Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » Epsilon » Rule reach and extension consistency in Flock and ETL
icon3.gif  Rule reach and extension consistency in Flock and ETL [message #1131287] Thu, 10 October 2013 11:39 Go to next message
Horacio Hoyos is currently offline Horacio HoyosFriend
Messages: 194
Registered: October 2009
Location: UK
Senior Member

Hi Everyone,

I hope this post creates an interesting discussion around the subject.

It might be too late (and surely break all the current Flock scripts out there), but for consistency (at least with ETL) the semantics of rule reach and extension in Flock should work the other way round, so migrate rules should:

1. Be "not greedy" by default (hence there should be a greedy annotation and not a strict)
2. Not inherit by default, so u will need an extends keyword and possibly an abstract annotation (if you want to skip instances of super-types from being migrated)

It would be nice to hear ideas/experience regarding this topic Very Happy


Horacio Hoyos Rodriguez
Epsilon Committer
Research Assistant
University Of York
Re: Rule reach and extension consistency in Flock and ETL [message #1131911 is a reply to message #1131287] Thu, 10 October 2013 22:05 Go to previous message
Dimitris Kolovos is currently offline Dimitris KolovosFriend
Messages: 1520
Registered: July 2009
Location: York, UK
Senior Member

Hi Horacio,

I can see the benefits of making these changes but I agree with Louis that overhauling the default semantics of the language at this point is probably not an option. One option for a future version of Flock could be to provide a runtime flag that inverts the default semantics.

Previous Topic:Iterator in an attribute
Next Topic:Type of input model in ETL project
Goto Forum:

Current Time: Mon Apr 23 15:45:34 GMT 2018

Powered by FUDForum. Page generated in 0.01915 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software