On Sun, Jul 4, 2021 at 2:59 PM Volkan Yazıcı <volkan.yaz...@gmail.com>
wrote:

> *[My answers are inline.]*
>
> On Thu, Jul 1, 2021 at 11:48 PM Raman Gupta <rocketra...@gmail.com> wrote:
>
> >
> > https://issues.apache.org/jira/browse/LOG4J2-3116
> > https://github.com/apache/logging-log4j2/pull/543
> >
>
> Thanks! I will check these out momentarily.
>
>
I'm having one issue with the tests, and replied to your comment in the
pull.


>
> > One administrative note -- as I am a committer (although thus far I have
> > only committed to the log4j2 kotlin repo) should I do dev="rgupta"
> instead
> > of adding myself in due-to?
> >
>
> Of course! Apologies that I have totally missed it.
>
> Some technical notes -- some of the functionality could be more efficient
> > with some changes to the resolvers, to avoid delegating so many of the
> > fields to the pattern resolver. Specifically:
> >
> > 1) the level resolver could be enhanced with a label mapping function,
> >
>
> PatternLayout enhances individual resolvers. For instance, %level directive
> accepts a boolean lowerCase argument. In JsonTemplateLayout, I want to
> follow a different approach and rather than enhancing individual resolvers,
> I want to introduce reusable components. For instance, I have recently
> introduced the CaseConverter resolver. One can combine this with
> LevelResolver to have the same effect with %level{lowerCase} in
> PatternLayout. Coming back to your remark about LevelResolver, similarly,
> rather than enhancing LevelResolver, I want to introduce a
> SubstitutionResolver, if you know what I mean. Nevertheless, please go
> ahead and create a ticket addressing your "problem". I can comment further
> there on how to approach it.
>
>
I was wondering if I could combine resolvers, but I didn't see any
documentation or examples showing that. Thanks for confirming that is your
preferred approach to the problem.

Issue for SubstitutionResolver:
https://issues.apache.org/jira/browse/LOG4J2-3119


>
> > 2) the message resolver could optionally output stack trace as well, and
> >
>
> Again, similar to the aforementioned "reusable components" strategy, I was
> thinking of creating a ConcatenationResolver. Would you mind creating a
> ticket for this as well, please?
>
>
https://issues.apache.org/jira/browse/LOG4J2-3120


>
> > 3) a serial number resolver could be added.
> >
>
> Do you mean as the one mentioned in LOG4J2-3067
> <https://issues.apache.org/jira/browse/LOG4J2-3067>?
>
>
Yes


>
> > However, in my opinion we should commit the basic functionality first,
> and
> > deal with those sorts of optimizations later.
> >
>
> Agreed! This said, let's create tickets for these feature requests so as
> not to forget them.
>

Done, see above.

Reply via email to