codeconsole opened a new pull request, #3113:
URL: https://github.com/apache/groovy/pull/3113
Static type checking rejects a value whose static type is `Closure<R>` (not
a closure literal) when it is assigned to a parameterized functional interface:
```groovy
import java.util.function.Supplier
@groovy.transform.CompileStatic
def m() {
Closure<String> c = { 'a' }
Supplier<String> s = c // Incompatible generic argument types. Cannot
assign groovy.lang.Closure<java.lang.String> to:
java.util.function.Supplier<java.lang.String>
s.get()
}
```
Raw `Supplier s = c` and closure literals are accepted, and dynamic Groovy
coerces the closure. The same error came from properties, subscripts,
ternaries, field assignments and generic methods whose result is a
`Closure<R>`, and for SAM types with parameters (`Function<Integer,String>`,
`Comparator<String>`, ...).
`checkCompatibleAssignmentTypes` already accepts Closure to SAM coercion;
the generics check that runs after it (`checkTypeGenerics`) compared
`Closure<String>` with `Supplier<String>` as a plain parameterized subtype and
failed. It now accepts a `Closure<R>` (or a Closure subclass) when the SAM
type's abstract method, parameterized by the target type, returns void or a
type that accepts `R`, using the same test as the return check for closure
literals. A mismatch such as `Closure<Integer>` to `Supplier<String>` is still
reported. Static compilation emits the same coercion as for the raw case, so
the value is converted to the interface.
## Tests
`CoercionSTCTest.testCoerceToFunctionalInterface22` and
`testCoerceToFunctionalInterface23` fail without the change, both in
`CoercionSTCTest` and in `CoercionStaticCompileTests`;
`testCoerceToFunctionalInterface24` covers the assignments that must stay
errors. `groovy.transform.stc.*`, `org.codehaus.groovy.classgen.asm.sc.*`,
`groovy.bugs.*` and `gls.*` pass. The Checkstyle and CodeNarc gates pass.
Seen in 4.0.33, 5.0.8 and 6.0.0.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
