You are not logged in.

Dear visitor, welcome to Palo Community Forum. If this is your first visit here, please read the Help. It explains in detail how this page works. To use all features of this page, you should consider registering. Please use the registration form, to register here or read more information about the registration process. If you are already registered, please login here.

  • "vargabor" is male
  • "vargabor" started this thread

Posts: 126

Date of registration: Nov 13th 2005

Location: Budapest, Hungary

Occupation: Wine production

  • Send private message

1

Tuesday, October 20th 2009, 3:58pm

Palo Marker Error - markered rule gives random false values

Markered rules sometime give wrong results.

i.e.

['09/08'] = ['2009'] / [['2008']]



2008 2009
values 1372476 1494248

['09/08'] equals normally 1,09, but sometimes it gives a random number like 342121144

Error is impossible to reproduce, server restart helps mostly (markers had been set before server start up).

Does anybody have the same problem?

System OS: WXP PRO SP3
Client: Excel 2003 SP3
Palo version: 3.0 Build 3933

This post has been edited 1 times, last edit by "vargabor" (Oct 20th 2009, 3:59pm)


  • "Michel" is male

Posts: 154

Date of registration: Jun 1st 2007

Location: Netherlands

  • Send private message

2

Tuesday, October 20th 2009, 4:51pm

RE: Palo Marker Error - markered rule gives random false values

Hi,

Results for calculations on C: level using markers are undefined. In PALO 2.5 this did not appear to be a problem (it seemed to work correctly), but in PALO 3.0 this doesn't work.
In your case your rule now should be split into two:
['09/08'] = N: ['2009'] / [['2008']]
['09/08'] = C: ['2009'] / ['2008']

It's a pain, but it works.

Michel

  • "vargabor" is male
  • "vargabor" started this thread

Posts: 126

Date of registration: Nov 13th 2005

Location: Budapest, Hungary

Occupation: Wine production

  • Send private message

3

Tuesday, October 20th 2009, 5:19pm

RE: Palo Marker Error - markered rule gives random false values

Thanks for the answer.

We had the same problem in 2.5 as well.

I can't understand this feature, usually it works and sometimes it goes wrong.

I will try to double the rules, but this will affect the cube performance.

  • "Michel" is male

Posts: 154

Date of registration: Jun 1st 2007

Location: Netherlands

  • Send private message

4

Tuesday, October 20th 2009, 9:38pm

RE: Palo Marker Error - markered rule gives random false values

Quoted

Originally posted by vargabor
I will try to double the rules, but this will affect the cube performance.


Hi,

Why do you think this will affect the cube perfomance? I would think this is still the same rule you had, but with a different notation (whatever notation used, the marker will always only be applicable to the N level).

Michel

Cuitlahua

Professional

  • "Cuitlahua" is male

Posts: 32

Date of registration: Feb 27th 2009

Location: Hungary

  • Send private message

5

Tuesday, October 20th 2009, 11:59pm

Hi All!

Anyhow, the marker technology seems to suffer from several problems. I have partly migrated a model from an another tool and I always met strange behaviour when I used markers.
I have reported for example 0004658 in the bug tracker. I would not call problem No 0003024 a minor issue either. Using markers within If function also resulted in weird events (marker was sent from the "true" side when the expression was false and the rule that came later to handle the false case remained unmarkered). In a former post I mentioned that ordering seemed to work incorrectly for markered values (maybe solved since then, I have not checked again).
I hope that the markers will work well soon, otherwise Palo degrades to a hobby-tool.

C.

  • "vargabor" is male
  • "vargabor" started this thread

Posts: 126

Date of registration: Nov 13th 2005

Location: Budapest, Hungary

Occupation: Wine production

  • Send private message

6

Wednesday, October 21st 2009, 11:05am

RE: Palo Marker Error - markered rule gives random false values

Quoted

Originally posted by Michel

Quoted

Originally posted by vargabor
I will try to double the rules, but this will affect the cube performance.


Hi,

Why do you think this will affect the cube perfomance? I would think this is still the same rule you had, but with a different notation (whatever notation used, the marker will always only be applicable to the N level).

Michel


That's true.

Rate this thread