pathfinder
pathfinder copied to clipboard
eqlabs
A Starknet full node written in Rust
This should be done last. Add caching to the transaction and event proofs. We don't store the merkle trie for these during sync - we only calculate them once off...
`/rpc/pathfinder/v0.1` should also include `/rpc/pathfinder/v0_1` to follow standard RPC conventions.
- use a test version of peer_aware::Client to inject responses - test happy path and if particular failures map to proper error variants - use rstest to reduce boilerplate duplication...
sync_handlers proptests proved to be hard to maintain and understand. Execution times are also a big burden.
[AWS Blockchain Node Runners](https://aws-samples.github.io/aws-blockchain-node-runners/) are a way to easily deploy blockchain nodes. [Here](https://github.com/aws-samples/aws-blockchain-node-runners/blob/main/docs/adding-new-nodes.md) are the instructions for how a new blueprint can be added for Pathfinder.
By default these tests are ignored, nevertheless they rely on goerli which leads to fails rn: ``` failures: v04::method::add_declare_transaction::tests::duplicate_transaction v04::method::add_declare_transaction::tests::duplicate_v3_transaction v04::method::add_declare_transaction::tests::insufficient_account_balance v04::method::add_declare_transaction::tests::insufficient_max_fee v04::method::add_deploy_account_transaction::tests::duplicate_transaction v04::method::add_deploy_account_transaction::tests::duplicate_v3_transaction v04::method::add_invoke_transaction::tests::duplicate_transaction v04::method::add_invoke_transaction::tests::duplicate_v3_transaction v06::method::add_declare_transaction::tests::duplicate_transaction v06::method::add_declare_transaction::tests::duplicate_v3_transaction v06::method::add_declare_transaction::tests::insufficient_account_balance v06::method::add_declare_transaction::tests::insufficient_max_fee...
Investigate the following, based on @Mirko-von-Leipzig [comment](https://github.com/eqlabs/pathfinder/pull/1838#discussion_r1514255815): > What do you think of having a separate stage just for state trie operations? As in, this stage just writes the state...
Sync can often by slow or fail to progress when the gateway is down, slow or rate-limiting the node. Currently we silently retry gateway requests until they eventually pass, which...
Starknet v0.13.0, v0.13.1 and v0.13.2 introduced several new block header fields, which we've made optional in our DTOs in order to support the version transitions. Once all networks have deployed...