Mazdak Farrokhzad

Represent `RowIter` as `NonZeroU32`

1

See https://github.com/clockworklabs/SpacetimeDB/pull/1622#discussion_r1733123333.

That is, add a test that exercises this counter-example: https://github.com/clockworklabs/SpacetimeDB/pull/1622/files#r1727653363

C#: *BTreeIndexBounds => *IndexScanRangeBounds

3

# Description of Changes In C# bindings, rename `*BTreeIndexBounds` to `*IndexScanRangeBounds` to be consistent with the Rust bindings. # API and ABI breaking changes These were marked as `Internal`, so...

Currently, when indices are created in the tx state, we end up putting committed state rows in the tx index. This forces us to do a bunch of unnecessary checks...

Currently, when the WASM module crashes, we must first do an expensive `__describe_module__` call, but we could instead store the extracted `RawModuleDef` in the host instead, so we only ever...

We pass the `ctx: ReducerContext` by move in `__call_reducer__` and in a few hops before we actually reach `reducer.invoke(&ctx, args)` where it's finally passed by reference. However, `size_of_val(&ctx)` is `400`,...

websocket: send/receive reducer & table ids instead of names (take 2)

2

# Description of Changes Receive and send reducer ids and table ids as opposed to names in the SDK. Alternative to https://github.com/clockworklabs/SpacetimeDB/pull/1883 which is more efficient, but also uses indices...

wip

1

# Description of Changes Please describe your change, mention any related tickets, and so on here. # API and ABI breaking changes If this is an API or ABI breaking...

TODO... - freelist?

Support `UniqueColumn::try_update`

1

Since there's `::try_insert`, we'd also like a `try_update` to fit with the new update ABI.