2022-08-24 13:44:18 +00:00
|
|
|
# Nimbus
|
|
|
|
# Copyright (c) 2018-2019 Status Research & Development GmbH
|
|
|
|
# Licensed under either of
|
|
|
|
# * Apache License, version 2.0, ([LICENSE-APACHE](LICENSE-APACHE) or
|
|
|
|
# http://www.apache.org/licenses/LICENSE-2.0)
|
|
|
|
# * MIT license ([LICENSE-MIT](LICENSE-MIT) or
|
|
|
|
# http://opensource.org/licenses/MIT)
|
|
|
|
# at your option. This file may not be copied, modified, or distributed except
|
|
|
|
# according to those terms.
|
|
|
|
|
|
|
|
import
|
2023-11-08 12:18:32 +00:00
|
|
|
std/[os, strformat, strutils],
|
2022-08-24 13:44:18 +00:00
|
|
|
eth/common,
|
|
|
|
stew/byteutils,
|
2023-02-15 10:14:40 +00:00
|
|
|
../../nimbus/sync/[protocol, snap/range_desc],
|
2022-08-24 13:44:18 +00:00
|
|
|
./gunzip
|
|
|
|
|
2023-11-08 12:18:32 +00:00
|
|
|
import
|
|
|
|
nimcrypto/utils except toHex
|
|
|
|
|
2022-08-24 13:44:18 +00:00
|
|
|
type
|
|
|
|
UndumpState = enum
|
|
|
|
UndumpHeader
|
|
|
|
UndumpStateRoot
|
|
|
|
UndumpBase
|
2022-09-02 18:16:09 +00:00
|
|
|
UndumpAccountList
|
2022-08-24 13:44:18 +00:00
|
|
|
UndumpProofs
|
|
|
|
UndumpCommit
|
|
|
|
UndumpError
|
2022-09-02 18:16:09 +00:00
|
|
|
UndumpSkipUntilCommit
|
2022-08-24 13:44:18 +00:00
|
|
|
|
2022-09-02 18:16:09 +00:00
|
|
|
UndumpAccounts* = object
|
2022-08-24 13:44:18 +00:00
|
|
|
## Palatable output for iterator
|
|
|
|
root*: Hash256
|
|
|
|
base*: NodeTag
|
|
|
|
data*: PackedAccountRange
|
2022-09-02 18:16:09 +00:00
|
|
|
seenAccounts*: int
|
|
|
|
seenStorages*: int
|
2022-08-24 13:44:18 +00:00
|
|
|
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
# Private helpers
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
|
|
|
|
template say(args: varargs[untyped]) =
|
|
|
|
# echo args
|
|
|
|
discard
|
|
|
|
|
|
|
|
proc toByteSeq(s: string): seq[byte] =
|
2022-09-03 18:15:35 +00:00
|
|
|
utils.fromHex(s)
|
2022-08-24 13:44:18 +00:00
|
|
|
|
|
|
|
proc fromHex(T: type Hash256; s: string): T =
|
|
|
|
result.data = ByteArray32.fromHex(s)
|
|
|
|
|
Prep for full sync after snap make 4 (#1282)
* Re-arrange fetching storage slots in batch module
why;
Previously, fetching partial slot ranges first has a chance of
terminating the worker peer 9due to network error) while there were
many inheritable storage slots on the queue.
Now, inheritance is checked first, then full slot ranges and finally
partial ranges.
* Update logging
* Bundled node information for healing into single object `NodeSpecs`
why:
Previously, partial paths and node keys were kept in separate variables.
This approach was error prone due to copying/reassembling function
argument objects.
As all partial paths, keys, and node data types are more or less handled
as `Blob`s over the network (using Eth/6x, or Snap/1) it makes sense to
hold these `Blob`s as named field in a single object (even if not all
fields are active for the current purpose.)
* For good housekeeping, using `NodeKey` type only for account keys
why:
previously, a mixture of `NodeKey` and `Hash256` was used. Now, only
state or storage root keys use the `Hash256` type.
