Future of 'type-annotations/type-annotations' forest

[Werner Dietl]

Eric,

>> Type annotations in the jdk9 repo are currently broken for serious
>> use. On Feb 7 I identified the invalid re-ordering performed in
>> ClassReader:
>>
>> http://mail.openjdk.java.net/pipermail/type-annotations-dev/2014-February/001592.html
>>
>> This problem still exists in jdk9 - the problematic changeset wasn't
>> rolled back, nor was the fix applied. I have fixed this problem in
>> type-annotations.
>> I was waiting to hear an update on this before pursuing the other
>> outstanding changes.
>
> I am aware of the fix that was pushed into the type annotations forest.
>  As we discussed when you developed the fix, I cannot push it into
> jdk9-dev as-is, because the development practices of the langtools team
> require that any changeset that introduces behavioral changes be
> accompanied by a test.
>
> I will take care of the test this time, so that you can run checkers on
> the contents of jdk9.  However, in the future, we ask that patches
> submitted against jdk9 conform to the langtools group's development
> practices.

It was your "cleanup" changeset that broke this.
I think the correct thing to do would have been to either roll back
your broken changeset or apply my fix.


>> You say that after the new changes type-annotations will be seriously
>> out of date. I don't see why this would be the case. I'm currently
>> keeping type-annotations in sync with jdk9, making sure that the
>> Checker Framework correctly works with type-annotations. Once these
>> further refactorings go into jdk9, they should be easy to pull into
>> type-annotations.
>
> I am posting a series of three (maybe just two) patches against jdk9,
> which rearchitect a significant portion of the frontend implementation.
>  The end result is that it may not be trivial to rebase all the changes
> you list below.

I don't see why this should be hard - I'm just pulling in the
changesets that you will push to jdk9.


>> - test/tools/javac/annotations/typeAnnotations/failures/AnnotatedImport.java
>> is being ignored in jdk9, even though it is working. Same for
>> AnnotatedPackage1 and CantAnnotateStaticClass3.
>
> The positions patch will probably fix a few of the @ignore'd tests.  So
> I'd planned to go through, figure out what's working, and turn them back on.
>
> (The rest should be unaffected by my work)

These tests are already working - so the @ignore in them might hide if
one of your changes breaks them.


>> How should I go about getting these changes into jdk9/jdk and jdk9/langtools?
>> Is it enough to send the above descriptions to compiler-dev and point
>> interested reviewers to a diff between the two repos? Should I attach
>> a diff?
>
> I have three major patches planned: remove AnnotatedType, which is
> already out for public review, the "positions patch", which is about to
> be, and a patch to remove code obsoleted by the positions patch.
>
> Here's what I'll do: we want to get the first two in ASAP, but the third
> is mostly for code cleanliness, and it may cause trouble rebasing some
> of those changes.  I'll hold off on the cleanup patch, and we'll look at
> bringing the changes over after.
>
> As for the reorder fix, I will commit that to jdk9 (with a test) before
> I commit the two major changes, so that you can run checkers with them.

This sounds like I shouldn't do anything with the changes in
type-annotations/langtools and you will take care that all meaningful
changes are transfered to jdk9.
Right?

Should I do something about the two small fixes in type-annotations/jdk?

cu, WMD.

-- 
http://www.google.com/profiles/wdietl