[great-users] 2 GReAT bugs

[Attila Vizhanyo]

Answers to problem #1:
You are correct in reusing existing rules through either GReAT
ExpressionRefs or GME's Subtype feature.
But an important rule is that the referred object must stay in the
*folder* of the reference.
 
There is an added constraint, which can be considered as a bug, and be
fixed if it is important for you.
The referred rule must not be isolated, that is it must be part of the
transformation.
 
What you do in this example is basically invoke the same rule three
times throughout the transformation.
If you make the referred rule part of the transformation, and
consequently move it to the Transformation1 folder, and create two
references or subtypes, then the output will contain three "Item2"
objects.
 
Answer to problem #2:
There was a bug in the GenerateGR interpreter which discarded the
'forAll' attribute for Case-s.
This has been fixed. 
 
Thanks,
Attila
 
-----Original Message-----
From: great-users-bounces at list.isis.vanderbilt.edu
[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of
Tivadar Szemethy
Sent: Sunday, November 27, 2005 9:30 PM
To: great-users
Subject: [great-users] 2 GReAT bugs
 
Hi,
 
The JIRA server is in a bad mood tonight and doesn't accept connections
so this goes to the list.
 
Pls. check the attached transformation (RuleReference.xme and the input
file House.xme. They use the HouseModel and Order paradigms from
GReAT\Samples).
 
Problem #1:
 
The resulting output file contains one "Item2" instead of three. I
believe this is due to some problem with handling ExpressionRefs. 
There's one Reference to Rule2 and two SubTypes of Rule2. (from Folder
Transformation2)
 
Problem #2:
 
Test case: Both the Case's and Rule3's 'forAll' attributes are set to
'False'. Still, Rule3 fires for 6 times instead of one.
If this is not a bug, please let me know how to execute Rule3 (following
a Test/Case) for only one match.
 
The results are the same with GReAT 1.5 and also with the previous
release.
 
Thanks,
 
Tivadar
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.isis.vanderbilt.edu/pipermail/great-users/attachments/20051129/0cfbb3f0/attachment.htm