nimbus-eth1/nimbus/evm/async/data_sources/json_rpc_data_source.nim

510 lines
25 KiB
Nim
Raw Normal View History

import
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
std/[sequtils, typetraits, options, times],
chronicles,
chronos,
nimcrypto,
stint,
stew/byteutils,
json_rpc/rpcclient,
eth/common,
eth/rlp,
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
eth/trie/hexary_proof_verification,
eth/p2p,
eth/p2p/rlpx,
eth/p2p/private/p2p_types,
../../../sync/protocol,
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
../../../db/[core_db, distinct_tries, incomplete_db, storage_types],
../data_sources
2023-04-17 08:22:51 +00:00
when defined(legacy_eth66_enabled):
import
2023-04-17 08:22:51 +00:00
../../../sync/protocol/eth66 as proto_eth66
from ../../../sync/protocol/eth66 import getNodeData
from ../../../rpc/rpc_utils import toHash
from web3 import Web3, BlockHash, BlockObject, FixedBytes, Address, ProofResponse, StorageProof, newWeb3, fromJson, fromHex, eth_getBlockByHash, eth_getBlockByNumber, eth_getCode, eth_getProof, blockId, `%`
2023-06-04 06:00:50 +00:00
from web3/ethtypes import Quantity
#from ../../../premix/downloader import request
#from ../../../premix/parser import prefixHex, parseBlockHeader, parseReceipt, parseTransaction
# Trying to do things the new web3 way:
from ../../../nimbus_verified_proxy/validate_proof import getAccountFromProof
export AsyncOperationFactory, AsyncDataSource
var durationSpentDoingFetches*: times.Duration
var fetchCounter*: int
func toHash*(s: string): Hash256 {.raises: [ValueError].} =
hexToPaddedByteArray[32](s).toHash
2023-06-12 04:29:03 +00:00
func toHash*(h: BlockHash): Hash256 {.raises: [].} =
distinctBase(h).toHash
func toWeb3BlockHash*(h: Hash256): BlockHash =
BlockHash(h.data)
func web3AddressToEthAddress(a: web3.Address): EthAddress =
distinctBase(a)
proc makeAnRpcClient*(web3Url: string): Future[RpcClient] {.async.} =
let myWeb3: Web3 = waitFor(newWeb3(web3Url))
return myWeb3.provider
#[
BlockObject* = ref object
number*: Quantity # the block number. null when its pending block.
hash*: Hash256 # hash of the block. null when its pending block.
parentHash*: Hash256 # hash of the parent block.
sha3Uncles*: Hash256 # SHA3 of the uncles data in the block.
logsBloom*: FixedBytes[256] # the bloom filter for the logs of the block. null when its pending block.
transactionsRoot*: Hash256 # the root of the transaction trie of the block.
stateRoot*: Hash256 # the root of the final state trie of the block.
receiptsRoot*: Hash256 # the root of the receipts trie of the block.
miner*: Address # the address of the beneficiary to whom the mining rewards were given.
difficulty*: UInt256 # integer of the difficulty for this block.
extraData*: DynamicBytes[0, 32] # the "extra data" field of this block.
gasLimit*: Quantity # the maximum gas allowed in this block.
gasUsed*: Quantity # the total used gas by all transactions in this block.
timestamp*: Quantity # the unix timestamp for when the block was collated.
nonce*: Option[FixedBytes[8]] # hash of the generated proof-of-work. null when its pending block.
size*: Quantity # integer the size of this block in bytes.
totalDifficulty*: UInt256 # integer of the total difficulty of the chain until this block.
transactions*: seq[TxHash] # list of transaction objects, or 32 Bytes transaction hashes depending on the last given parameter.
uncles*: seq[Hash256] # list of uncle hashes.
