I think that Javassist is instrumenting field access incorrectly inside static
methods (when the static method accesses an instance variable of an object
passed as a parameter). That's why it affects these $access methods so much
... that's pretty much what those do.
View the original post :
The class that was affected has changed a bit since my first encounter with the
problem. No, even though I've restored the inner class that spurred the problem
in the first place, I'm no longer seeing the VerifyError. I guess I'll post
again if and when it recurs.
View the original post :
ht
Yes, I think access$0 is generated by the normal compiler.
I'm going to try and reproduce this, and I'll upload the normal and enhanced
.class files.
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4015811#4015811
Reply to the post :
http://www.jboss.com/ind
Hi,
The bytecode of access$0 in GridDataSopurce seems broken.
If access$0 is generated by Javassist, can you find the source
code given to Javassist for generating this method?
I don't think access$0 is not a method implicitly generated
by Javassist.
Thanks
View the original post :
http://www
And now I can see that the forum has truncated my (very long) message. Here's
the most relevant part at the end:
| public void initializer();
| Code:
|Stack=1, Locals=1, Args_size=1
|0: aload_0
|1: pop
|2: nop
|3: nop
|4: return
| LineNumb
Ah, never mind that note at the bottom, I found the compiler-generated code for
access$0.
| static org.apache.tapestry.grid.GridDataSource
access$0(org.apache.tapestry.corelib.components.Grid);
| Code:
|Stack=1, Locals=3, Args_size=1
|0: aload_0
|1: astore_1
|2
