[abc] RE: NPE in coffie / long weaving time

Hi again.

So, I was able to track this down to this position:

It seems that some of the array entries in the line number attachment
are "null". Does anyone have an idea if this is a bug or if one should
handle cases of having null in there in this method?

Anyway - so I was able to circumvent the NPE by "-keep-line-numbers no".
But now it seems that abc really takes ages to weave aprove from jar to
jar. I have had it running on my lab machine for several hours now with
about 90% time and now also memory consumption. I doubt that it is ever
going to terminate... What's the longest weave time you have seen so
far?

/Eric

> -----Original Message-----
> From: Eric Bodden
> Sent: Wednesday, January 18, 2006 11:06 AM
> To: abc@comlab.ox.ac.uk
> Subject: NPE in coffie
>
> Hi. As I wrote earlier I am trying to get Aprove to weave
> with abc. By now I managed to get hold of all the necessary
> jars (interestingly, ajc would require less classes to be on
> the classpath for weaving - but maybe that's do to the fact
> that they allow separate compilation and weaving), but now I
> am stuck because of the following nullpointer exception:
>
> Exception in thread "main" polyglot.util.InternalCompilerError:
> Exception while processing <aprove.Framework.Logic.SetOfFilterMask:
> aprove.Framework.Logic.SetOfFilterMask
> andOrOr(aprove.Framework.Logic.SetOfFilterMask,aprove.Framewor
> k.Logic.SetOfFilterMask,boolean)>
> at abc.main.Main.weave(Main.java:901)
> at abc.main.Main.run(Main.java:519)
> at abc.main.Main.main(Main.java:206) Caused by:
> java.lang.NullPointerException
> at java.util.Hashtable.get(Hashtable.java:336)
> at soot.coffi.CFG.adjustLineNumberTable(CFG.java:991)
> at soot.coffi.CFG.eliminateJsrRets(CFG.java:462)
> at soot.coffi.CFG.<init>(CFG.java:89)
> at
> soot.coffi.CoffiMethodSource.getBody(CoffiMethodSource.java:87)
> at soot.SootMethod.getBodyFromMethodSource(SootMethod.java:80)
> at soot.SootMethod.retrieveActiveBody(SootMethod.java:304)
> at abc.main.Main.weave(Main.java:897)
> ... 2 more
>
> I think it should be caused by Coffie. Can anybody help me
> with this one?
>
> Cheers,
> Eric
>
>
Received on Thu Jan 19 04:57:48 2006