Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-10-07 Thread Edson Tirelli
Lol, I guess I need a cup of coffee. Yes, that was what I meant. Edson 2011/10/7 Michael Anstis > Edson, you mean 5.3.Final? > > sent on the move > > On 7 Oct 2011 21:08, "Edson Tirelli" wrote: > >> >>This is fixed in master. Will be in 3.1 final release. >> >>Edson >> >> 201

Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-10-07 Thread Michael Anstis
Edson, you mean 5.3.Final? sent on the move On 7 Oct 2011 21:08, "Edson Tirelli" wrote: > >This is fixed in master. Will be in 3.1 final release. > >Edson > > 2011/10/7 mike9322 > >> >> laune wrote: >> > >> > There's a bug in the incremental build as done by the KnowledgeAgent. If >> >

Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-10-07 Thread Edson Tirelli
This is fixed in master. Will be in 3.1 final release. Edson 2011/10/7 mike9322 > > laune wrote: > > > > There's a bug in the incremental build as done by the KnowledgeAgent. If > > the > > KA configuration property drools.agent.newInstance is set to false, some > > type information is no

Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-10-07 Thread mike9322
laune wrote: > > There's a bug in the incremental build as done by the KnowledgeAgent. If > the > KA configuration property drools.agent.newInstance is set to false, some > type information is not put into the new KBase. > I added the following comment to the JIRA related to this issue (https:/

Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-07-14 Thread Wolfgang Laun
On 13 July 2011 17:09, John wrote: > Interesting... I have no idea why the KnowledgeBuilder would compile the > DRL > differently than the KnowledgeAgent, There's a bug in the incremental build as done by the KnowledgeAgent. If the KA configuration property drools.agent.newInstance is set to fa

Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-07-13 Thread Wolfgang Laun
On 13 July 2011 17:09, John wrote: > Interesting... I have no idea why the KnowledgeBuilder would compile the > DRL > differently than the KnowledgeAgent, but I don't think I had this problem > in > 5.1.1. Yes, it's OK in 5.1.1; I've verified this with my simplified setup using file: rather tha

Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-07-13 Thread John
Interesting... I have no idea why the KnowledgeBuilder would compile the DRL differently than the KnowledgeAgent, but I don't think I had this problem in 5.1.1. I'm up and running now with the type fully declared in DRL and I can live with that. Thanks for looking into this. -- View this message

Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-07-13 Thread Wolfgang Laun
If the event type is fully declared in DRL, the NPE does not occur. What is the difference between DRL being compiled in the course of a ChangeSet in contrast to being compiled by a KnowledgeBuilder? -W On 13 July 2011 15:05, Wolfgang Laun wrote: > I'm pretty sure that this is a bug. When co

Re: [rules-users] NPE on fact insertion if rule has sliding window

2011-07-13 Thread Wolfgang Laun
I'm pretty sure that this is a bug. When compiling a DRL from a Chenge Set, the resulting package in the Knowledge Base does not contain declare'd types that do not contain fields; it only contains declares with at least one field. This results in a ClassObjectTypeConf with a typeDecl == null, and

[rules-users] NPE on fact insertion if rule has sliding window

2011-07-12 Thread John
Hello, I've been messing with this for days now and can't get to the bottom of this. I'm using drools 5.2.0 Final and setting up my KnowledgeBase with a KnowledgeAgent which is pulling a package from guvnor (also 5.2.0 Final). I do NOT get this error if I use a KnowledgeBuilder and add the exact