Cerc refactor (#43)

* pointing to unpublished versions

* pointing to unpublished versions

* updated for bumped go-eth-state-node-iterator and other cerc refactors

README.md

@@ -2,7 +2,7 @@
 
 [![Go Report Card](https://goreportcard.com/badge/github.com/vulcanize/eth-ipfs-state-validator)](https://goreportcard.com/report/github.com/vulcanize/eth-ipfs-state-validator)
 
-> Uses [ipfs-ethdb](https://github.com/vulcanize/ipfs-ethdb/tree/master/postgres) to validate completeness of IPFS Ethereum state data
+> Uses [ipfs-ethdb](https://github.com/cerc-io/ipfs-ethdb/tree/master/postgres) to validate completeness of IPFS Ethereum state data
 
 ## Background
 

cmd/validateTrie.go

@@ -24,7 +24,7 @@ import (
 	"github.com/spf13/cobra"
 	"github.com/spf13/viper"
 
-	validator "github.com/vulcanize/eth-ipfs-state-validator/v4/pkg"
+	validator "github.com/cerc-io/eth-ipfs-state-validator/v4/pkg"
 )
 
 // validateTrieCmd represents the validateTrie command
@@ -59,14 +59,19 @@ It can operate at three levels:
 }
 
 func validateTrie() {
-	v, err := newValidator()
+	params := validator.Params{
+		Workers:        viper.GetUint("validator.workers"),
+		RecoveryFormat: viper.GetString("validator.recoveryFormat"),
+	}
+	v, err := newValidator(params)
 	if err != nil {
 		logWithCommand.Fatal(err)
 	}
 	stateRootStr := viper.GetString("validator.stateRoot")
 	storageRootStr := viper.GetString("validator.storageRoot")
 	contractAddrStr := viper.GetString("validator.address")
-	switch strings.ToLower(viper.GetString("validator.type")) {
+	traversal := strings.ToLower(viper.GetString("validator.type"))
+	switch traversal {
 	case "f", "full":
 		if stateRootStr == "" {
 			logWithCommand.Fatal("must provide a state root for full state validation")
@@ -98,26 +103,28 @@ func validateTrie() {
 			logWithCommand.Fatalf("Storage trie for contract %s and root %s not complete\r\nerr: %v", addr.String(), storageRoot.String(), err)
 		}
 		logWithCommand.Infof("Storage trie for contract %s and root %s is complete", addr.String(), storageRoot.String())
+	default:
+		logWithCommand.Fatalf("Invalid traversal level: '%s'", traversal)
 	}
 
 	stats := v.GetCacheStats()
 	logWithCommand.Debugf("groupcache stats %+v", stats)
 }
 
-func newValidator() (*validator.Validator, error) {
+func newValidator(params validator.Params) (*validator.Validator, error) {
 	ipfsPath := viper.GetString("ipfs.path")
 	if ipfsPath == "" {
 		db, err := validator.NewDB()
 		if err != nil {
 			logWithCommand.Fatal(err)
 		}
-		return validator.NewPGIPFSValidator(db), nil
+		return validator.NewPGIPFSValidator(db, params), nil
 	}
 	bs, err := validator.InitIPFSBlockService(ipfsPath)
 	if err != nil {
 		return nil, err
 	}
-	return validator.NewIPFSValidator(bs), nil
+	return validator.NewIPFSValidator(bs, params), nil
 }
 
 func init() {
@@ -128,10 +135,14 @@ func init() {
 	validateTrieCmd.PersistentFlags().String("storage-root", "", "Root of the storage trie we wish to validate; for storage validation")
 	validateTrieCmd.PersistentFlags().String("address", "", "Contract address for the storage trie we wish to validate; for storage validation")
 	validateTrieCmd.PersistentFlags().String("ipfs-path", "", "Path to IPFS repository; if provided operations move through the IPFS repo otherwise Postgres connection params are expected in the provided config")
+	validateTrieCmd.PersistentFlags().Int("workers", 4, "number of concurrent workers to use")
+	validateTrieCmd.PersistentFlags().String("recovery-format", validator.DefaultRecoveryFormat, "format pattern for recovery files")
 
 	viper.BindPFlag("validator.stateRoot", validateTrieCmd.PersistentFlags().Lookup("state-root"))
 	viper.BindPFlag("validator.type", validateTrieCmd.PersistentFlags().Lookup("type"))
 	viper.BindPFlag("validator.storageRoot", validateTrieCmd.PersistentFlags().Lookup("storage-root"))
 	viper.BindPFlag("validator.address", validateTrieCmd.PersistentFlags().Lookup("address"))
+	viper.BindPFlag("validator.workers", validateTrieCmd.PersistentFlags().Lookup("workers"))
+	viper.BindPFlag("validator.recoveryFormat", validateTrieCmd.PersistentFlags().Lookup("recovery-format"))
 	viper.BindPFlag("ipfs.path", validateTrieCmd.PersistentFlags().Lookup("ipfs-path"))
 }

go.mod

@@ -1,23 +1,239 @@
-module github.com/vulcanize/eth-ipfs-state-validator/v4
+module github.com/cerc-io/eth-ipfs-state-validator/v4
 
