Remove duplicate -Xlint:options flags. #14788
Conversation
github-project-automation
bot
moved this to Open
in OpenSearch Lucene & Core Performance Tracking
|
This PR does not have an entry in lucene/CHANGES.txt. Consider adding one. If the PR doesn't need a changelog entry, then add the skip-changelog label to it and you will stop receiving this reminder on future updates to the PR. |
| // TODO: this depends on the order of argument configuration... | ||
| int releaseIndex = options.compilerArgs.indexOf("--release") | ||
| options.compilerArgs.removeAt(releaseIndex) | ||
| options.compilerArgs.removeAt(releaseIndex) | ||
|
|
||
| // Remove conflicting options for the linter. #14782 | ||
| options.compilerArgs.removeAll("-Xlint:options") |
There was a problem hiding this comment.
@breskeby is there any nicer way to do this kind of compilerArgs manipulation that doesn't
depend on the ordering of those configuration callbacks? I can think of a custom argument provider but this would be even more confusing than what it is now, I think.
There was a problem hiding this comment.
I don't have a better idea than you. I can think of a smarter ArgumentProvider but no easy fix
There was a problem hiding this comment.
If this works it is ok as a first step. I had same idea.
Indeed the problem with configuration order is hard to handle. Currently we rely on order of configuration.
Maybe @breskeby has an idea.
I will comment on JDK issue that there is kind of a regression because the order how the options are applied changed. Maybe they should have a defined order.
I was not able to test it yesterday in train, I just opened issue. My bad.
Greetings from bbuzz2025.
|
Can you backport this change? |
|
I will apply and backport, no worries, Uwe. Enjoy a beer in Berlin! |
github-project-automation
bot
moved this from Open
to Merged
in OpenSearch Lucene & Core Performance Tracking
|
If I remember: the order of configuration is still predictable by the order of including them into the main build script. I always felt bad with that, but it worked. Just want to confirm it also applies to changes in main. |
|
Yes, I think this can be relied upon. I don't think it's mentioned anywhere, officially, but I think it's the case. |
Fixes #14782 by not emitting duplicate lint flags to javac.