Re: [abc-users] adding jimple

You just have to add a fresh Local to the method's Body. Then you
assign the result to the Local and pass the local as an arument to the
EqExpr.

Eric

2009/4/20 Amjad Nusayr <anusayr_at_cs.nmsu.edu>:
> Thanks Eric,
> Another quick question. The method I'm calling returns an integer, How can I
> store the return value from my invocation? I want to use it in a "EqExpr"
> that compares it to an int.
>
> Amjad
>
> Eric Bodden wrote:
>>
>> Hi Amjad.
>>
>> You have to make Soot aware of your helper class, using your AbcExtension.
>> See:
>>
>> AbcExtension.addBasicClassesToSoot()
>>
>> Eric
>>
>> 2009/4/20 Amjad Nusayr <anusayr_at_cs.nmsu.edu>:
>>
>>>
>>> Hi,
>>> I trying to add jimple stmts to the IR code generated in abc using
>>> MyResidue.codeGen method. The jimple code I writing should call a method
>>> from a helper class I created.
>>> I used the code from this link to aid me.
>>> http://www.sable.mcgill.ca/soot/tutorial/profiler2/index.html
>>>
>>> abc compiles but when run abc with my extension using the following
>>> command
>>> line
>>> +soot -f J -soot -argfile files.lst
>>>
>>> I get
>>> Exception in thread "main" polyglot.util.InternalCompilerError: unhandled
>>> exception during we
>>> ving/optimisation
>>>      at abc.main.CompileSequence.runSequence(CompileSequence.java:142)
>>>      at abc.main.Main.run(Main.java:406)
>>>      at abc.main.Main.main(Main.java:144)
>>> Caused by: polyglot.util.InternalCompilerError: World.aj:29,3-32:4: Error
>>> during matching
>>>      at
>>>
>>> abc.weaving.matching.AdviceApplication.doShadows(AdviceApplication.java:209)
>>>      at abc.main.AbcExtension.findMethodShadows(AbcExtension.java:541)
>>>      at abc.bbs.AbcExtension.findMethodShadows(AbcExtension.java:230)
>>>      at
>>>
>>> abc.weaving.matching.AdviceApplication.doMethod(AdviceApplication.java:275)
>>>      at
>>>
>>> abc.weaving.matching.AdviceApplication.computeAdviceLists(AdviceApplication.java:3
>>> 2)
>>>      at
>>>
>>> abc.weaving.aspectinfo.GlobalAspectInfo.computeAdviceLists(GlobalAspectInfo.java:4
>>> 2)
>>>      at abc.main.CompileSequence.weave(CompileSequence.java:447)
>>>      at abc.main.CompileSequence.runSequence(CompileSequence.java:115)
>>>      ... 2 more
>>> Caused by: java.lang.ExceptionInInitializerError
>>>      at
>>> abc.bbs.weaving.aspectinfo.ProbabilityD.matchesAt(ProbabilityD.java:52)
>>>      at
>>> abc.weaving.aspectinfo.ShadowPointcut.matchesAt(ShadowPointcut.java:38)
>>>      at abc.weaving.aspectinfo.AndPointcut.matchesAt(AndPointcut.java:65)
>>>      at abc.weaving.aspectinfo.AndPointcut.matchesAt(AndPointcut.java:65)
>>>      at abc.weaving.aspectinfo.AndPointcut.matchesAt(AndPointcut.java:65)
>>>      at
>>> abc.weaving.matching.AdviceApplication$2.run(AdviceApplication.java:115)
>>>      at
>>>
>>> abc.weaving.matching.AdviceApplication.doShadows(AdviceApplication.java:192)
>>>      ... 9 more
>>> Caused by: java.lang.RuntimeException: This operation requires resolving
>>> level SIGNATURES but
>>> Helper is at resolving level HIERARCHY
>>> If you are extending Soot, try to add the following call before calling
>>> soot.Main.main(..):
>>> Scene.v().addBasicClass(Helper,SIGNATURES);
>>> Otherwise, try whole-program mode (-w).
>>>      at soot.SootClass.checkLevel(SootClass.java:121)
>>>      at soot.SootClass.addMethod(SootClass.java:570)
>>>      at soot.coffi.Util.resolveFromClassFile(Util.java:249)
>>>      at soot.CoffiClassSource.resolve(CoffiClassSource.java:39)
>>>      at soot.SootResolver.bringToHierarchy(SootResolver.java:194)
>>>      at soot.SootResolver.bringToSignatures(SootResolver.java:219)
>>>      at soot.SootResolver.processResolveWorklist(SootResolver.java:139)
>>>      at soot.SootResolver.resolveClass(SootResolver.java:124)
>>>      at soot.Scene.loadClass(Scene.java:390)
>>>      at soot.Scene.loadClassAndSupport(Scene.java:375)
>>>      at
>>>
>>> abc.bbs.weaving.residues.ProbabilityDResidue.<clinit>(ProbabilityDResidue.java:37)
>>>      ... 16 more
>>>
>>>
>>> Could someone tell me what this means and how to fix it.
>>>
>>> thanks
>>> Amjad
>>>
>>>
>>>
>>>
>>
>>
>>
>>
>
>
>

-- 
Eric Bodden
Sable Research Group, McGill University
Montréal, Québec, Canada