bump nimbus-build-system to use Nim v2.0.10

[tersec]

No description provided.

[arnetheduck]

newPayload issue is from #2685

[tersec]

newPayload issue is from #2685

Yeah, it's a type alias confusion:

type
  FixedBytes[N: static int] = distinct array[N, byte]

template makeFixedBytesN(N: static int) =
  # Create specific numbered instantiations along with helpers
  type `Bytes N` = FixedBytes[N]

  const `zeroBytes N` = system.default(`Bytes N`)
  template default(T: type `Bytes N`): `Bytes N` =
    # reuse single constant for precomputed N
    `zeroBytes N`

  template `bytes N`(s: static string): `Bytes N` =
    `Bytes N`.fromHex(s)

makeFixedBytesN(32)

type Hash32 = distinct Bytes32

const zeroHash32 = system.default(Hash32) ## Hash32 value consisting of all zeroes

template hash32(s: static string): Hash32 =
  s.to(Hash32)

proc engine_newPayloadV3(parentBeaconBlockRoot: Hash32): int = 1

template newPayload(
    parentBeaconBlockRoot: Bytes32): int =
  engine_newPayloadV3(
    parentBeaconBlockRoot)
  0

#discard newPayload(default(Hash32))
discard newPayload(default(Bytes32))

[tersec]

If one tries to compile fluffy/tests/rpc_tests/test_portal_rpc_client.nim, one gets:

nimbus-eth1/fluffy/tests/rpc_tests/test_portal_rpc_client.nim(119, 1) template/generic instantiation of `procSuite` from here
nimbus-eth1/vendor/nim-testutils/testutils/unittests.nim(9, 5) template/generic instantiation of `suite` from here
nimbus-eth1/fluffy/tests/rpc_tests/test_portal_rpc_client.nim(268, 3) template/generic instantiation of `asyncTest` from here
nimbus-eth1/vendor/nim-unittest2/unittest2.nim(1128, 24) template/generic instantiation of `failingOnExceptions` from here
nimbus-eth1/vendor/nim-unittest2/unittest2.nim(1132, 26) template/generic instantiation of `failingOnExceptions` from here
nimbus-eth1/vendor/nim-testutils/testutils/unittests.nim(15, 21) template/generic instantiation of `async` from here
nimbus-eth1/fluffy/tests/rpc_tests/test_portal_rpc_client.nim(290, 7) template/generic instantiation of `check` from here
nimbus-eth1/fluffy/tests/rpc_tests/test_portal_rpc_client.nim(292, 29) template/generic instantiation of `check` from here
nimbus-eth1/vendor/nimbus-build-system/vendor/Nim/lib/system/comparisons.nim(334, 13) template/generic instantiation of `==` from here
nimbus-eth1/vendor/nimbus-build-system/vendor/Nim/lib/system.nim(1719, 13) Warning: toMDigest is deprecated [Deprecated]
nimbus-eth1/vendor/nim-json-rpc/json_rpc/private/server_handler_wrapper.nim(93, 30) Error: request to generate code for .compileTime proc: $

nim-lang/Nim#24228

[tersec]

If one tries to compile fluffy/tests/rpc_tests/test_portal_rpc_client.nim, one gets:

nimbus-eth1/fluffy/tests/rpc_tests/test_portal_rpc_client.nim(119, 1) template/generic instantiation of `procSuite` from here
nimbus-eth1/vendor/nim-testutils/testutils/unittests.nim(9, 5) template/generic instantiation of `suite` from here
nimbus-eth1/fluffy/tests/rpc_tests/test_portal_rpc_client.nim(268, 3) template/generic instantiation of `asyncTest` from here
nimbus-eth1/vendor/nim-unittest2/unittest2.nim(1128, 24) template/generic instantiation of `failingOnExceptions` from here
nimbus-eth1/vendor/nim-unittest2/unittest2.nim(1132, 26) template/generic instantiation of `failingOnExceptions` from here
nimbus-eth1/vendor/nim-testutils/testutils/unittests.nim(15, 21) template/generic instantiation of `async` from here
nimbus-eth1/fluffy/tests/rpc_tests/test_portal_rpc_client.nim(290, 7) template/generic instantiation of `check` from here
nimbus-eth1/fluffy/tests/rpc_tests/test_portal_rpc_client.nim(292, 29) template/generic instantiation of `check` from here
nimbus-eth1/vendor/nimbus-build-system/vendor/Nim/lib/system/comparisons.nim(334, 13) template/generic instantiation of `==` from here
nimbus-eth1/vendor/nimbus-build-system/vendor/Nim/lib/system.nim(1719, 13) Warning: toMDigest is deprecated [Deprecated]
nimbus-eth1/vendor/nim-json-rpc/json_rpc/private/server_handler_wrapper.nim(93, 30) Error: request to generate code for .compileTime proc: $

nim-lang/Nim#24228

Addressed by

• bump nim-json-serialization for Nim 2.0.10 compatibility #2693

[tersec]