* Always accept latest pivot (and not a slightly older one)
why;
For testing it was tried to use a slightly older pivot state root than
available. Some anecdotal tests seemed to suggest an advantage so that
more peers are willing to serve on that older pivot. But this could not
be confirmed in subsequent tests (still anecdotal, though.)
As a side note, the distance of the latest pivot to its predecessor is
at least 128 (or whatever the constant `minPivotBlockDistance` is
assigned to.)
* Reshuffle name components for some file and function names
why:
Clarifies purpose:
"storages" becomes: "storage slots"
"store" becomes: "range fetch"
* Stash away currently unused modules in sub-folder named "notused"
2022-10-27 13:49:28 +00:00
|
|
|
proc fromHex(T: type NodeKey; s: string): T =
|
|
|
|
ByteArray32.fromHex(s).T
|
|
|
|
|
2022-08-24 13:44:18 +00:00
|
|
|
proc fromHex(T: type NodeTag; s: string): T =
|
|
|
|
UInt256.fromBytesBE(ByteArray32.fromHex(s)).T
|
|
|
|
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
# Public capture
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
|
2022-09-02 18:16:09 +00:00
|
|
|
proc dumpAccounts*(
|
2022-08-24 13:44:18 +00:00
|
|
|
root: Hash256;
|
|
|
|
base: NodeTag;
|
|
|
|
data: PackedAccountRange;
|
|
|
|
): string =
|
|
|
|
## Dump accounts data in parseable Ascii text
|
|
|
|
proc ppStr(blob: Blob): string =
|
2023-11-08 12:18:32 +00:00
|
|
|
blob.toHex
|
2022-08-24 13:44:18 +00:00
|
|
|
|
2023-02-15 10:14:40 +00:00
|
|
|
proc ppStr(proof: SnapProof): string =
|
2023-03-03 20:01:59 +00:00
|
|
|
proof.to(Blob).ppStr
|
2023-02-15 10:14:40 +00:00
|
|
|
|
2022-08-24 13:44:18 +00:00
|
|
|
proc ppStr(hash: Hash256): string =
|
2023-11-08 12:18:32 +00:00
|
|
|
hash.data.toHex
|
2022-08-24 13:44:18 +00:00
|
|
|
|
Prep for full sync after snap make 4 (#1282)
* Re-arrange fetching storage slots in batch module
why;
Previously, fetching partial slot ranges first has a chance of
terminating the worker peer 9due to network error) while there were
many inheritable storage slots on the queue.
Now, inheritance is checked first, then full slot ranges and finally
partial ranges.
* Update logging
* Bundled node information for healing into single object `NodeSpecs`
why:
Previously, partial paths and node keys were kept in separate variables.
This approach was error prone due to copying/reassembling function
argument objects.
As all partial paths, keys, and node data types are more or less handled
as `Blob`s over the network (using Eth/6x, or Snap/1) it makes sense to
hold these `Blob`s as named field in a single object (even if not all
fields are active for the current purpose.)
* For good housekeeping, using `NodeKey` type only for account keys
why:
previously, a mixture of `NodeKey` and `Hash256` was used. Now, only
state or storage root keys use the `Hash256` type.
* Always accept latest pivot (and not a slightly older one)
why;
For testing it was tried to use a slightly older pivot state root than
available. Some anecdotal tests seemed to suggest an advantage so that
more peers are willing to serve on that older pivot. But this could not
be confirmed in subsequent tests (still anecdotal, though.)
As a side note, the distance of the latest pivot to its predecessor is
at least 128 (or whatever the constant `minPivotBlockDistance` is
assigned to.)