baseFeePerGas*: Option[UInt256] # EIP-1559
withdrawalsRoot*: Option[Hash256] # EIP-4895
excessBlobGas*: Option[UInt256] # EIP-4844
]#
2023-06-04 06:00:50 +00:00
func fromQty(x: Option[Quantity]): Option[uint64] =
if x.isSome:
some(x.get().uint64)
else:
none(uint64)
func blockHeaderFromBlockObject(o: BlockObject): BlockHeader =
let nonce: BlockNonce = if o.nonce.isSome: distinctBase(o.nonce.get) else: default(BlockNonce)
BlockHeader(
parentHash: o.parentHash.toHash,
ommersHash: o.sha3Uncles.toHash,
coinbase: o.miner.web3AddressToEthAddress,
stateRoot: o.stateRoot.toHash,
txRoot: o.transactionsRoot.toHash,
receiptRoot: o.receiptsRoot.toHash,
bloom: distinctBase(o.logsBloom),
difficulty: o.difficulty,
blockNumber: distinctBase(o.number).u256,
gasLimit: int64(distinctBase(o.gasLimit)),
gasUsed: int64(distinctBase(o.gasUsed)),
timestamp: initTime(int64(distinctBase(o.timestamp)), 0),
extraData: distinctBase(o.extraData),
#mixDigest: o.mixHash.toHash, # AARDVARK what's this?
nonce: nonce,
fee: o.baseFeePerGas,
withdrawalsRoot: o.withdrawalsRoot.map(toHash),
blobGasUsed: fromQty(o.blobGasUsed),
excessBlobGas: fromQty(o.excessBlobGas)
)
proc fetchBlockHeaderWithHash*(rpcClient: RpcClient, h: Hash256): Future[BlockHeader] {.async.} =
let t0 = now()
let blockObject: BlockObject = await rpcClient.eth_getBlockByHash(h.toWeb3BlockHash, false)
durationSpentDoingFetches += now() - t0
fetchCounter += 1
return blockHeaderFromBlockObject(blockObject)
proc fetchBlockHeaderWithNumber*(rpcClient: RpcClient, n: BlockNumber): Future[BlockHeader] {.async.} =
let t0 = now()
let bid = blockId(n.truncate(uint64))
let blockObject: BlockObject = await rpcClient.eth_getBlockByNumber(bid, false)
durationSpentDoingFetches += now() - t0
fetchCounter += 1
return blockHeaderFromBlockObject(blockObject)
#[
proc parseBlockBodyAndFetchUncles(rpcClient: RpcClient, r: JsonNode): Future[BlockBody] {.async.} =
var body: BlockBody
for tn in r["transactions"].getElems:
body.transactions.add(parseTransaction(tn))
for un in r["uncles"].getElems:
let uncleHash: Hash256 = un.getStr.toHash
let uncleHeader = await fetchBlockHeaderWithHash(rpcClient, uncleHash)
body.uncles.add(uncleHeader)
return body
proc fetchBlockHeaderAndBodyWithHash*(rpcClient: RpcClient, h: Hash256): Future[(BlockHeader, BlockBody)] {.async.} =
let t0 = now()
let r = request("eth_getBlockByHash", %[%h.prefixHex, %true], some(rpcClient))
durationSpentDoingFetches += now() - t0
fetchCounter += 1
if r.kind == JNull:
error "requested block not available", blockHash=h
raise newException(ValueError, "Error when retrieving block header and body")
let header = parseBlockHeader(r)
let body = await parseBlockBodyAndFetchUncles(rpcClient, r)
return (header, body)
proc fetchBlockHeaderAndBodyWithNumber*(rpcClient: RpcClient, n: BlockNumber): Future[(BlockHeader, BlockBody)] {.async.} =
let t0 = now()
let r = request("eth_getBlockByNumber", %[%n.prefixHex, %true], some(rpcClient))
