Summary: | [assist] Wrong proposals order when completing on an item of an array | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | [Eclipse Project] JDT | Reporter: | Frederic Fusier <frederic_fusier> | ||||||
Component: | Core | Assignee: | David Audel <david_audel> | ||||||
Status: | VERIFIED FIXED | QA Contact: | |||||||
Severity: | normal | ||||||||
Priority: | P3 | CC: | jerome_lanneluc | ||||||
Version: | 3.4 | ||||||||
Target Milestone: | 3.4 M6 | ||||||||
Hardware: | PC | ||||||||
OS: | Windows XP | ||||||||
Whiteboard: | |||||||||
Attachments: |
|
Smaller test case: package p; public class Test { void foo(Object a) { Object zzvar1 = null; int zzvar2 = 0; if (a == zz) { } } } zzvar2 shouldn't be the first proposal. Created attachment 92412 [details]
Proposed fix
Released for 3.4M6 Test added CompletionTests#testCompletionAfterEqualEqual1() 'type' is no longer proposed at the end of the list. However this could still be improved. I entered bug 223899 for this. In the meantime, marking this bug as verified for 3.4M6 using I20080325-0100 |
Created attachment 92055 [details] Project to reproduce the problem Build I20080305-1100. Unzip the attached jar file, import the 'Completion' project in a workspace, open the class 'Test' and complete where indicated. You'll get a long list of proposals... The one I was looking for was 'type', and I'm really surprised to find it only near the end of the list. As I was in an identity test with ASTNode, I would not expect to see all 'int' proposals placed before the 'ASTNode' one...