* Reshuffle name components for some file and function names
why:
Clarifies purpose:
"storages" becomes: "storage slots"
"store" becomes: "range fetch"
* Stash away currently unused modules in sub-folder named "notused"
2022-10-27 13:49:28 +00:00
|
|
|
proc ppStr(key: NodeKey): string =
|
2023-11-08 12:18:32 +00:00
|
|
|
key.ByteArray32.toHex
|
Prep for full sync after snap make 4 (#1282)
* Re-arrange fetching storage slots in batch module
why;
Previously, fetching partial slot ranges first has a chance of
terminating the worker peer 9due to network error) while there were
many inheritable storage slots on the queue.
Now, inheritance is checked first, then full slot ranges and finally
partial ranges.
* Update logging
* Bundled node information for healing into single object `NodeSpecs`
why:
Previously, partial paths and node keys were kept in separate variables.
This approach was error prone due to copying/reassembling function
argument objects.
As all partial paths, keys, and node data types are more or less handled
as `Blob`s over the network (using Eth/6x, or Snap/1) it makes sense to
hold these `Blob`s as named field in a single object (even if not all
fields are active for the current purpose.)
* For good housekeeping, using `NodeKey` type only for account keys
why:
previously, a mixture of `NodeKey` and `Hash256` was used. Now, only
state or storage root keys use the `Hash256` type.
* Always accept latest pivot (and not a slightly older one)
why;
For testing it was tried to use a slightly older pivot state root than
available. Some anecdotal tests seemed to suggest an advantage so that
more peers are willing to serve on that older pivot. But this could not
be confirmed in subsequent tests (still anecdotal, though.)
As a side note, the distance of the latest pivot to its predecessor is
at least 128 (or whatever the constant `minPivotBlockDistance` is
assigned to.)
* Reshuffle name components for some file and function names
why:
Clarifies purpose:
"storages" becomes: "storage slots"
"store" becomes: "range fetch"
* Stash away currently unused modules in sub-folder named "notused"
2022-10-27 13:49:28 +00:00
|
|
|
|
2022-08-24 13:44:18 +00:00
|
|
|
result = "accounts " & $data.accounts.len & " " & $data.proof.len & "\n"
|
|
|
|
|
|
|
|
result &= root.ppStr & "\n"
|
|
|
|
result &= base.to(Hash256).ppStr & "\n"
|
|
|
|
|
|
|
|
for n in 0 ..< data.accounts.len:
|
Prep for full sync after snap make 4 (#1282)
* Re-arrange fetching storage slots in batch module
why;
Previously, fetching partial slot ranges first has a chance of
terminating the worker peer 9due to network error) while there were
many inheritable storage slots on the queue.
Now, inheritance is checked first, then full slot ranges and finally
partial ranges.
* Update logging
* Bundled node information for healing into single object `NodeSpecs`
why:
Previously, partial paths and node keys were kept in separate variables.
This approach was error prone due to copying/reassembling function
argument objects.
As all partial paths, keys, and node data types are more or less handled
as `Blob`s over the network (using Eth/6x, or Snap/1) it makes sense to
hold these `Blob`s as named field in a single object (even if not all
fields are active for the current purpose.)
* For good housekeeping, using `NodeKey` type only for account keys
why:
previously, a mixture of `NodeKey` and `Hash256` was used. Now, only
state or storage root keys use the `Hash256` type.
* Always accept latest pivot (and not a slightly older one)
why;
For testing it was tried to use a slightly older pivot state root than
available. Some anecdotal tests seemed to suggest an advantage so that
more peers are willing to serve on that older pivot. But this could not
be confirmed in subsequent tests (still anecdotal, though.)
As a side note, the distance of the latest pivot to its predecessor is
at least 128 (or whatever the constant `minPivotBlockDistance` is
assigned to.)
