[Soot-list] Why create temp$xx variables?

[Eric Bodden]

I am afraid, I don't quite understand. You disable constant folding so
that you can test your own constant folder. This leads to code
(obviously) that has many temporary variables because constants have
not been folded. So that should give an excellent test case for your
constant folder. I don't see what the problem is.

Eric

2009/4/21 Peng Du <imdupeng at gmail.com>:
> Thanks, Eric and Patrick.
>
> Yes, I did turned off all the optimizations to see what the result would be.
> The reason why I turned off those optimizations is: I want to write a custom
> constant propagator with Shimple. Then I ran into this temp$ issue and
> couldn't figure out why. If SOOT propagates and fold those constants, I
> can't test my algorithm then.
>
> How can I get around this?
>
> Thanks again.
>
>
> Peng
>
> On Tue, Apr 21, 2009 at 8:20 AM, Patrick Lam <plam at sable.mcgill.ca> wrote:
>>
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> You shouldn't get the temp$ variables if you generate Jimple in the
>> normal way. Are you explicitly turning off the default optimizations
>> (the copy propagation and constant folding that Eric mentioned)? They're
>> not optimizations per se, they just ensure that you get reasonable Jimple.
>>
>> pat
>> -----BEGIN PGP SIGNATURE-----
>> Version: GnuPG v1.4.9 (GNU/Linux)
>> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
>>
>> iEYEARECAAYFAkntyDkACgkQNTVxnNGEyOizbACdEZuAXJVSGtRezxZMi/SCW0F9
>> yF0AoKbkdjZTtLbhhCjBZrQDFG9NCmIf
>> =BJpO
>> -----END PGP SIGNATURE-----
>
>



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