There is a UNIVERSE.INCLUDE file FILEINFO.H which
includes a definition of what would be in each dynamic array field when using:
FILEINFO(Fvar, 99) or FILEINFO(Fvar, FINFO$HDRLAYOUT)
For example if you executed:
FileInfo = FILEINFO(Fvar, 99)
you'd have:
FileInfo1 = File
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On Behalf Of Glenn Herbert
Sent: Tuesday, February 17, 2004 10:32 AM
To: [EMAIL PROTECTED]
Subject: RE: UV write fatal error
Ok. So there WAS a bscan done on the T25 file prior to attempting the
write. Since this is
So what exactly is the particular data item that is causing the grief?
At 02:50 PM 02/18/2004, you wrote:
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On Behalf Of Glenn Herbert
Sent: Tuesday, February 17, 2004 10:32 AM
To: [EMAIL PROTECTED]
Subject: RE: UV write
Cool. I found the include file with the following comment:
* 04/23/99 24742 GMH Add special FINFO$HDRLAYOUT and access keywords ;-)
But my include file only includes keys up to the following line:
EQUATE FH$RESIZEPATH TO 95
I am running 9.6. Were the extra keys added in a later release
Hmmm. It's a very curious situation. Your data doesn't appear to be
anything out of the ordinary (though it's neat about your munging).
Have you tried using the blook utility to scan through the btree file to
see if anything is out of the ordinary? From the OS level, just issue
blook
Ray:
I don't much agree with your analysis. Definitive statements such as
...top-posting still makes things more difficult for just about everybody
else and ...there really are no valid reasons to top-post,... and
...Insisting on top-posting...is just plain rude and self serving. are
incorrect
Tuesday, March 16th, at the Spectrum show, we will be having an
informal get together at 5:00 PM. We'll all meet in the Exhibit Hall at
the Key Ally table (right near the door, across from IBM) as the
exhibits break for the evening. Unfortunately, we can not fund dinner
(this time) but we'd
Tuesday, March 16th, at the Spectrum show, we will be having an
informal get together at 5:00 PM. We'll all meet in the Exhibit Hall at
the Key Ally table (right near the door, across from IBM) as the
exhibits break for the evening. Unfortunately, we can not fund dinner
(this time) but we'd
Maybe passing a matrix in parentheses DOES pass by value - it passes the
value of the matrix reference! Which still contains pointers to the
matrix elements ... :-)
With respect to the docu, I suspect it's meaningless when talking about
matrices because it's not talking about what you think
Hello, Stuart.
I say this is a compiler flaw rather than a documentation flaw. While it
may prove challenging to dig through the parse-tree to figure out when
to make use of the implicit temporary, it should be done for matrix
components as it is for scalars. The compiler can do it far more
EQUATE CONST LIT ':'
CALL xxx(CONST zzz(n))
Good call!
**
This email message and any files transmitted with it are confidential
and intended solely for the use of addressed recipient(s). If you have
I say this is a compiler flaw rather than a documentation flaw. While it
may prove challenging to dig through the parse-tree to figure out when
to make use of the implicit temporary, it should be done for matrix
components as it is for scalars. The compiler can do it far more
reliably for us
I am looking for a version control application for U2 software development.
Ideally it would have a check-out/check-in function, ability to track
changes, migration dates, etc
Does anyone have any suggestions?
Jeff Steinmetz
Vice President, Store Systems
Guitar Center, Inc
5795 Lindero
I am looking for a version control application for U2 software development.
Ideally it would have a check-out/check-in function, ability to track
changes, migration dates, etc
Hi Jeff,
Have you looked at CVS? It's available for a variety of platforms and
should handle what you're looking for.
Jeff,
Someone in Oz - maybe PRISM www.pri.com.au IIRC - does a U2 integration with
Microsoft Source Safe.
And someone on this list has created and sells a U2-specifi product. Check
the list archives.
HTH
Mike
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On
Hello, Stuart.
Hmmm. The technique shown deals with string, real and whole number
variables. What about files, subroutines, sockets, etc? And performing
string concatenation would coerce numerics into strings, which may slow
things further down stream.
Best regards,
Gyle
-Original
Rats -- Spectrum coincides with the Datatel User Group conference this year.
So, if you happen to be going to DUG in DC rather than Spectrum in LV,
please let me know you are a u2-users member if you see me working in the
Entrinsik (Informer) booth or find me sitting next to the lobby in need of a
Jeff,
You can also look at Susan Josyln's offering:
http://sjplus.com/sjphomepage.html
- Charles Barouch
[EMAIL PROTECTED]
Larry Hiscock wrote:
I am looking for a version control application for U2 software development.
Ideally it would have a check-out/check-in function, ability to
on [EMAIL PROTECTED] wrote::
I am using Visual Schema Generator to SQLize some UniData
files
[snip]
I am getting the
message Died in UCI::SQLExecDirect() with SQLSTATE S1001, Native
error:0 [IBM][SQL Client][UNIDATA] Invalid procedure call:
SCHEMA_ATTRIBUTE_MAP. Does anyone know what
James Hogan wrote:
From time to time we have customers who break a Unidata
parameter and the
program they are running will crash with errors such as No
more entries in
MI table in 'LCT -n'.
[snip]
I have had a look at the commands sms, gstt and lstt.
With some clever
scripting these
44 matches
Mail list logo
