RFR: JDK-8306819: Consider disabling the compiler's default active annotation processing
Ethan McCue
ethan at mccue.dev
Tue Oct 31 15:30:01 UTC 2023
I'm still unclear why this change is not going through the same general
process as JEP 451.
On Fri, Oct 20, 2023 at 9:45 PM Joe Darcy <darcy at openjdk.org> wrote:
> On Sun, 10 Sep 2023 22:57:30 GMT, Rob Bygrave <duke at openjdk.org> wrote:
>
> > > This patch isn't aimed to remove AP. This aims to avoid unintentional
> AP pollutions in classpath
> >
> > Yes, I understand that.
> >
> > > Such classpath separation have already been promoted by build tools
> like Gradle.
> >
> > **_BUT_** it has **_NOT_** been promoted by **_Maven_**. In the bug
> report and in this PR I do not see any explicit recognition of how this
> change will impact the community that is using Maven and I am suggesting
> that this is a significant proportion of the community.
> >
> > For example, all the libraries that use annotation processing that
> document maven use could be impacted along the lines that their
> documentation will now be at least misleading (and sometimes incorrect).
> >
> > This change might indeed be a good long term change but it seems to be
> being made without any recognition or understanding of how impactful it
> will be.
>
> Maven is listed as a participant in the OpenJDK Quality Outreach effort (
> https://wiki.openjdk.org/display/quality/Quality+Outreach).
>
> The prior change of the javac emitting the note was explicitly raised on
> the quality discuss alias in July 2023 (
> https://mail.openjdk.org/pipermail/quality-discuss/2023-July/thread.html),
> ahead of the GA of JDK 21 in September.
>
> The change in this PR is likewise publicized this the October 2023 quality
> discuss message (
> https://mail.openjdk.org/pipermail/quality-discuss/2023-October/001129.html),
> months ahead of the expected GA of JDK 23 in March 2024.
>
> -------------
>
> PR Comment: https://git.openjdk.org/jdk/pull/14432#issuecomment-1773592750
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/compiler-dev/attachments/20231031/7714b571/attachment.htm>
More information about the compiler-dev
mailing list