Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 468471

Summary: blindly fixed aggregation build due to broken categories
Product: Community Reporter: David Williams <david_williams>
Component: Cross-ProjectAssignee: Ben Gamble <bgamble>
Status: RESOLVED FIXED QA Contact:
Severity: normal    
Priority: P3 CC: bgamble
Version: unspecified   
Target Milestone: ---   
Hardware: PC   
OS: Linux   
Whiteboard:
Attachments:
Description Flags
screen shot showing original model 'validate' errors. none

Description David Williams CLA 2015-05-27 09:13:47 EDT 
The aggregation build has been broken for a while based on a "broken model". 

This results when someone changes something like categories in their own file, but not the simrel.b3aggr file. 

By "blindly" I mean I fixed just by removing broken things from categories. 

The "owners" of these components will have to correct categories in the way they intend them to be. 

This primarily was related to BIRT, but I think a few in SOA too ... and, some others? I'll attach screen shot of original error, not sure it will help too much, since in "emf serialization" form.
Comment 1 David Williams CLA 2015-05-27 09:14:50 EDT 
Created attachment 253837 [details]
screen shot showing original model 'validate' errors.
Comment 2 Ben Gamble CLA 2015-05-28 00:47:29 EDT 
I don't know how your stuff works but this is the thing that says it's supposed to validate aggregation and it reported success when I fixed up my categories so if I don't hear otherwise I will assume that I fixed them up correctly.

https://hudson.eclipse.org/simrel/job/simrel.mars.runaggregator.VALIDATE/314/
Comment 3 David Williams CLA 2015-05-28 00:56:01 EDT 
(In reply to Ben Gamble from comment #2)
> I don't know how your stuff works but this is the thing that says it's
> supposed to validate aggregation and it reported success when I fixed up my
> categories so if I don't hear otherwise I will assume that I fixed them up
> correctly.
> 
> https://hudson.eclipse.org/simrel/job/simrel.mars.runaggregator.VALIDATE/314/

That's how it works. "validate aggregation" ALSO does a "validate" on the model. 

If you are in b3 aggregator editor, "validate" is a very fast operation. 
And "validate aggregation" is a fairly long operation (but does 'validate the model', and in addition validations all the requirements are met, and "fit together", conceptually similar to "can they all be installed into the same IDE". 

Thanks, 

I think I removed a few "dead features" in the SOA category ... but that might have been the correct fix? So, fine with me to leave this bug as "fixed".