Perry Kundert

Yes, sorry -- it grew as I realized there was a major bug in bitbuffer_add_row. I think it's worth taking the plunge, though. Future decoders will be quite a bit...

Brought up-to-date with master; fixed a defect in another protocol, schraeder.c that prevented building...

I've corrected a bunch of decoders that were based on the assumption that the first row was always empty, but there might be more. I ran the rtl_433_tests, and I...

OK, I've discovered one interesting thing. Using HashSet/Map in committed entries is a no-no -- it doesn't De/Serialize deterministically, so the same entry committed, deserialized then re-serialized and hash-checked will...

This test still fails. If any Zome init function does a create_entry, and there are multiple init functions, the source-chain will be corrupted. Commenting out one line (the create_entry in...

Hi, @thedavidmeister; this test still fails, and shows at least 1 serious issue exists in the handling of the source-chain by init functions. This will fail the wasm_test (run as...

@zippy @thedavidmeister I think I have made a breakthrough; after considerable mucking around with removing functionality from my Zomes until they worked -- I have confirmed that *any* Zome function...

@zippy @thedavidmeister Another data-point: one of my Zome functions was failing (after it returned!) with an Unreachable error. No logging from the validate_ functions; not called? I removed (commented out)...

@zippy OK, this defect seems to be the most urgent for me, now -- and it seems to prevent anyone from writing a non-trivial Zome (one with more than a...

@thedavidmeister After merging the latest `develop` with your hdk3 API improvements, we're getting a bit better error context: ``` FATAL PANIC PanicInfo { payload: Any, message: Some( assertion failed: `Ok(Err(WasmError(Host("Wasm...