[Soot-list] a possible problem for LoopFinder

[Eric Bodden]

This was a tiny bug in MHGDomintorsFinder. It's fixed now in SVN
revision 3102. Thanks for reporting this!

Eric

On 28/01/2008, Peng Li <lipeng360 at gmail.com> wrote:
> HI
> I am using the soot build-in
> LoopFinder(soot.jimple.toolkits.annotation.logic.LoopFinder) to
> identify all the loops in my code. However, I found there is a problem
> for this LoopFinder.
>
> If a loop happens at the end of a method and there is no more code
> after the loop,  the LoopFinder is not able to find that loop. For
> example, in the following two methods , methodA and methodB, the
> LoopFinder is no problem to find the loop in the MethodB  because
> there are some code after the loop. However, the LoopFinder cannnot
> find the loop in MethodA
>
> MethodA ()
> {
> for (int i=0; i<10; i++)
> {
>     int k=0;
>     k++;
> }
> }
>
> MethodB ()
> {
> for (int i=0; i<10; i++)
> {
>     int k=0;
>     k++;
> }
> System.out.println("find loop");
> }
>
> We tried to trace the problem by looking the source of the LoopFinder,
> which used MHGDominatorsFinder to find all the loops, but it seems
> that there is no reason why the loop in MethodA cannot be found...
>
> Cheers
> Peng
> _______________________________________________
> Soot-list mailing list
> Soot-list at sable.mcgill.ca
> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
>


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