-go 1.15
+go 1.18
 
 require (
-	github.com/ethereum/go-ethereum v1.10.17
-	github.com/ipfs/go-blockservice v0.1.7
-	github.com/ipfs/go-cid v0.0.7
-	github.com/ipfs/go-filestore v1.0.0 //indirect
-	github.com/ipfs/go-ipfs v0.10.0
-	github.com/ipfs/go-ipfs-blockstore v1.0.1
-	github.com/ipfs/go-ipfs-ds-help v1.0.0
+	github.com/cerc-io/go-eth-state-node-iterator v1.1.7
+	github.com/cerc-io/ipfs-ethdb/v4 v4.0.8-alpha
+	github.com/ethereum/go-ethereum v1.10.23
+	github.com/ipfs/go-blockservice v0.4.0
+	github.com/ipfs/go-cid v0.2.0
+	github.com/ipfs/go-ipfs-blockstore v1.2.0
+	github.com/ipfs/go-ipfs-ds-help v1.1.0
+	github.com/ipfs/kubo v0.14.0
 	github.com/jmoiron/sqlx v1.3.5
-	github.com/lib/pq v1.10.5
+	github.com/lib/pq v1.10.6
 	github.com/mailgun/groupcache/v2 v2.3.0
-	github.com/multiformats/go-multihash v0.1.0
+	github.com/multiformats/go-multihash v0.2.0
 	github.com/onsi/ginkgo v1.16.5
 	github.com/onsi/gomega v1.19.0
-	github.com/sirupsen/logrus v1.8.1
+	github.com/sirupsen/logrus v1.9.0
 	github.com/spf13/cobra v1.4.0
 	github.com/spf13/viper v1.11.0
-	github.com/vulcanize/ipfs-ethdb/v4 v4.0.0-alpha
+	golang.org/x/sync v0.0.0-20210220032951-036812b2e83c
 )
+
+require (
+	bazil.org/fuse v0.0.0-20200117225306-7b5117fecadc // indirect
+	github.com/Stebalien/go-bitfield v0.0.1 // indirect
+	github.com/VictoriaMetrics/fastcache v1.6.0 // indirect
+	github.com/alecthomas/units v0.0.0-20210927113745-59d0afb8317a // indirect
+	github.com/benbjohnson/clock v1.3.0 // indirect
+	github.com/beorn7/perks v1.0.1 // indirect
+	github.com/blang/semver/v4 v4.0.0 // indirect
+	github.com/btcsuite/btcd v0.22.1 // indirect
+	github.com/btcsuite/btcd/btcec/v2 v2.2.0 // indirect
+	github.com/btcsuite/btcd/chaincfg/chainhash v1.0.1 // indirect
+	github.com/cenkalti/backoff v2.2.1+incompatible // indirect
+	github.com/cenkalti/backoff/v4 v4.1.3 // indirect
+	github.com/cespare/xxhash/v2 v2.1.2 // indirect
+	github.com/cheekybits/genny v1.0.0 // indirect
+	github.com/containerd/cgroups v1.0.3 // indirect
+	github.com/coreos/go-systemd/v22 v22.3.2 // indirect
+	github.com/crackcomm/go-gitignore v0.0.0-20170627025303-887ab5e44cc3 // indirect
+	github.com/cskr/pubsub v1.0.2 // indirect
+	github.com/davidlazar/go-crypto v0.0.0-20200604182044-b73af7476f6c // indirect
+	github.com/decred/dcrd/dcrec/secp256k1/v4 v4.0.1 // indirect
+	github.com/docker/go-units v0.4.0 // indirect
+	github.com/dustin/go-humanize v1.0.0 // indirect
+	github.com/elastic/gosigar v0.14.2 // indirect
+	github.com/facebookgo/atomicfile v0.0.0-20151019160806-2de1f203e7d5 // indirect
+	github.com/flynn/noise v1.0.0 // indirect
+	github.com/francoispqt/gojay v1.2.13 // indirect
+	github.com/fsnotify/fsnotify v1.5.4 // indirect
+	github.com/go-logr/logr v1.2.3 // indirect
+	github.com/go-logr/stdr v1.2.2 // indirect
+	github.com/go-ole/go-ole v1.2.6 // indirect
+	github.com/go-stack/stack v1.8.0 // indirect
+	github.com/go-task/slim-sprig v0.0.0-20210107165309-348f09dbbbc0 // indirect
+	github.com/godbus/dbus/v5 v5.1.0 // indirect
+	github.com/gogo/protobuf v1.3.2 // indirect
+	github.com/golang/protobuf v1.5.2 // indirect
+	github.com/golang/snappy v0.0.4 // indirect
+	github.com/google/gopacket v1.1.19 // indirect
+	github.com/google/uuid v1.3.0 // indirect
+	github.com/gorilla/websocket v1.5.0 // indirect
+	github.com/hannahhoward/go-pubsub v0.0.0-20200423002714-8d62886cc36e // indirect
+	github.com/hashicorp/errwrap v1.1.0 // indirect
