[Soot-list] Best class to extend new type of statement for parameter

[Steven Arzt]

Hi Utsav,

 

I’m always a bit skeptical about introducing new Unit types into Soot. The
problem is that analyses depend on covering an exhaustive set of types. If
you introduce a new one that is not a subclass of an existing leak class
(i.e. one that has no further subclasses at the moment), these analyses
might miss things. The next problem arises with the IO routines: Soot will
not know when and how to  produce your new unit type when reading in
class/java/dex/
 files. The same will happen when writing them back out in
the end.

 

Best regards,

  Steven

 

Von: soot-list-bounces at sable.mcgill.ca
[mailto:soot-list-bounces at sable.mcgill.ca] Im Auftrag von Utsav Chokshi
Gesendet: Montag, 24. März 2014 13:11
An: soot-list at sable.mcgill.ca
Betreff: [Soot-list] Best class to extend new type of statement for
parameter

 

Hello soot-community,
Which is best class (that implements unit interface) to extend, for
introducing new stmt/unit type in soot for fieldrefs.
Basically I want new class that encloses member variable (i.e. sootfield)and
related properties and implement unit interface.
I am currently thinking about abstractRefTypeInst ( I know inst is related
with grimple, not with jimple.) 
Or shall I simply extend AbstractUnit?

- Utsav Chokshi
[ Intern ,TCS Innovation Labs , Pune, India]

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.mcgill.ca/pipermail/soot-list/attachments/20140324/15f08229/attachment.html