dcodeIO

The cookie parameter has remained from previous implementations, where multiple operations during visiting were possible (increment / decrement reference count). Hasn't been removed in anticipation of potential future cases where,...

Regarding the failing tests, the most straight-forward option would be to unroll the first loop iteration if incompatible local states are detected. In most cases, this will not be necessary,...

Keeping this PR open as part of [2020 vacuum](https://github.com/AssemblyScript/assemblyscript/issues/1277) since I'm planning to look into this further. In general it appears good to do as it simplifies some parts of...

When I thought about narrowing in the past, the mechanism I had in mind was to add a new `Flow#localTypes` in addition to `Flow#localFlags`, that would include narrowed types if...

Quite similar to why nullability checks are limited to locals. Also applies to narrowing if we don't want to run into unsoundness. ```ts ... let cv = c.v; if (cv...

Still thinking whether the nullable flag should be combined with this, so the following can become one ```ts thenFlow.inheritNonnullIfTrue(condExpr); thenFlow.inheritNarrowedTypeIfTrue(condExpr); ``` Right now this looks like unnecessary duplication of basically...

This is still missing the compiler utility to make it actually compile, mostly shortcutting. Leaving it open but converting to a draft as part of [2020 vacuum](https://github.com/AssemblyScript/assemblyscript/issues/1277), so we can...

So far, `Math` is quite intentionally identical to JS `Math`, also because the implementations can be switched with `--use`. Making one generic and the other not seems problematic in this...

Could be nvm-windows somehow interacting with npm, as [discussed here](https://github.com/npm/cli/issues/4234). On a glimpse, seems like the same issue to me.

Thanks Aaron! Suggesting to leave this PR open until we have an implementation to test positive against. Still good to have these tests pinned here so we know what we...