[
https://issues.apache.org/jira/browse/BCEL-362?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17578983#comment-17578983
]
Alexander Kriegisch edited comment on BCEL-362 at 8/12/22 1:58 PM:
-------------------------------------------------------------------
I did dig into the code (did you?), as much as I as a dilletante could. IMO,
this is simply a matter of an incomplete implementation, not a bug as such. My
reasoning was that a maintainer should know the code better than a mere user
and first-time issue creator by orders of magnitude.
We are not talking about keeping up with Java's twice-a-year release cycle
here. Java 11 and the advent of condy are a couple of years old. Java 11 was
LTS and has since been superseded by LTS version 17. I am not even complaining
that the implementation is incomplete, given the fact that condy is not a
frequently used feature. All I am asking is that today, in 2022, you deign to
look into the issue I raised. The time you spent discussing this with me might
on par with the time you would have needed to analyse the issue. But that is,
of course, just a guess. It feels like you are looking for reasons to reject
it. Even if I would take the time to try and understand this complex product
BCEL and be able to make it work for my special case, the fix would probably be
suboptimal at best, incomplete or outright wrong in the worst case.
Anyway, thanks for discussing the issue with me. Your effort is appreciated.
was (Author: kriegaex):
I did dig into the code (did you?), as much as I as a dilletante could. IMO,
this is simply a matter of an incomplete implementation, not a bug as such. My
reasoning was that a maintainer should know the code better than a mere user
and first-time issue creator by orders of magnitude.
We are not talking about keeping up with Java's twice-a-year release cycle
here. Java 11 and the advent of condy are a couple of years old. Java 11 was
LTS and has since been superseeded by LTS version 17. I am not even complaining
that the implementation is incomplete, given the fact that condy is not a
frequently used feature. All I am asking is that today, in 2022, you deign to
look into the issue I raised. The time you spent discussing this with me might
on par with the time you would have needed to analyse the issue. But that is,
of course, just a guess. It feels like you are looking for reasons to reject
it. Even if I would take the time to try and understand this complex product
BCEL and be able to make it work for my special case, the fix would probably be
suboptimal at best, incomplete or outright wrong in the worst case.
Anyway, thanks for discussing the issue with me. Your effort is appreciated.
> Handling for Constants.CONSTANT_Dynamic is missing in class ConstantPool
> ------------------------------------------------------------------------
>
> Key: BCEL-362
> URL: https://issues.apache.org/jira/browse/BCEL-362
> Project: Commons BCEL
> Issue Type: Bug
> Reporter: Alexander Kriegisch
> Priority: Major
>
> _Disclaimer: I am not a byte code buff, so if I am misunderstanding anything
> here, please forgive me for raising a stupid issue, try to be patient and
> explain it to me._
> When looking at the class
> {{[ConstantPool|https://github.com/apache/commons-bcel/blob/5165671d2bc4bed17918051f1689ee5fd4c3425a/src/main/java/org/apache/bcel/classfile/ConstantPool.java]}},
> it seems as if condy is not being handled in methods
> * {{constantToString(Constant)}},
> * {{addConstant(Constant, ConstantPool)}},
> * {{addConstantDynamic(int, int)}} (method does not exist, not sure if it
> would be necessary).
> I would have expected something like {{case Constants.CONSTANT_Dynamic}} with
> proper handling for this case in the first two methods.
> When using BCEL in order to parse class files using condy, like JaCoCo
> creates them, this causes errors. FYI, we are using BCEL code in AspectJ, and
> whenever people combine it with JaCoCo, corresponding problems occur.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
