Summary: | [search] Strange search engine behaviour | ||
---|---|---|---|
Product: | [Eclipse Project] JDT | Reporter: | Dirk Baeumer <dirk_baeumer> |
Component: | Core | Assignee: | Frederic Fusier <frederic_fusier> |
Status: | VERIFIED FIXED | QA Contact: | |
Severity: | minor | ||
Priority: | P3 | ||
Version: | 3.1 | ||
Target Milestone: | 3.1 M5 | ||
Hardware: | PC | ||
OS: | Windows XP | ||
Whiteboard: |
Description
Dirk Baeumer
2004-12-13 18:58:09 EST
foo(null) is an ambiguous call (you should have a compiler errors for that, shouldn't you?). So, the current behavior is to store a ProblemReferenceBinding with an original MethodBinding to the first method which may match the call. I guess the first method in the list of class A is foo(String)... To demonstrate this, write foo(Integer) before foo(String) in class A... and then you get 2 references for foo(Integer) and 0 for foo(String)! I'm not sure if we can change this behavior but, I'll have a look on it if time permit. Waiting this, I reduce the severity as this behavior disappears when compiler errors are fixed... Fixed. Now search engine reports 2 potential matches for both methods. [jdt-core-internal] Change done in MethodLocator.matchMethod(MethodBinding). Test case added in JavaSearchBugsTests: testMethodReferenceBug80890() Verified for 3.1M5 |