* Reshuffle name components for some file and function names
why:
Clarifies purpose:
"storages" becomes: "storage slots"
"store" becomes: "range fetch"
* Stash away currently unused modules in sub-folder named "notused"
2022-10-27 13:49:28 +00:00
|
|
|
result &= data.accounts[n].accKey.ppStr & " "
|
2022-08-24 13:44:18 +00:00
|
|
|
result &= data.accounts[n].accBlob.ppStr & "\n"
|
|
|
|
|
2022-09-02 18:16:09 +00:00
|
|
|
if 0 < data.proof.len:
|
|
|
|
result &= "# ----\n"
|
|
|
|
for n in 0 ..< data.proof.len:
|
|
|
|
result &= data.proof[n].ppStr & "\n"
|
2022-08-24 13:44:18 +00:00
|
|
|
|
|
|
|
result &= "commit\n"
|
|
|
|
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
# Public undump
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
|
2022-09-02 18:16:09 +00:00
|
|
|
iterator undumpNextAccount*(gzFile: string): UndumpAccounts =
|
2022-08-24 13:44:18 +00:00
|
|
|
var
|
|
|
|
state = UndumpHeader
|
2022-09-02 18:16:09 +00:00
|
|
|
data: UndumpAccounts
|
2022-08-24 13:44:18 +00:00
|
|
|
nAccounts = 0u
|
|
|
|
nProofs = 0u
|
2022-09-02 18:16:09 +00:00
|
|
|
seenAccounts = 0
|
|
|
|
seenStorages = 0
|
2022-08-24 13:44:18 +00:00
|
|
|
|
|
|
|
if not gzFile.fileExists:
|
|
|
|
raiseAssert &"No such file: \"{gzFile}\""
|
|
|
|
|
|
|
|
for lno,line in gzFile.gunzipLines:
|
|
|
|
if line.len == 0 or line[0] == '#':
|
|
|
|
continue
|
|
|
|
var flds = line.split
|
|
|
|
#echo ">>> ",
|
|
|
|
# " lno=", lno,
|
|
|
|
# " state=", state,
|
|
|
|
# " nAccounts=", nAccounts,
|
|
|
|
# " nProofs=", nProofs,
|
|
|
|
# " flds=", flds
|
|
|
|
|
|
|
|
case state:
|
2022-09-02 18:16:09 +00:00
|
|
|
of UndumpSkipUntilCommit:
|
|
|
|
if flds.len == 1 and flds[0] == "commit":
|
|
|
|
state = UndumpHeader
|
|
|
|
|
2022-08-24 13:44:18 +00:00
|
|
|
of UndumpHeader, UndumpError:
|
|
|
|
if flds.len == 3 and flds[0] == "accounts":
|
|
|
|
nAccounts = flds[1].parseUInt
|
|
|
|
nProofs = flds[2].parseUInt
|
|
|
|
data.reset
|
|
|
|
state = UndumpStateRoot
|
2022-09-02 18:16:09 +00:00
|
|
|
seenAccounts.inc
|
|
|
|
continue
|
|
|
|
if 1 < flds.len and flds[0] == "storages":
|
|
|
|
seenStorages.inc
|
|
|
|
state = UndumpSkipUntilCommit
|
2022-08-24 13:44:18 +00:00
|
|
|
continue
|
|
|
|
if state != UndumpError:
|
|
|
|
state = UndumpError
|
|
|
|
say &"*** line {lno}: expected header, got {line}"
|
|
|
|
|
|
|
|
of UndumpStateRoot:
|
|
|
|
if flds.len == 1:
|
|
|
|
data.root = Hash256.fromHex(flds[0])
|
|
|
|
state = UndumpBase
|
|
|
|
continue
|
|
|
|
state = UndumpError
|
|
|
|
say &"*** line {lno}: expected state root, got {line}"
|
|
|
|
|
|
|
|
of UndumpBase:
|
|
|
|
if flds.len == 1:
|
|
|
|
data.base = NodeTag.fromHex(flds[0])
|
2022-09-02 18:16:09 +00:00
|
|
|
if 0 < nAccounts:
|
|
|
|
state = UndumpAccountList
|
|
|
|
continue
|
|
|
|
if 0 < nProofs:
|
|
|
|
state = UndumpProofs
|
|
|
|
continue
|
|
|
|
state = UndumpCommit
|
2022-08-24 13:44:18 +00:00
|
|
|
continue
|
|
|
|
state = UndumpError
|
|
|
|
say &"*** line {lno}: expected account base, got {line}"
|
|
|
|
|
2022-09-02 18:16:09 +00:00
|
|
|
of UndumpAccountList:
|
2022-08-24 13:44:18 +00:00
|
|
|
if flds.len == 2:
|
|
|
|
data.data.accounts.add PackedAccount(
|
Prep for full sync after snap make 4 (#1282)
* Re-arrange fetching storage slots in batch module
why;
Previously, fetching partial slot ranges first has a chance of
terminating the worker peer 9due to network error) while there were
many inheritable storage slots on the queue.