durationSpentDoingFetches += now() - t0
fetchCounter += 1
if r.kind == JNull:
error "requested block not available", blockNumber=n
raise newException(ValueError, "Error when retrieving block header and body")
let header = parseBlockHeader(r)
let body = await parseBlockBodyAndFetchUncles(rpcClient, r)
return (header, body)
]#
proc fetchBlockHeaderAndBodyWithHash*(rpcClient: RpcClient, h: Hash256): Future[(BlockHeader, BlockBody)] {.async.} =
doAssert(false, "AARDVARK not implemented")
proc fetchBlockHeaderAndBodyWithNumber*(rpcClient: RpcClient, n: BlockNumber): Future[(BlockHeader, BlockBody)] {.async.} =
doAssert(false, "AARDVARK not implemented")
func mdigestFromFixedBytes*(arg: FixedBytes[32]): MDigest[256] =
MDigest[256](data: distinctBase(arg))
func mdigestFromString*(s: string): MDigest[256] =
mdigestFromFixedBytes(FixedBytes[32].fromHex(s))
type
AccountProof* = seq[seq[byte]]
proc fetchAccountAndSlots*(rpcClient: RpcClient, address: EthAddress, slots: seq[UInt256], blockNumber: BlockNumber): Future[(Account, AccountProof, seq[StorageProof])] {.async.} =
let t0 = now()
debug "Got to fetchAccountAndSlots", address=address, slots=slots, blockNumber=blockNumber
#let blockNumberHexStr: HexQuantityStr = encodeQuantity(blockNumber)
let blockNumberUint64 = blockNumber.truncate(uint64)
let a = web3.Address(address)
let bid = blockId(blockNumber.truncate(uint64))
debug "About to call eth_getProof", address=address, slots=slots, blockNumber=blockNumber
let proofResponse: ProofResponse = await rpcClient.eth_getProof(a, slots, bid)
debug "Received response to eth_getProof", proofResponse=proofResponse
let acc = Account(
nonce: distinctBase(proofResponse.nonce),
balance: proofResponse.balance,
storageRoot: mdigestFromFixedBytes(proofResponse.storageHash),
codeHash: mdigestFromFixedBytes(proofResponse.codeHash)
)
debug "Parsed response to eth_getProof", acc=acc
let mptNodesBytes: seq[seq[byte]] = proofResponse.accountProof.mapIt(distinctBase(it))
durationSpentDoingFetches += now() - t0
fetchCounter += 1
return (acc, mptNodesBytes, proofResponse.storageProof)
proc fetchCode*(client: RpcClient, blockNumber: BlockNumber, address: EthAddress): Future[seq[byte]] {.async.} =
let t0 = now()
let a = web3.Address(address)
let bid = blockId(blockNumber.truncate(uint64))
let fetchedCode: seq[byte] = await client.eth_getCode(a, bid)
durationSpentDoingFetches += now() - t0
fetchCounter += 1
return fetchedCode
const bytesLimit = 2 * 1024 * 1024
const maxNumberOfPeersToAttempt = 3
proc fetchUsingGetTrieNodes(peer: Peer, stateRoot: Hash256, paths: seq[SnapTriePaths]): Future[seq[seq[byte]]] {.async.} =
let r = await peer.getTrieNodes(stateRoot, paths, bytesLimit)
if r.isNone:
raise newException(CatchableError, "AARDVARK: received None in GetTrieNodes response")
else:
return r.get.nodes
proc fetchUsingGetNodeData(peer: Peer, nodeHashes: seq[Hash256]): Future[seq[seq[byte]]] {.async.} =
#[