2024-10-04T05:46:32.3318985Z /home/runner/work/nimbus-eth1/nimbus-eth1/nimbus/db/ledger/backend/accounts_ledger.nim(458, 44) template/generic instantiation of `valueOr` from here
2024-10-04T05:46:32.3341242Z /home/runner/work/nimbus-eth1/nimbus-eth1/nimbus/db/ledger/backend/accounts_ledger.nim(465, 20) template/generic instantiation of `put` from here
2024-10-04T05:46:34.3562192Z /home/runner/work/nimbus-eth1/nimbus-eth1/fluffy/tools/eth_data_exporter/cl_data_exporter.nim(68, 5) template/generic instantiation of `async` from here
2024-10-04T05:46:34.3569794Z /home/runner/work/nimbus-eth1/nimbus-eth1/fluffy/tools/eth_data_exporter/cl_data_exporter.nim(79, 23) template/generic instantiation of `awaitWithTimeout` from here
2024-10-04T05:46:34.3580645Z /home/runner/work/nimbus-eth1/nimbus-eth1/fluffy/tools/eth_data_exporter/cl_data_exporter.nim(83, 9) Error: ambiguous call; both chronicles.error(eventName: static[string], props: varargs[untyped]) [template declared in /home/runner/work/nimbus-eth1/nimbus-eth1/vendor/nim-chronicles/chronicles.nim(378, 12)] and macros.error(msg: string, n: NimNode) [proc declared in /home/runner/work/nimbus-eth1/nimbus-eth1/vendor/nimbus-build-system/vendor/Nim/lib/core/macros.nim(436, 6)] match for: (string)
2024-10-04T05:46:34.4725122Z make: *** [Makefile:301: eth_data_exporter] Error 1
2024-10-04T05:46:34.4729188Z make: *** Waiting for unfinished jobs....
2024-10-04T05:46:40.4031761Z /home/runner/work/nimbus-eth1/nimbus-eth1/fluffy/tools/eth_data_exporter/cl_data_exporter.nim(68, 5) template/generic instantiation of `async` from here
2024-10-04T05:46:40.4038231Z /home/runner/work/nimbus-eth1/nimbus-eth1/fluffy/tools/eth_data_exporter/cl_data_exporter.nim(79, 23) template/generic instantiation of `awaitWithTimeout` from here
2024-10-04T05:46:40.4047242Z /home/runner/work/nimbus-eth1/nimbus-eth1/fluffy/tools/eth_data_exporter/cl_data_exporter.nim(83, 9) Error: ambiguous call; both chronicles.error(eventName: static[string], props: varargs[untyped]) [template declared in /home/runner/work/nimbus-eth1/nimbus-eth1/vendor/nim-chronicles/chronicles.nim(378, 12)] and macros.error(msg: string, n: NimNode) [proc declared in /home/runner/work/nimbus-eth1/nimbus-eth1/vendor/nimbus-build-system/vendor/Nim/lib/core/macros.nim(436, 6)] match for: (string)
2024-10-04T05:46:40.5138720Z make: *** [Makefile:301: portal_bridge] Error 1

Related to nim-lang/Nim#24231 but distinct: it persists in the version-2-2/devel branches.

There's an obvious workaround of using chronicles.error ... in relevant places but would be nice to be able to avoid that.

[metagn]

Error: ambiguous call; both chronicles.error(eventName: static[string], props: varargs[untyped]) and macros.error(msg: string, n: NimNode) match for: (string)

I think the issue here is that not-provided arguments count for the disambiguation (in sumGeneric: NimNode = ref object = ref + object = 2, varargs[untyped] = varargs + untyped = 1 + 0 but static[string] = static + string = 2). There would still be a problem if chronicles.error used string instead of static[string] but this is expected. Edit: Opened nim-lang/Nim#24232 for this

I think macros.error is too broadly named, import macros except error then qualifying macros.error if necessary would probably be easier. Renaming it in Nim would be nice too.

[tersec]

I think the issue here is that not-provided arguments count for the disambiguation (in sumGeneric: NimNode = ref object = ref + object = 2, varargs[untyped] = varargs + untyped = 1 + 0 but static[string] = static + string = 2). There would still be a problem if chronicles.error used string instead of static[string] but this is expected.

I think macros.error is too broadly named, import macros except error then qualifying macros.error if necessary would probably be easier. Renaming it in Nim would be nice too.

The except approach seems reasonable, and can be done where it's imported rather than throughout the codebase. One challenge with these quite general symbols is that they end up visible throughout both nimbus-eth1 and nimbus-eth2 repos' code because both use import/export networks. e.g., to avoid generics sandwich issues. So one ends up with some code which doesn't directly know or care about macros accidentally having macros symbols from 3 import layers down bubble up. A solution (such as chronicles.foo) isn't just ugly, it scales badly too. By contrast, the except approach seems reasonable.

But, yeah, it would be nicer not to require. error is some prime symbol real estate macros is using there.