Gas-based limit implementation to retrieve unproposed transactions. The limit is maintained by the constant 'GAS_LIMIT_UNPROPOSED_TXS', currently calculated as the product of the average total gas used and a gas limit multiplier equivalent to the existing TX_CAP value (i.e, 50).
The average total gas was obtained through analysis of empirical transaction test data, using a gas analysis tool.
The aim of this approach is to allow a gradual and controlled transition towards an optimal gas-based system. This minimizes potential adverse effects brought by changes to TX_CAP's implementation and provides the benefits of using gas to limit the number of txs received. This implementation will be fine-tuned until the discovery of the most efficient formula for determining the unproposed transactions' gas limit.
Tests have been added to verify the implementation's correctness by running transactions against it. To run the tests, run the following command from the bin/darkfid directory:
cargo test --release --bin darkfid tests::unproposed_txs
- Fix issue where `half` was calculated incorrectly for odd values of
`value`
- Fix issue where sending a `value` of 1 and `half_split` of true would
result in two empty value transfers
Note that this is only a client issue. The wasm back-end would cause the
invalid transfer to fail as the tx outputs from the client would not
match the inputs
clippy linting at the workspace level for all crates
Configure all workspaces to use lints from the top-level Cargo.toml
file
Add example lints that the project could configure to improve security
and reliability.
Configure lints to warn level. Using deny level makes other binaries
fail to compile if even one of them has a failure.
No lints are added in this commit. Future changes can enable and fix
lints