let r: Option[seq[seq[byte]]] = none[seq[seq[byte]]]() # AARDVARK await peer.getNodeData(nodeHashes)
if r.isNone:
raise newException(CatchableError, "AARDVARK: received None in GetNodeData response")
else:
echo "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA fetchUsingGetNodeData received nodes: " & $(r.get.data)
return r.get.data
]#
# AARDVARK whatever
return @[]
proc findPeersAndMakeSomeCalls[R](peerPool: PeerPool, protocolName: string, protocolType: typedesc, initiateAttempt: (proc(p: Peer): Future[R] {.gcsafe.})): Future[seq[Future[R]]] {.async.} =
var attempts: seq[Future[R]]
while true:
#info("AARDVARK: findPeersAndMakeSomeCalls about to loop through the peer pool", count=peerPool.connectedNodes.len)
for nodeOfSomeSort, peer in peerPool.connectedNodes:
if peer.supports(protocolType):
info("AARDVARK: findPeersAndMakeSomeCalls calling peer", protocolName, peer)
attempts.add(initiateAttempt(peer))
if attempts.len >= maxNumberOfPeersToAttempt:
break
#else:
# info("AARDVARK: peer does not support protocol", protocolName, peer)
if attempts.len == 0:
warn("AARDVARK: findPeersAndMakeSomeCalls did not find any peers; waiting and trying again", protocolName, totalPeerPoolSize=peerPool.connectedNodes.len)
await sleepAsync(5000)
else:
if attempts.len < maxNumberOfPeersToAttempt:
warn("AARDVARK: findPeersAndMakeSomeCalls did not find enough peers, but found some", protocolName, totalPeerPoolSize=peerPool.connectedNodes.len, found=attempts.len)
break
return attempts
proc findPeersAndMakeSomeAttemptsToCallGetTrieNodes(peerPool: PeerPool, stateRoot: Hash256, paths: seq[SnapTriePaths]): Future[seq[Future[seq[seq[byte]]]]] =
findPeersAndMakeSomeCalls(peerPool, "snap", protocol.snap, (proc(peer: Peer): Future[seq[seq[byte]]] = fetchUsingGetTrieNodes(peer, stateRoot, paths)))
#[
proc findPeersAndMakeSomeAttemptsToCallGetNodeData(peerPool: PeerPool, stateRoot: Hash256, nodeHashes: seq[Hash256]): Future[seq[Future[seq[seq[byte]]]]] =
findPeersAndMakeSomeCalls(peerPool, "eth66", eth66, (proc(peer: Peer): Future[seq[seq[byte]]] = fetchUsingGetNodeData(peer, nodeHashes)))
]#
proc fetchNodes(peerPool: PeerPool, stateRoot: Hash256, paths: seq[SnapTriePaths], nodeHashes: seq[Hash256]): Future[seq[seq[byte]]] {.async.} =
let attempts = await findPeersAndMakeSomeAttemptsToCallGetTrieNodes(peerPool, stateRoot, paths)
#let attempts = await findPeersAndMakeSomeAttemptsToCallGetNodeData(peerPool, stateRoot, nodeHashes)
let completedAttempt = await one(attempts)
let nodes: seq[seq[byte]] = completedAttempt.read
info("AARDVARK: fetchNodes received nodes", nodes)
return nodes
proc verifyFetchedAccount(stateRoot: Hash256, address: EthAddress, acc: Account, accProof: seq[seq[byte]]): Result[void, string] =
let accKey = toSeq(keccakHash(address).data)
let accEncoded = rlp.encode(acc)
let accProofResult = verifyMptProof(accProof, stateRoot, accKey, accEncoded)
case accProofResult.kind
of ValidProof:
return ok()
of MissingKey:
# For an account that doesn't exist yet, which is fine.
return ok()
of InvalidProof:
return err(accProofResult.errorMsg)
type
CodeFetchingInfo = tuple[blockNumber: BlockNumber, address: EthAddress]
proc fetchCode(client: RpcClient, p: CodeFetchingInfo): Future[seq[byte]] {.async.} =
let (blockNumber, address) = p
let fetchedCode = await fetchCode(client, blockNumber, address)
return fetchedCode
proc verifyFetchedCode(fetchedCode: seq[byte], desiredCodeHash: Hash256): Result[void, Hash256] =
let fetchedCodeHash = keccakHash(fetchedCode)
2023-06-04 06:00:50 +00:00
if desiredCodeHash == fetchedCodeHash:
ok()
else:
err(fetchedCodeHash)
proc fetchAndVerifyCode(client: RpcClient, p: CodeFetchingInfo, desiredCodeHash: Hash256): Future[seq[byte]] {.async.} =
let fetchedCode: seq[byte] = await fetchCode(client, p)
let verificationRes = verifyFetchedCode(fetchedCode, desiredCodeHash)
if verificationRes.isOk():
return fetchedCode
else:
let fetchedCodeHash = verificationRes.error
error("code hash values do not match", p=p, desiredCodeHash=desiredCodeHash, fetchedCodeHash=fetchedCodeHash)
raise newException(CatchableError, "async code received code for " & $(p.address) & " whose hash (" & $(fetchedCodeHash) & ") does not match the desired hash (" & $(desiredCodeHash) & ")")
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
proc putCode*(db: CoreDbRef, codeHash: Hash256, code: seq[byte]) =
when defined(geth):
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
db.kvt.put(codeHash.data, code)
else:
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
db.kvt.put(contractHashKey(codeHash).toOpenArray, code)
proc putCode*(trie: AccountsTrie, codeHash: Hash256, code: seq[byte]) =
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
putCode(trie.db, codeHash, code)
proc storeCode(trie: AccountsTrie, p: CodeFetchingInfo, desiredCodeHash: Hash256, fetchedCode: seq[byte]) =
trie.putCode(desiredCodeHash, fetchedCode)
proc assertThatWeHaveStoredCode(trie: AccountsTrie, p: CodeFetchingInfo, codeHash: Hash256) =
# FIXME-Adam: this is a bit wrong because we're not checking it against the blockNumber, only the address. (That is,
# if the code for this address has *changed* (which is unlikely), this check isn't the right thing to do.)
let maybeFoundCode = trie.maybeGetCode(p.address)
if maybeFoundCode.isNone:
error("code didn't get put into the db", p=p, codeHash=codeHash)
doAssert false, "code didn't get put into the db"
else:
let foundCode = maybeFoundCode.get
let foundCodeHash = keccakHash(foundCode)
if foundCodeHash != codeHash:
error("code does not have the right hash", p=p, codeHash=codeHash, foundCode=foundCode)
doAssert false, "code does not have the right hash"
proc assertThatWeHaveStoredAccount(trie: AccountsTrie, address: EthAddress, fetchedAcc: Account, isForTheNewTrie: bool = false) =
let foundAcc = ifNodesExistGetAccount(trie, address).get
if fetchedAcc != foundAcc:
error "account didn't come out the same", address=address, fetchedAcc=fetchedAcc, foundAcc=foundAcc, isForTheNewTrie=isForTheNewTrie
doAssert false, "account didn't come out the same"
doAssert(trie.hasAllNodesForAccount(address), "Can I check the account this way, too?")
proc verifyFetchedSlot(accountStorageRoot: Hash256, slot: UInt256, fetchedVal: UInt256, storageMptNodes: seq[seq[byte]]): Result[void, string] =
if storageMptNodes.len == 0:
# I think an empty storage proof is okay; I see lots of these
# where the account is empty and the value is zero.
return ok()
else:
let storageKey = toSeq(keccakHash(toBytesBE(slot)).data)
let storageValueEncoded = rlp.encode(fetchedVal)
let storageProofResult = verifyMptProof(storageMptNodes, accountStorageRoot, storageKey, storageValueEncoded)
case storageProofResult.kind
of ValidProof:
return ok()
of MissingKey:
