[rfc][icedtea-web] More testing for IcedTeaScriptablePluginObject.cc + memory leak fixes

[Pavel Tisnovsky]

Hi Adam,

both patches looks correct, ok for HEAD :-)
I suppose that your later patches are applicable only when that two patches are applied/pushed, right?

Cheers,
Pavel


----- Adam Domurad <adomurad at redhat.com> wrote:
> Hi all, I am looking into making the 'getvalue' NPAPI hook non-blocking. 
> This involves the Java-in-Javascript wrappers, so I'm trying to put more 
> of it under test.
> 
> Included is a leak fix and extra C++ test infrastructure + tests. It is 
> working very well at detecting leaks due to 'new' without a 
> corresponding 'delete'.
> 
> Before the fix two tests should report 1 leak with 'new'. After fix all 
> tests should pass.
> 
> Test ChangeLog:
> 2013-XX-XX  Adam Domurad  <adomurad at redhat.com>
> 
>      * plugin/icedteanp/IcedTeaPluginUtils.cc: Add global state clearing
>      utility functions.
>      * plugin/icedteanp/IcedTeaPluginUtils.h: Same.
>      * tests/cpp-unit-tests/IcedTeaScriptablePluginObjectTest.cc: Test
>      scriptable object creation and destruction.
>      * tests/cpp-unit-tests/browser_mock.cc
>      (mock_createobject): New, mocks NPAPI 'createobject'.
>      * tests/cpp-unit-tests/MemoryLeakDetector.h: New, memory leak detection
>      utility class.
>      * tests/cpp-unit-tests/main.cc
>      (ReportTestFinish): Print which tests resulted in memory leaks.
> 
> 
> Fix ChangeLog:
> 
> 2013-XX-XX  Adam Domurad  <adomurad at redhat.com>
> 
>      * plugin/icedteanp/IcedTeaScriptablePluginObject.cc
>  
> (IcedTeaScriptablePluginObject::get_scriptable_java_package_object): Fix
>      memory leak due to allocated NPClass.
>      (IcedTeaScriptableJavaPackageObject::get_scriptable_java_object):
>      Same.
> 
> 
> Cheers,
> -Adam
>