Now, inheritance is checked first, then full slot ranges and finally
partial ranges.
* Update logging
* Bundled node information for healing into single object `NodeSpecs`
why:
Previously, partial paths and node keys were kept in separate variables.
This approach was error prone due to copying/reassembling function
argument objects.
As all partial paths, keys, and node data types are more or less handled
as `Blob`s over the network (using Eth/6x, or Snap/1) it makes sense to
hold these `Blob`s as named field in a single object (even if not all
fields are active for the current purpose.)
* For good housekeeping, using `NodeKey` type only for account keys
why:
previously, a mixture of `NodeKey` and `Hash256` was used. Now, only
state or storage root keys use the `Hash256` type.
* Always accept latest pivot (and not a slightly older one)
why;
For testing it was tried to use a slightly older pivot state root than
available. Some anecdotal tests seemed to suggest an advantage so that
more peers are willing to serve on that older pivot. But this could not
be confirmed in subsequent tests (still anecdotal, though.)
As a side note, the distance of the latest pivot to its predecessor is
at least 128 (or whatever the constant `minPivotBlockDistance` is
assigned to.)
* Reshuffle name components for some file and function names
why:
Clarifies purpose:
"storages" becomes: "storage slots"
"store" becomes: "range fetch"
* Stash away currently unused modules in sub-folder named "notused"
2022-10-27 13:49:28 +00:00
|
|
|
accKey: NodeKey.fromHex(flds[0]),
|
2022-08-24 13:44:18 +00:00
|
|
|
accBlob: flds[1].toByteSeq)
|
|
|
|
nAccounts.dec
|
2022-09-02 18:16:09 +00:00
|
|
|
if 0 < nAccounts:
|
|
|
|
continue
|
|
|
|
if 0 < nProofs:
|
2022-08-24 13:44:18 +00:00
|
|
|
state = UndumpProofs
|
2022-09-02 18:16:09 +00:00
|
|
|
continue
|
|
|
|
state = UndumpCommit
|
2022-08-24 13:44:18 +00:00
|
|
|
continue
|
|
|
|
state = UndumpError
|
|
|
|
say &"*** line {lno}: expected account data, got {line}"
|
|
|
|
|
|
|
|
of UndumpProofs:
|
|
|
|
if flds.len == 1:
|
2023-03-03 20:01:59 +00:00
|
|
|
data.data.proof.add flds[0].toByteSeq.to(SnapProof)
|
2022-08-24 13:44:18 +00:00
|
|
|
nProofs.dec
|
|
|
|
if nProofs <= 0:
|
|
|
|
state = UndumpCommit
|
|
|
|
continue
|
|
|
|
state = UndumpError
|
|
|
|
say &"*** expected proof data, got {line}"
|
|
|
|
|
|
|
|
of UndumpCommit:
|
|
|
|
if flds.len == 1 and flds[0] == "commit":
|
2022-09-02 18:16:09 +00:00
|
|
|
data.seenAccounts = seenAccounts
|
|
|
|
data.seenStorages = seenStorages
|
2022-08-24 13:44:18 +00:00
|
|
|
yield data
|
|
|
|
state = UndumpHeader
|
|
|
|
continue
|
|
|
|
state = UndumpError
|
|
|
|
say &"*** line {lno}: expected commit, got {line}"
|
|
|
|
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
# End
|
|
|
|
# ------------------------------------------------------------------------------
|