[Soot-list] Requiring main method for doing inter-procedural analysis Heros/Soot

Zhoulai zell08v at gmail.com
Sat Jan 11 16:14:04 EST 2014


Hi,

Have you read this http://www.bodden.de/2012/07/26/soot-custom-entry-points/
of Eric? It is not for Heros but should serve for your cases as well?

Otherwise, I think that you can write an auxiliary function that invokes
your customized entry point. In this way, to analyze this auxiliary
function (which is named 'main' in a class dedicated for it) should have
the similar effect as analyzing the original class with your entry point.

To extend this functionality into Soot/Heros would should be useful for
testing our analysis for libraries, but imho, I am not sure if it can serve
for real-life applications as we are most of time concerned about the
callgraph from 'main'.

Zhoulai
(web-site: http://zhoulaifu.com)

On Sat, Jan 11, 2014 at 8:40 PM, Guru Devanla <gurudev.devanla at gmail.com>wrote:

> Hello  Sooters,
>
> While doing inter-procedural analysis I see that the Heros/Soot api always
> needs an 'main' function. While I do understand this helps build the graph
> for the ensuing analysis, isn't there a way to start from different points
> if a 'main' method is not available.  I am wondering if this is somehow
> intractable and if any solutions exist.
>
> Any pointers to this would be great, since when I try to analyze very many
> libraries I end up with an error.
>
> Also, if there are any suggestion I would love to extend this
> functionality into Soot/Heros as it would immensely benefit me and hopeful
> a lot others right away.
>
> Thanks
> Guru
>
> _______________________________________________
> Soot-list mailing list
> Soot-list at sable.mcgill.ca
> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.mcgill.ca/pipermail/soot-list/attachments/20140111/eb3bed9f/attachment.html 


More information about the Soot-list mailing list