RFR: 8254692: (se) Clarify the behaviour of the non-abstract SelectorProvider::inheritedChannel [v3]

[Fernando Guallini]

On Tue, 20 Oct 2020 15:51:09 GMT, Alan Bateman <alanb at openjdk.org> wrote:

>> No disagreement about testCustomProvider not being the most appropriate place to test the default implementation of
>> inheritedChannel. What I was suggesting was to move *all* the assertions related to the default implementation of
>> SelectorProvider to a single new test located under test/jdk/java/nio/channels/spi/SelectorProvider/, rather than
>> splitting the tests for the default implementation across several different tests and directory hierarchies.
>
> A new test/jdk/java/nio/channels/spi/SelectorProvider that covers all the assertions would be great, thanks!

Thank you Alan and Chris for reviewing, I moved all the assertions related to the default implementation from
ProtocolFamilies to a new test class as suggested and removed unused imports in ProtocolFamilies

-------------

PR: https://git.openjdk.java.net/jdk/pull/745