[Javapathfinder-svn] SF.net SVN: javapathfinder:[1479] branches/mango/MangoJPF

[<fra...@us...>]

Revision: 1479
          http://javapathfinder.svn.sourceforge.net/javapathfinder/?rev=1479&view=rev
Author:   frankrimlinger
Date:     2009-05-02 20:22:52 +0000 (Sat, 02 May 2009)

Log Message:
-----------
Working through details of jpf/scanner invocation phase transition.

Notes: 1) invokevirtual and invokeinterface are almost the same, except invokeinterface does not require access control. 2) mango currently simulates a thrown exception by "returning" an exception object to caller, and then "catching" return values and branching to the handlers or rethrowing as appropriate. The question then becomes, since jpf is now in the driver's seat as far as control flow goes, can we scrap the old "exception catchers" and just model the observed jpf behavior.  This really is the way to go, because if the map from jpf simulation to the mango control flow graph is not invertable, there will be problems down the road.

Modified Paths:
--------------
    branches/mango/MangoJPF/.settings/org.eclipse.jdt.core.prefs
    branches/mango/MangoJPF/javapathfinder-mango-bridge/mango/scanner/InvocationChoiceGenerator.java
    branches/mango/MangoJPF/javapathfinder-mango-bridge/mango/scanner/ScanChoiceGenerator.java
    branches/mango/MangoJPF/javapathfinder-mango-bridge/mango/scanner/bytecode/INVOKEINTERFACE.java
    branches/mango/MangoJPF/javapathfinder-mango-bridge/mango/scanner/bytecode/INVOKESTATIC.java
    branches/mango/MangoJPF/javapathfinder-mango-bridge/mango/scanner/bytecode/INVOKEVIRTUAL.java


This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.