# This is for a slot that doesn't have anything stored at it, but that's fine.
return ok()
of InvalidProof:
return err(storageProofResult.errorMsg)
proc assertThatWeHaveStoredSlot(trie: AccountsTrie, address: EthAddress, acc: Account, slot: UInt256, fetchedVal: UInt256, isForTheNewTrie: bool = false) =
if acc.storageRoot == EMPTY_ROOT_HASH and fetchedVal.isZero:
# I believe this is okay.
discard
else:
let foundVal = ifNodesExistGetStorage(trie, address, slot).get
if (fetchedVal != foundVal):
error("slot didn't come out the same", address=address, slot=slot, fetchedVal=fetchedVal, foundVal=foundVal, isForTheNewTrie=isForTheNewTrie)
doAssert false, "slot didn't come out the same"
proc verifyFetchedBlockHeader(fetchedHeader: BlockHeader, desiredBlockNumber: BlockNumber): Result[void, BlockNumber] =
# *Can* we do anything to verify this header, given that all we know
# is the desiredBlockNumber and we want to run statelessly so we don't
# know what block hash we want?
2023-06-04 06:00:50 +00:00
ok()
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
proc storeBlockHeader(chainDB: CoreDbRef, header: BlockHeader) =
chainDB.persistHeaderToDbWithoutSetHeadOrScore(header)
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
proc assertThatWeHaveStoredBlockHeader(chainDB: CoreDbRef, blockNumber: BlockNumber, header: BlockHeader) =
let h = chainDB.getBlockHash(blockNumber)
doAssert(h == header.blockHash, "stored the block header for block " & $(blockNumber))
proc raiseExceptionIfError[V, E](whatAreWeVerifying: V, r: Result[void, E]) =
if r.isErr:
error("async code failed to verify", whatAreWeVerifying=whatAreWeVerifying, err=r.error)
raise newException(CatchableError, "async code failed to verify: " & $(whatAreWeVerifying) & ", error is: " & $(r.error))
const shouldDoUnnecessarySanityChecks = true
# This proc fetches both the account and also optionally some of its slots, because that's what eth_getProof can do.
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
proc ifNecessaryGetAccountAndSlots*(client: RpcClient, db: CoreDbRef, blockNumber: BlockNumber, stateRoot: Hash256, address: EthAddress, slots: seq[UInt256], justCheckingAccount: bool, justCheckingSlots: bool, newStateRootForSanityChecking: Hash256): Future[void] {.async.} =
let trie = initAccountsTrie(db, stateRoot, false) # important for sanity checks
let trie2 = initAccountsTrie(db, newStateRootForSanityChecking, false) # important for sanity checks
let doesAccountActuallyNeedToBeFetched = not trie.hasAllNodesForAccount(address)
let slotsToActuallyFetch = slots.filter(proc(slot: UInt256): bool = not (trie.hasAllNodesForStorageSlot(address, slot)))
if (not doesAccountActuallyNeedToBeFetched) and (slotsToActuallyFetch.len == 0):
# Already have them, no need to fetch either the account or the slots
discard
else:
let (acc, accProof, storageProofs) = await fetchAccountAndSlots(client, address, slotsToActuallyFetch, blockNumber)
# We need to verify the proof even if we already had this account,
# to make sure the data is valid.
let accountVerificationRes = verifyFetchedAccount(stateRoot, address, acc, accProof)
let whatAreWeVerifying = ("account proof", address, acc)
raiseExceptionIfError(whatAreWeVerifying, accountVerificationRes)
if not doesAccountActuallyNeedToBeFetched:
# We already had the account, no need to populate the DB with it again.
discard
else:
if not justCheckingAccount:
populateDbWithBranch(db, accProof)
if shouldDoUnnecessarySanityChecks:
assertThatWeHaveStoredAccount(trie, address, acc, false)
if doesAccountActuallyNeedToBeFetched: # this second check makes no sense if it's not the first time
assertThatWeHaveStoredAccount(trie2, address, acc, true)
doAssert(slotsToActuallyFetch.len == storageProofs.len, "We should get back the same number of storage proofs as slots that we asked for. I think.")
