RE: Re: Re: LocalVariableSorter

A piece of code and the sample input .class might be useful in
communicating the issue (and solution here)

> -----Original Message-----
> From: Nikolas Nehmer [mailto:[email protected]]
> Sent: Wednesday, September 15, 2010 10:58 AM
> To: Eugene Kuleshov
> Cc: [email protected]
> Subject: [asm] Re: Re: LocalVariableSorter
> 
> Hi Eugene,
> 
> in fact, I don't need the variable to be in the localVar table. But as
> mentioned in my first mail, just using the newLocal method provided by
> the localVarSorter doesn't work. Obviously the Sorter is doing
> "something" (i.e. it seems to be invoked in the right way )- e.g.
> shifting the indexPositions of exisitng variables in the table -
> leaving
> "space" for the created variable. But if I try to access the new
> variable with a store operation, the store operation's index in the
> bytecode changes automatically to a value that is defined in the
> variable table (in contrast to the new variable that is not defined in
> the table).
> ISTORE 1 automatically changes to ISTORE 0 e.g. when looking at
> bytecode
> after the instrumentation process using a TraceClassVisitor. Of course
> this usually ends in a verify error because of typing problems. So
> basically my problem is not to insert new entries to the table, but to
> prevent the bytecode to change by some magic. It seems as if the index
> in the store operation is changing because "something" decided that at
> this point in time no variable could be accessed at the given index
> position. Any recommendations to solve that problem?
> 
> Eugene Kuleshov schrieb:
> > Nikolas,
> >
> >   If you need your new variable in the local var table debug info,
> > then you should call visitLocalVariable() method
> >
> >   regards,
> >   Eugene
> >
> >
> > On Wed, Sep 15, 2010 at 12:26 PM, Nikolas Nehmer
> > <[email protected]> wrote:
> >
> >> Hi,
> >>
> >> I have a question regarding the LocalVariableSorter. I`m trying to
> add an
> >> additional local variable calling the newLocal method. Looking at
> the
> >> localVariable entries in the resulting bytecode, one can see that
> the
> >> variable indexes are "shifted" according to the added variable - but
> there
> >> is no additional entry added to the localVariable table. If I'm
> trying to
> >> access the added variable at the corresponding index position, the
> index
> >> within the store (let's asume it's a store statement that I'm using
> to
> >> access the newly defined variable) statement is shifted (by some
> magic) to a
> >> close value that is present (index 1 --> index 2) in the local
> variable
> >> table. Any ideas?
> >>
> >> Best regards
> >> Nikolas
> >>
> >>
> >> --
> >> You receive this message as a subscriber of the [email protected] mailing
> list.
> >> To unsubscribe: mailto:[email protected]
> >> For general help: mailto:[email protected]?subject=help
> >> OW2 mailing lists service home page: http://www.ow2.org/wws
> >>
> >>
> >>
> 
> 
> --
> Dipl.-Inf. Nikolas Nehmer,
> Dept. of Computer Science. University of Kaiserslautern
> P.O. Box 3049, 67653 Kaiserslautern, Germany.
> room 36/308, phone: +49 (0) 631 - 205 2644, fax: +49 (0) 631 - 205 3299
> mail: [email protected]

-- 
You receive this message as a subscriber of the [email protected] mailing list.
To unsubscribe: mailto:[email protected]
For general help: mailto:[email protected]?subject=help
OW2 mailing lists service home page: http://www.ow2.org/wws