Summary: | Generics in parent class removes markers from child classes | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | [Tools] AspectJ | Reporter: | William Headrick <eclipse> | ||||||
Component: | Compiler | Assignee: | aspectj inbox <aspectj-inbox> | ||||||
Status: | RESOLVED FIXED | QA Contact: | |||||||
Severity: | critical | ||||||||
Priority: | P2 | CC: | aclement | ||||||
Version: | 1.5.3 | ||||||||
Target Milestone: | 1.6.2 | ||||||||
Hardware: | PC | ||||||||
OS: | Windows NT | ||||||||
Whiteboard: | |||||||||
Attachments: |
|
Description
William Headrick
2007-07-24 18:03:53 EDT
Created attachment 74512 [details]
Test Project
The files in this jar should show the problem... I marked the lines where I think there should be a marker (advice) with the comment "// Should be a marker here..."
Created attachment 74513 [details]
Test Project
The files in this jar should show the problem... I marked the lines where I think there should be a marker (advice) with the comment "// Should be a marker here..."
I forgot to say that if you remove all of the <...> from the code, all of the markers/advice is correct. The problem only occurs with the use of Generics. :( Wish I'd got to this bug sooner. Quite serious. I've now fixed it. thanks for the clear test case. Can't believe more people haven't hit it. The problem was not deferring to the backing generic member to determine annotation matches when working with a parameterized method. |