Type annotations cleanup bugs

[Alex Buckley]

// Switching to type-annotations-dev so everyone involved can follow.

Eric,

Did you see my two mails last week to type-annotations-dev about javac 
bugs on receiver parameters and class literals?

JDK-8027264 should be closed, there is no bug there.

Alex

On 10/25/2013 7:45 AM, Eric McCorkle wrote:
> I closed two of them.  Generally, with these issues coming from my combo
> test, it's entirely possible I could be wrong.  It's clear I need to
> retool the test a bit.
>
> As for the new/constructor invocation one (JDK-8027264), I'd like to
> discuss that one more, but after the groundwork is in place to actually
> support a source type annotation producing multiple bytecode type
> annotations.
>
> On 10/24/13 15:02, Werner Dietl wrote:
>> Eric,
>>
>> thanks for collecting these bug reports.
>>
>> Three of them seem invalid to me and I've added comments to them
>> explaining why.
>>
>> The one about unifying code would be nice to have and I added a comment
>> what the complication is.
>>
>> cu, WMD.
>>
>> On 10/24/2013 01:42 PM, Michel Trudeau wrote:
>>> Nice work Eric.
>>>
>>> Anyone can get started on this list today,  for each sub-task
>>>
>>> - Work on fix
>>> - Get it review
>>> - Once ready to push
>>>    - we will move the task to fixversion 8
>>>    - get the approval from SQE to push it
>>>    - push the fix
>>>
>>> Thanks,
>>> Michel
>>>
>>> On Oct 24, 2013, at 10:38 AM, Eric McCorkle <eric.mccorkle at oracle.com> wrote:
>>>
>>> I've filed bugs for all the type annotations issues I know about, and
>>> also for my suggested fixes to address the remaining issues.
>>>
>>> They are all under a task here:
>>> https://bugs.openjdk.java.net/browse/JDK-8027181
>>>
>>> There are probably more specific bugs; as I said, the combo test is
>>> effectively an infinite source of bugs for all timelines less than two
>>> weeks.
>>>
>>> If any of these need to be marked tbd_minor, or critical watchlist, or
>>> whatever else, I'm fine with that.  I plan on fixing as many as I can
>>> regardless.
>>>
>>> I plan to pull together an outline for my suggested roadmap some time in
>>> the next week.  That should clarify things.
>>> <eric_mccorkle.vcf>
>>>