1:55 p.m.
We are still preparing some release notes, and overviews for this
release, before we announce it. But for the brave here are some links to
binaries:
http://labs.jboss.com/file-access/default/members/jbossrules/downloads/3....
http://labs.jboss.com/file-access/default/members/jbossrules/downloads/3....
http://labs.jboss.com/file-access/default/members/jbossrules/downloads/3....
The only drl incompatabilities, that i'm aware of, at this stage is:
Primitives are no longer auto-boxed, as we can handle primitives
natively now - so code that expected a wrapped primitive will fail to
compile.
We now implement shadow facts, which means for the moment classes and
methods must not be final and implement a default constructor.
To indicate this is a "project" preview, it is being released under the
drools project name, when it comes time to final, we'll tag and release
as jboss rules.
Mark
Edson Tirelli wrote:
Mark,
Follows the list of main features added to core/compiler I was able
to get from JIRA.
Do you want me to ellabora on each item or the JIRA is enough and
we focus on documentation? I think it would be good to start a wiki
page on migration from 3.0.x to 3.2. What do you think?
[]s
Edson
LANGUAGE IMPROVEMENTS
* New conditional elements: from, collect, accumulate, forall
http://jira.jboss.com/jira/browse/JBRULES-486
http://jira.jboss.com/jira/browse/JBRULES-337
http://jira.jboss.com/jira/browse/JBRULES-341
http://jira.jboss.com/jira/browse/JBRULES-218
* Full support to FOL: nesting of elements now complete
http://jira.jboss.com/jira/browse/JBRULES-218
* Support to connective constraints:
http://jira.jboss.com/jira/browse/JBRULES-216
* Parser improvements:
- Keyword conflicts resolved
http://jira.jboss.com/jira/browse/JBRULES-214
- Better support to line breaks
http://jira.jboss.com/jira/browse/JBRULES-269
- Support to escaped characters inside strings
http://jira.jboss.com/jira/browse/JBRULES-519
* Primitive type support: no more autoboxing/unboxing
http://jira.jboss.com/jira/browse/JBRULES-527
* Support to Templates:
http://jira.jboss.com/jira/browse/JBRULES-330
* Support to additional predicate syntax:
http://jira.jboss.com/jira/browse/JBRULES-556
CORE IMPROVEMENTS:
* Shadow facts:
http://jira.jboss.com/jira/browse/JBRULES-44
* Support to constraints between fields of the same object:
http://jira.jboss.com/jira/browse/JBRULES-515
* Performance improvements
http://jira.jboss.com/jira/browse/JBRULES-484
http://jira.jboss.com/jira/browse/JBRULES-498
http://jira.jboss.com/jira/browse/JBRULES-526
* Specialized "exists" node for performance:
http://jira.jboss.com/jira/browse/JBRULES-547
11:02 a.m.
New subject: [rules-dev] M1 sneak preview
Hmm... shadow facts are going to mess up quite a bit of my code, then. I
have several classes which subclass a base class with several final methods
(to ensure that the contract of these classes cannot accidently be broken --
so, we have a few final methods in an abstract base class and several
methods subclasses defined to customize the behaviors).
Is there a way to get around this? I'm reluctant to muck up my POJO logic,
but I do understand the advantages of shadow facts.
--- Michael
1:34 p.m.
New subject: [rules-dev] M1 sneak preview
There are ways, but its always a compromise. To work directly with pojos
you need to retract the value before modification and then assert it
after modification. We can possibly instrument the drl to do an auto
retract and assert on using a setter, but it means you cannot modify a
fact external, without manually retracting and asserting it. This would
also deprecate modify() in the drl. Over time we'll try and provide
several ways to slice and dice this, as I don't think there is one clear
answer for everyone.
Mark
Michael Suzio wrote:
Hmm... shadow facts are going to mess up quite a bit of my code,
then. I have several classes which subclass a base class with several
final methods (to ensure that the contract of these classes cannot
accidently be broken -- so, we have a few final methods in an abstract
base class and several methods subclasses defined to customize the
behaviors).
Is there a way to get around this? I'm reluctant to muck up my POJO
logic, but I do understand the advantages of shadow facts.
--- Michael
------------------------------------------------------------------------
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
6497
days inactive
6498
days old
2 comments
2 participants
participants (2)
-
Mark Proctor -
Michael Suzio