[Soot-list] Soot and Serialization

sun ding xyz031702 at hotmail.com
Fri Jul 27 23:16:37 EDT 2012


Dear Darko and Eric,

Thanks a lot for providing the resource and suggestions to me! Let me go through it and feedback to u.

Best Regards,
Ding Sun

在 27 Jul, 2012,4:32 AM,"Darko Marinov" <marinov at illinois.edu> 写道:

>> There was a paper [1] by Darko Marinov
> The work on CoDeSe was led by Milos Gligoric not me. :)
> 
> The CoDeSe system itself does not require the classes to have the Serializable marker interface.  But to compare CoDeSe with the standard Java serialization, Milos had a piece of code that adds "implements Serializable" to every class in a .jar file.  You can find more at [1].
> 
> Ding Sun, please feel free to email Milos and/or me if you need more info.
> 
> Darko
> 
> 
> On Thu, 26 Jul 2012, Eric Bodden wrote:
> 
>> Hi Ding Sun.
>> 
>> No there is nothing that I know of. There was a paper [1] by Darko
>> Marinov (in CC) on fast deserialization. I forgot whether they require
>> those annotations. Darko, could you comment on this?
>> 
>> Best wishes,
>> Eric
>> 
>> [1] http://mir.cs.illinois.edu/codese/
>> 
>> On 26 July 2012 10:25, dingsun <xyz031702 at hotmail.com> wrote:
>>> Dear all,
>>> 
>>> During the static analysis, sometimes I want to serialized some Soot objects
>>> into my database. So later I could recover the context and these runtime
>>> objects.
>>> 
>>> I found not every soot class could be serialized.  Some classes have not
>>> implemented the "Serializable" interface. For example,
>>> soot.util.SingletonList.   So currently I failed to achive my goal.
>>> 
>>> I'm wondering is there a fast method to enable all the classes in Soot-2.5
>>> serializable? Thanks very much.
>>> 
>>> Best Regards,
>>> Ding Sun
>>> 
>>> _______________________________________________
>>> Soot-list mailing list
>>> Soot-list at sable.mcgill.ca
>>> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list


More information about the Soot-list mailing list