Kyle Gray

I don't like the repetition of the `@` character, but agree that the `?` and `$` syntax feels unnecessary. We shouldn't allow named parameters to start with a number anyways.

@Emyrk I just update the proposal with a slightly different syntax, which allows for specifying type information without change the nullability of a parameter.

Tracking here https://github.com/kyleconroy/sqlc/issues/1661

The majority of the code is kept in the `internal` package so that it's easy for us to change the internals of sqlc without causing backwards incompatible changes. Plugins have...

There are a few ways to workaround this issue. Understand and interpreting `DO` functions is going to be pretty far off. Instead of having sqlc rely on your migrations directory,...

Fixed by https://github.com/sqlc-dev/sqlc/pull/3006

The fix has been merged, but going to keep this open until we do a release that contains the fix.

@shkvortez I can't reproduce this [behavior in the playground](https://play.sqlc.dev/p/1ef9f53de0cdb5cbc33784724a08f7d8016a811a44809731ac82dc564c81a644). Can you take a look at the example and help me out?

Tracking JSON unmarshaling of custom structs here: https://github.com/sqlc-dev/sqlc/issues/2761 and composite type support here #2760.

I've confirmed that downgrading to mingw 11.2.0 fixes the issue. https://github.com/sqlc-dev/sqlc/pull/2757