RFR: 8339296: Record deconstruction pattern in switch fails to compile

[Maurizio Cimadamore]

On Fri, 27 Sep 2024 12:21:06 GMT, Jan Lahoda <jlahoda at openjdk.org> wrote:

>> src/jdk.compiler/share/classes/com/sun/tools/javac/code/Type.java line 2424:
>> 
>>> 2422:         }
>>> 2423: 
>>> 2424:         @Override @DefinedBy(Api.LANGUAGE_MODEL)
>> 
>> Modelling-wise, I'm not sure. It almost seems as if `UnknownType` should be subclassed by `ErrorType` and not the other way around. Is there a specific reason you went for this?
>
> I would see the `UnknownType` as a specific kind of an erroneous type (and hence `UnknownType` being a subtype or erroneous, as it is here).
> 
> But, even from a practical point of view - `ErrorType` basically must extend `ClassType`, I think, so implanting `UnknownType` in there seems difficult.

I guess I'm a bit surprised by the typeKind - which is OTHER (you added the override to keep compatibility with what was there before). In that sense, it doesn't seem like `UnknownType` "is a" `ErrorType`, because its tag seems "less general" than the one in `ErrorType`.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/20990#discussion_r1778644189