colexechash: introduce a type alias for keyID

[yuzefovich]

Release note: None

[cockroach-teamcity]

This change is 

[michae2]

Thanks for doing this!

Reviewable status: complete! 0 of 0 LGTMs obtained (waiting on @michae2 and @yuzefovich)

---

pkg/sql/colexec/colexechash/hashtable.go, line 73 at r1 (raw file):

//
// keyID of 0 is reserved to indicate the end of the hash chain.
type keyID uint64

Ah, crap, I misled you. I was wrong, this should probably be a "type alias declaration" not a "type definition" (TIL they are different in go!). In other words, this should probably be:

type keyID = uint64

And by adding that equals sign, we should be able to use []uint64 as []keyID and vice-versa (even without explicit casts). Then you won't need the unsafe.Pointer stuff and in general that will make life much easier. Sorry about that!

Code quote:

type keyID uint64

[yuzefovich]

Reviewable status: complete! 0 of 0 LGTMs obtained (waiting on @michae2)

---

pkg/sql/colexec/colexechash/hashtable.go, line 73 at r1 (raw file):

Previously, michae2 (Michael Erickson) wrote…

Ah, crap, I misled you. I was wrong, this should probably be a "type alias declaration" not a "type definition" (TIL they are different in go!). In other words, this should probably be:

type keyID = uint64

And by adding that equals sign, we should be able to use []uint64 as []keyID and vice-versa (even without explicit casts). Then you won't need the unsafe.Pointer stuff and in general that will make life much easier. Sorry about that!

Oops, this makes sense, I don't know why I didn't think of this myself right away, I guess I wasn't consciously aware of these differences too. PTAL.

[michae2]

Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on @michae2)

[yuzefovich]

TFTR!

bors r+

[craig]

Build succeeded:

• GitHub CI (Cockroach)