for storageProof in storageProofs:
let slot: UInt256 = storageProof.key
let fetchedVal: UInt256 = storageProof.value
let storageMptNodes: seq[seq[byte]] = storageProof.proof.mapIt(distinctBase(it))
let storageVerificationRes = verifyFetchedSlot(acc.storageRoot, slot, fetchedVal, storageMptNodes)
let whatAreWeVerifying = ("storage proof", address, acc, slot, fetchedVal)
raiseExceptionIfError(whatAreWeVerifying, storageVerificationRes)
if not justCheckingSlots:
populateDbWithBranch(db, storageMptNodes)
# I believe this is done so that we can iterate over the slots. See
# persistStorage in accounts_cache.nim.
let slotAsKey = createTrieKeyFromSlot(slot)
let slotHash = keccakHash(slotAsKey)
let slotEncoded = rlp.encode(slot)
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
db.kvt.put(slotHashToSlotKey(slotHash.data).toOpenArray, slotEncoded)
if shouldDoUnnecessarySanityChecks:
assertThatWeHaveStoredSlot(trie, address, acc, slot, fetchedVal, false)
assertThatWeHaveStoredSlot(trie2, address, acc, slot, fetchedVal, true)
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
proc ifNecessaryGetCode*(client: RpcClient, db: CoreDbRef, blockNumber: BlockNumber, stateRoot: Hash256, address: EthAddress, justChecking: bool, newStateRootForSanityChecking: Hash256): Future[void] {.async.} =
await ifNecessaryGetAccountAndSlots(client, db, blockNumber, stateRoot, address, @[], false, false, newStateRootForSanityChecking) # to make sure we've got the codeHash
let trie = initAccountsTrie(db, stateRoot, false) # important for sanity checks
let acc = ifNodesExistGetAccount(trie, address).get
let desiredCodeHash = acc.codeHash
let p = (blockNumber, address)
if not(trie.hasAllNodesForCode(address)):
let fetchedCode = await fetchAndVerifyCode(client, p, desiredCodeHash)
if not justChecking:
storeCode(trie, p, desiredCodeHash, fetchedCode)
if shouldDoUnnecessarySanityChecks:
assertThatWeHaveStoredCode(trie, p, desiredCodeHash)
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
proc ifNecessaryGetBlockHeaderByNumber*(client: RpcClient, chainDB: CoreDbRef, blockNumber: BlockNumber, justChecking: bool): Future[void] {.async.} =
let maybeHeaderAndHash = chainDB.getBlockHeaderWithHash(blockNumber)
if maybeHeaderAndHash.isNone:
let fetchedHeader = await fetchBlockHeaderWithNumber(client, blockNumber)
let headerVerificationRes = verifyFetchedBlockHeader(fetchedHeader, blockNumber)
let whatAreWeVerifying = ("block header by number", blockNumber, fetchedHeader)
raiseExceptionIfError(whatAreWeVerifying, headerVerificationRes)
if not justChecking:
storeBlockHeader(chainDB, fetchedHeader)
if shouldDoUnnecessarySanityChecks:
assertThatWeHaveStoredBlockHeader(chainDB, blockNumber, fetchedHeader)
