mirror of
https://github.com/carbon-language/carbon-lang.git
synced 2026-10-03 20:15:48 +01:00
Dana Jansensandjosh11b
402dc2c064
- A test that should fail that looks to see we poison impls when we do a concrete lookup, so you can't define an impl specialization after we looked for it. This currently passes but should fail. - A test with a final specialization with a type constant written before a generic function using it. The generic function should be able to know the concrete type of the constant. This currently fails, and was discussed in open discussion here: https://docs.google.com/document/d/1Iut5f2TQBrtBNIduF4vJYOKfw7MbS8xH_J01_Q4e6Rk/edit?resourcekey=0-mc_vh5UzrzXfU4kO-3tOjA&tab=t.0#heading=h.swr8311y952x - A test with a specialization written after a generic function, which will be used symbolically so the type constant will not be known. This fails and should continue to, though the error diagnostic may change in time. - A test with a specialization written after a generic function, and which returns a value typed as the type constant from that specialization. The generic is called with types that should cause it to use that specialization in the specific, so the caller gets back the type expected. This currently fails but should pass. --------- Co-authored-by: josh11b <15258583+josh11b@users.noreply.github.com>