On Tue, 22 Nov 2022 18:39:43 GMT, John Hendrikx <jhendr...@openjdk.org> wrote:
> - Remove unsupported/unnecessary SuppressWarning annotations
> - Remove reduntant type specifications (use diamond operator)
> - Remove unused or duplicate imports
> - Remove unnecessary casts (type is already correct type or can be autoboxed)
> - Remove unnecessary semi-colons (at end of class definitions, or just
> repeated ones)
> - Remove redundant super interfaces (interface that is already inherited)
> - Remove unused type parameters
> - Remove declared checked exceptions that are never thrown
> - Add missing `@Override` annotations
It's clear that the unnecessary cast removal is causing many discussions, so I
think its best to back out that change. The danger in many unnecessary casts is
that they can become **real** casts after a seemingly innocuous change, so it
is mostly a benefit to make refactors easier to reason about. Unnecessary
casts turning into down casts are the ones that will cause trouble:
float x, y, z;
float a = (float)x * (float)y * (float)z; // unnecessary casts
However, after a refactor:
double x, y, z;
float a = (float)x * (float)y * (float)z; // down casts (no warning!)
The extra precision that one may have hoped to gain is still lost.
Similarly:
Integer number = ... ;
Integer x = (Integer) number; // unnecessary cast
Refactor:
Number number = ... ;
Integer x = (Integer) number; // down cast (no warning!)
The code may now raise a `ClassCastException` here if `Number` is not an
`Integer`. Without the unnecessary cast this would be a compile error to be
investigated.
-------------
PR: https://git.openjdk.org/jfx/pull/960
