dbTalk Databases Forums  

how to suppress warning messages

comp.databases.pick comp.databases.pick


Discuss how to suppress warning messages in the comp.databases.pick forum.



Reply
 
Thread Tools Display Modes
  #1  
Old   
Matt Harting
 
Posts: n/a

Default how to suppress warning messages - 12-15-2005 , 09:11 AM






We are running UV 9.5.1.1f (soon to upgrade to 10.1.11). We are
executing a basic pgm and capturing the output from the pgm. The pgm
capturing the output then parses the output. We are running into a
problem where a system message "Non-numeric data, zero assumed" is also
being captured in the output and causing our parsing pgm problems.

I know we need to clean up the data to fix the error, but is there a
way to prevent the message from being generated by the system?

Thanks . . . Matt


Reply With Quote
  #2  
Old   
None
 
Posts: n/a

Default Re: how to suppress warning messages - 12-15-2005 , 09:51 AM






I heard from a reliable IBM source that a user has asked for an option
to suppress this message. This horrifies me. Let's hope it's not my
bank balance that has been assumed to be zero! (I have seen the same
message in a live banking application).

This message indicates a program error, not a problem with the data.
Fix the program and it will go away.

Martin Phillips, Ladybridge Systems.


Reply With Quote
  #3  
Old   
None
 
Posts: n/a

Default Re: how to suppress warning messages - 12-15-2005 , 10:38 AM



OK, I misread the message... "Non-numeric data", not "Variable
unassigned".

Yes, it probably is the data. My comment regarding users who see this
as an acceptable situation still stands but obviously doesn't apply
here.


Martin Phillips, Ladybridge Systems.


Reply With Quote
  #4  
Old   
Simon Verona
 
Posts: n/a

Default Re: how to suppress warning messages - 12-15-2005 , 12:10 PM



It's one thing that MV seems to lack... data-integrity checking at
update..

I would love to be able to set an attribute as being "numeric only" and have
an error flagged if you assigned a non-numeric to the value at the point of
updating a dynamic array... I know that means a massive change in
architecture (binding a dynamic array to the file that it "represents" as
well as the overhead of getting the data-schema and matching the data)...
but this is one area that most SQL systems excel compared to multivalue.

It would certainly get rid of this problem, which probably is bad data.. The
problem of course, is not in the program generating the error, but in the
program that originally wrote it, if you can find it!

Regards
Simon
"None" <MartinPhillips (AT) ladybridge (DOT) com> wrote

Quote:
OK, I misread the message... "Non-numeric data", not "Variable
unassigned".

Yes, it probably is the data. My comment regarding users who see this
as an acceptable situation still stands but obviously doesn't apply
here.


Martin Phillips, Ladybridge Systems.




Reply With Quote
  #5  
Old   
B Faux
 
Posts: n/a

Default Re: how to suppress warning messages - 12-15-2005 , 12:20 PM




"None" <MartinPhillips (AT) ladybridge (DOT) com> wrote

Quote:
OK, I misread the message... "Non-numeric data", not "Variable
unassigned".

Yes, it probably is the data. My comment regarding users who see this
as an acceptable situation still stands but obviously doesn't apply
here.


Martin Phillips, Ladybridge Systems.

At least it's UV and not QM...
But this looks like a message that would apply only to a mathematical
calculation being attempted on unknown or non-numeric values. Aren't
run-time errors fun!!!

Two things come to mind:
1) Check the dictionary items used to generate the original select list,
does anything get calculated? If yes, then this is where your problem is,
if no then...
2) The calculation is being attempted is in the basic program as Martin
suggests, and hence, this is where your problem is.

Easy fix for option 1) If the problem is in the original select because of a
calculation in a dictionary item, then eliminate those items from your data
capture by adding something like
<ex>
....WITHOUT BOGUS.ATTR = ""...
<ex\>
Or '...BOGUS.ATTR. > "9999"...' even perhaps creating a special dictionary
item that points to the offending attribute for easier isolation.

Easy fix for option 2) Problem in basic program calculation, add a plus zero
test to precede the actual calculation. If the problem is a null that
should be a zero, then this will 'fix' it by forcing the null to zero,
optionally you could force the same thing into a dict item on the original
select which would pass the zero into the program through the capture and
obviate the need to change the program beyond modifying the select
statement.
<ex>
.... BOGUS.ATTR = BOGUS.ATTR + 0...
<ex\>
This usually does not cause an error, if the initial value of 'BOGUS.ATTR'
is null. However, if the original value is indeed non-numeric then you will
happily get two 'Non-Numeric...' errors rather than just one. I say
'happily' because it's predictable, not necessarily desirable.

BFaux




Reply With Quote
  #6  
Old   
Mark Brown
 
Posts: n/a

Default Re: how to suppress warning messages - 12-15-2005 , 02:41 PM



I'm not certain, but you could try using the (S option on whatever command
the capturing program is executing.

Mark

"Matt Harting" <mharting (AT) micahtek (DOT) com> wrote

Quote:
We are running UV 9.5.1.1f (soon to upgrade to 10.1.11). We are
executing a basic pgm and capturing the output from the pgm. The pgm
capturing the output then parses the output. We are running into a
problem where a system message "Non-numeric data, zero assumed" is also
being captured in the output and causing our parsing pgm problems.

I know we need to clean up the data to fix the error, but is there a
way to prevent the message from being generated by the system?

Thanks . . . Matt




Reply With Quote
  #7  
Old   
Excalibur
 
Posts: n/a

Default Re: how to suppress warning messages - 12-15-2005 , 03:21 PM



Ah Yes The good old "Sue ME" option.
RUN BP MYPROGRAM (S) and bingo ALL errors disappear.
Fix the data mate.
Peter McMurray
"Mark Brown" <mbrown (AT) drexelmgt (DOT) com> wrote

Quote:
I'm not certain, but you could try using the (S option on whatever command
the capturing program is executing.

Mark

"Matt Harting" <mharting (AT) micahtek (DOT) com> wrote in message
news:1134659468.309060.93000 (AT) g43g2000cwa (DOT) googlegroups.com...
We are running UV 9.5.1.1f (soon to upgrade to 10.1.11). We are
executing a basic pgm and capturing the output from the pgm. The pgm
capturing the output then parses the output. We are running into a
problem where a system message "Non-numeric data, zero assumed" is also
being captured in the output and causing our parsing pgm problems.

I know we need to clean up the data to fix the error, but is there a
way to prevent the message from being generated by the system?

Thanks . . . Matt






Reply With Quote
  #8  
Old   
Tony Gravagno
 
Posts: n/a

Default Re: how to suppress warning messages - 12-16-2005 , 02:02 PM



"Excalibur" wrote:
Quote:
Ah Yes The good old "Sue ME" option.
RUN BP MYPROGRAM (S) and bingo ALL errors disappear.
Fix the data mate.
Peter McMurray
Hehe, fix data or as I remember how one Trial Balance solved errors:
DEBITS = CREDITS
PRINT DEBITS:" ":CREDITS:" -- In balance !"

T

Quote:
"Mark Brown" wrote
I'm not certain, but you could try using the (S option on whatever command
the capturing program is executing.


Reply With Quote
Reply




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off



Powered by vBulletin Version 3.5.3
Copyright ©2000 - 2012, Jelsoft Enterprises Ltd.