[Soot-list] Output of source line annotations in jimple JasminClass

[Hal Hildebrand]

No worries. This helps alot!   Thanks!

Sent from my tricorder

On Aug 30, 2010, at 8:54 AM, Eric Bodden
<bodden at st.informatik.tu-darmstadt.de> wrote:

> Hi Hal.
>
> Sorry about the late response. I have now patched the other
> JasminClass too, as follows:
>
> @@ -1236,6 +1237,8 @@
>     void emitStmt(Stmt stmt)
>     {
> +        LineNumberTag lnTag = (LineNumberTag) stmt.getTag("LineNumberTag");
> +        if(lnTag != null) emit(".line " + lnTag.getLineNumber());
>         stmt.apply(new AbstractStmtSwitch()
>         {
>             public void caseAssignStmt(AssignStmt s)
>
> I think this should work.
>
> Eric
>
> --
> Dr. Eric Bodden
> Software Technology Group, Technische Universität Darmstadt, Germany
> Tel: +49 6151 16-5478    Fax: +49 6151 16-5410
> Mailing Address: S2|02 A209, Hochschulstraße 10, 64289 Darmstadt
>
>
>
> On 16 August 2010 04:32, Hal Hildebrand <hal.hildebrand at gmail.com> wrote:
>> I'm finding that the recently discussed modification to the baf JasminClass to output source line attributes is not sufficient in all scenarios.  The problem comes about from these lines in the PackManager:
>>
>>
>>            case Options.output_format_class :
>>            case Options.output_format_jasmin :
>>                if (c.containsBafBody())
>>                    new soot.baf.JasminClass(c).print(writerOut);
>>                else
>>                    new soot.jimple.JasminClass(c).print(writerOut);
>>                break;
>>
>> As you can see, it's only if the class has a baf body will the baf.JasminClass patch actually work.  In my case, I'm doing transformations in the JTP phase which produce classes with jimple bodies for the methods.  Consequently, the classfile output takes the else branch and completely by passes the patch to output source line annotations.
>>
>> Thus, my question is how to go about patching jimple.JasminClass in the same fashion such that source line annotations are emitted correctly...
>>
>> Thanks,
>>
>> Hal
>>
>> _______________________________________________
>> Soot-list mailing list
>> Soot-list at sable.mcgill.ca
>> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
>>