Roman Shanin

Imo it kinda strange to have `Register` and `Unregister` for assets, so implicit balance update through `Mint`, `Burn`, `Transfer` should be enough.

Simplest solution is to just increase `transaction_gossip_period_ms`, but it still has a problem that the same transactions are sent over and over again.

Transaction i've sent to iroha: transfer.json: ```json { "Transfer": { "Asset": { "Numeric": { "source": "rose##ed0120CE7FA46C9DCE7EA4B125E2E36BDB63EA33073E7590AC92816AE1E861B7048B03@wonderland", "object": "10", "destination": "ed0120CE7FA46C9DCE7EA4B125E2E36BDB63EA33073E7590AC92816AE1E861B7048B03@wonderland" } } } } ``` transaction.json: ```bash yes transfer.json |...

> I'm more concerned about if transactions are assumed to be forwarded in one shot to all of validators Can you clarify your concern?

> But Torii tests are failing. Tests are fixed + added extra test for different wrong body errors, the problem was mainly because pytests didn't set proper headers.

@AlexStroke please take a look at torii pytests

Ignore first commit, it's extracted in separate PR and here only to make unit tests work

Just a thought but maybe we can store our entites in zerocopy way so that we can omit costly serialization and deserialization? smt like [apache arrow](https://arrow.apache.org/docs/format/Columnar.htm) But this would be...

i agree, still it would lower the overhead, sadly we can't share the data with executor

I agree, out of curiosity i've executed `validate_block` benchmark with wasm profiling support: [results](https://share.firefox.dev/3XSIum5). ``` cd default_executor RUSTFLAGS="-C force-frame-pointers=on" cargo +nightly build -Z build-std -Z build-std-features=panic_immediate_abort -Z unstable-options --target wasm32-unknown-unknown...