# Used in asynchronous on-demand-data-fetching mode.
proc realAsyncDataSource*(peerPool: PeerPool, client: RpcClient, justChecking: bool): AsyncDataSource =
AsyncDataSource(
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
ifNecessaryGetAccount: (proc(db: CoreDbRef, blockNumber: BlockNumber, stateRoot: Hash256, address: EthAddress, newStateRootForSanityChecking: Hash256): Future[void] {.async.} =
await ifNecessaryGetAccountAndSlots(client, db, blockNumber, stateRoot, address, @[], false, false, newStateRootForSanityChecking)
),
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
ifNecessaryGetSlots: (proc(db: CoreDbRef, blockNumber: BlockNumber, stateRoot: Hash256, address: EthAddress, slots: seq[UInt256], newStateRootForSanityChecking: Hash256): Future[void] {.async.} =
await ifNecessaryGetAccountAndSlots(client, db, blockNumber, stateRoot, address, slots, false, false, newStateRootForSanityChecking)
),
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
ifNecessaryGetCode: (proc(db: CoreDbRef, blockNumber: BlockNumber, stateRoot: Hash256, address: EthAddress, newStateRootForSanityChecking: Hash256): Future[void] {.async.} =
await ifNecessaryGetCode(client, db, blockNumber, stateRoot, address, justChecking, newStateRootForSanityChecking)
),
Unified database frontend integration (#1670) * Nimbus folder environment update details: * Integrated `CoreDbRef` for the sources in the `nimbus` sub-folder. * The `nimbus` program does not compile yet as it needs the updates in the parallel `stateless` sub-folder. * Stateless environment update details: * Integrated `CoreDbRef` for the sources in the `stateless` sub-folder. * The `nimbus` program compiles now. * Premix environment update details: * Integrated `CoreDbRef` for the sources in the `premix` sub-folder. * Fluffy environment update details: * Integrated `CoreDbRef` for the sources in the `fluffy` sub-folder. * Tools environment update details: * Integrated `CoreDbRef` for the sources in the `tools` sub-folder. * Nodocker environment update details: * Integrated `CoreDbRef` for the sources in the `hive_integration/nodocker` sub-folder. * Tests environment update details: * Integrated `CoreDbRef` for the sources in the `tests` sub-folder. * The unit tests compile and run cleanly now. * Generalise `CoreDbRef` to any `select_backend` supported database why: Generalisation was just missed due to overcoming some compiler oddity which was tied to rocksdb for testing. * Suppress compiler warning for `newChainDB()` why: Warning was added to this function which must be wrapped so that any `CatchableError` is re-raised as `Defect`. * Split off persistent `CoreDbRef` constructor into separate file why: This allows to compile a memory only database version without linking the backend library. * Use memory `CoreDbRef` database by default detail: Persistent DB constructor needs to import `db/core_db/persistent why: Most tests use memory DB anyway. This avoids linking `-lrocksdb` or any other backend by default. * fix `toLegacyBackend()` availability check why: got garbled after memory/persistent split. * Clarify raw access to MPT for snap sync handler why: Logically, `kvt` is not the raw access for the hexary trie (although this holds for the legacy database)
2023-08-04 11:10:09 +00:00
ifNecessaryGetBlockHeaderByNumber: (proc(chainDB: CoreDbRef, blockNumber: BlockNumber): Future[void] {.async.} =
await ifNecessaryGetBlockHeaderByNumber(client, chainDB, blockNumber, justChecking)
),
# FIXME-Adam: This will be needed later, but for now let's just get the basic methods in place.
#fetchNodes: (proc(stateRoot: Hash256, paths: seq[seq[seq[byte]]], nodeHashes: seq[Hash256]): Future[seq[seq[byte]]] {.async.} =
# return await fetchNodes(peerPool, stateRoot, paths, nodeHashes)
#),
fetchBlockHeaderWithHash: (proc(h: Hash256): Future[BlockHeader] {.async.} =
return await fetchBlockHeaderWithHash(client, h)
),
fetchBlockHeaderWithNumber: (proc(n: BlockNumber): Future[BlockHeader] {.async.} =
return await fetchBlockHeaderWithNumber(client, n)
),
fetchBlockHeaderAndBodyWithHash: (proc(h: Hash256): Future[(BlockHeader, BlockBody)] {.async.} =
return await fetchBlockHeaderAndBodyWithHash(client, h)
),
fetchBlockHeaderAndBodyWithNumber: (proc(n: BlockNumber): Future[(BlockHeader, BlockBody)] {.async.} =
return await fetchBlockHeaderAndBodyWithNumber(client, n)
)
)