Closed waywardmonkeys closed 3 months ago
I'm unable to test this locally, so some care should be exercised.
In #77, updating to Rust 1.79 enabled a new clippy lint about how transmute
calls are annotated and instead of doing what it suggested, I thought it would be better to get rid of the transmute entirely instead.
One option was to just call font.coverage.leaves.push(leaf.map)
instead of transmuting, since the map
has the right type and is pub
, so no transmute
would be required.
That felt a bit odd though to do and I thought maybe this approach would be better since it would use the code that already exists within fontconfig-cache-parser
.
Is there an issue with this not inlining that code? (If so, could flag that method in a future release of fontconfig-cache-parser
as #[inline]
perhaps?)
Or, going by my mental track record of late after illness, I could just be entirely wrong here.
(Another option might have been to use even more code from fontconfig-cache-parser
, but I don't know enough at that point about this and the requirements to be sure...)
Instead of transmuting the
CharSetLeaf
to the underlying value and then doing the operations on that, use the operation as already exposed viaCharSetLeaf::contains_byte
.