Bug 400710

Summary:

[1.8][compiler] synthetic access to default method generates wrong code

Product:

[Eclipse Project] JDT

Reporter:

Stephan Herrmann <stephan.herrmann>

Component:

Core

Assignee:

Stephan Herrmann <stephan.herrmann>

Status:

RESOLVED FIXED

QA Contact:

Severity:

normal

Priority:

CC:

jarthana

Version:

4.3

Target Milestone:

BETA J8

Hardware:

All

OS:

All

Whiteboard:

Bug Depends on:

Bug Blocks:

380501

Attachments:

Description	Flags
tentative fix	none

Description Stephan Herrmann

2013-02-13 11:13:43 EST

The following code generates wrong code:

interface I {
    public void foo() default {
        System.out.println("default");
    }
}
public class C implements I {
    public static void main(String[] args) {
        C c = new C();
        c.foo();
    }
}

At runtime this produces:
Exception in thread "main" java.lang.NoSuchMethodError: java.lang.Object.foo()V
	at C.foo(C.java:1)
	at C.main(C.java:9)

The reason is in the way a synthetic method (SuperMethodAccess) for foo is generated: it is assumed that the super method must be in accessMethod.declaringClass.superclass(), which leads to trying to invoke java.lang.Object.foo(), which obviously doesn't exist.

Comment 1 Stephan Herrmann

2013-02-13 11:26:00 EST

Created attachment 227029 [details]
tentative fix

This patch fixes the immediate issue but is not sufficiently tested.

Comment 2 Stephan Herrmann

2013-02-15 05:46:17 EST

Jay, does my patch help for your test re static methods in interfaces?
Should I test and release this one first, or wait for you?

Comment 3 Stephan Herrmann

2013-02-16 05:04:45 EST

If released the immediate patch via commit c22ae4e2acd996824ff94a2b0edaa0df69d9b3b1

Further cross-checking deferred to bug 391376.

Comment 4 Jay Arthanareeswaran

2013-02-18 04:20:38 EST

(In reply to comment #2)
> Jay, does my patch help for your test re static methods in interfaces?
> Should I test and release this one first, or wait for you?

Yes, Stephan. Thank you!