+	github.com/hashicorp/go-multierror v1.1.1 // indirect
+	github.com/hashicorp/golang-lru v0.5.5-0.20210104140557-80c98217689d // indirect
+	github.com/hashicorp/hcl v1.0.0 // indirect
+	github.com/holiman/bloomfilter/v2 v2.0.3 // indirect
+	github.com/huin/goupnp v1.0.3 // indirect
+	github.com/inconshreveable/mousetrap v1.0.0 // indirect
+	github.com/ipfs/bbloom v0.0.4 // indirect
+	github.com/ipfs/go-bitfield v1.0.0 // indirect
+	github.com/ipfs/go-bitswap v0.8.0 // indirect
+	github.com/ipfs/go-block-format v0.0.3 // indirect
+	github.com/ipfs/go-cidutil v0.1.0 // indirect
+	github.com/ipfs/go-datastore v0.5.1 // indirect
+	github.com/ipfs/go-delegated-routing v0.3.0 // indirect
+	github.com/ipfs/go-ds-measure v0.2.0 // indirect
+	github.com/ipfs/go-fetcher v1.6.1 // indirect
+	github.com/ipfs/go-filestore v1.2.0 //indirect
+	github.com/ipfs/go-fs-lock v0.0.7 // indirect
+	github.com/ipfs/go-graphsync v0.13.1 // indirect
+	github.com/ipfs/go-ipfs-chunker v0.0.5 // indirect
+	github.com/ipfs/go-ipfs-delay v0.0.1 // indirect
+	github.com/ipfs/go-ipfs-exchange-interface v0.2.0 // indirect
+	github.com/ipfs/go-ipfs-exchange-offline v0.3.0 // indirect
+	github.com/ipfs/go-ipfs-files v0.1.1 // indirect
+	github.com/ipfs/go-ipfs-keystore v0.0.2 // indirect
+	github.com/ipfs/go-ipfs-pinner v0.2.1 // indirect
+	github.com/ipfs/go-ipfs-posinfo v0.0.1 // indirect
+	github.com/ipfs/go-ipfs-pq v0.0.2 // indirect
+	github.com/ipfs/go-ipfs-provider v0.7.1 // indirect
+	github.com/ipfs/go-ipfs-routing v0.2.1 // indirect
+	github.com/ipfs/go-ipfs-util v0.0.2 // indirect
+	github.com/ipfs/go-ipld-cbor v0.0.5 // indirect
+	github.com/ipfs/go-ipld-format v0.4.0 // indirect
+	github.com/ipfs/go-ipld-legacy v0.1.1 // indirect
+	github.com/ipfs/go-ipns v0.1.2 // indirect
+	github.com/ipfs/go-log v1.0.5 // indirect
+	github.com/ipfs/go-log/v2 v2.5.1 // indirect
+	github.com/ipfs/go-merkledag v0.6.0 // indirect
+	github.com/ipfs/go-metrics-interface v0.0.1 // indirect
+	github.com/ipfs/go-mfs v0.2.1 // indirect
+	github.com/ipfs/go-namesys v0.5.0 // indirect
+	github.com/ipfs/go-path v0.3.0 // indirect
+	github.com/ipfs/go-peertaskqueue v0.7.1 // indirect
+	github.com/ipfs/go-unixfs v0.4.0 // indirect
+	github.com/ipfs/go-unixfsnode v1.4.0 // indirect
+	github.com/ipfs/go-verifcid v0.0.1 // indirect
+	github.com/ipfs/interface-go-ipfs-core v0.7.0 // indirect
+	github.com/ipld/edelweiss v0.1.4 // indirect
+	github.com/ipld/go-codec-dagpb v1.4.0 // indirect
+	github.com/ipld/go-ipld-prime v0.17.0 // indirect
+	github.com/jackpal/go-nat-pmp v1.0.2 // indirect
+	github.com/jbenet/go-temp-err-catcher v0.1.0 // indirect
+	github.com/jbenet/goprocess v0.1.4 // indirect
+	github.com/klauspost/compress v1.15.1 // indirect
+	github.com/klauspost/cpuid/v2 v2.0.12 // indirect
+	github.com/koron/go-ssdp v0.0.2 // indirect
+	github.com/libp2p/go-buffer-pool v0.0.2 // indirect
+	github.com/libp2p/go-cidranger v1.1.0 // indirect
+	github.com/libp2p/go-doh-resolver v0.4.0 // indirect
+	github.com/libp2p/go-eventbus v0.2.1 // indirect
+	github.com/libp2p/go-flow-metrics v0.0.3 // indirect
+	github.com/libp2p/go-libp2p v0.20.3 // indirect
+	github.com/libp2p/go-libp2p-asn-util v0.2.0 // indirect
+	github.com/libp2p/go-libp2p-core v0.16.1 // indirect
+	github.com/libp2p/go-libp2p-discovery v0.7.0 // indirect
+	github.com/libp2p/go-libp2p-kad-dht v0.16.0 // indirect
+	github.com/libp2p/go-libp2p-kbucket v0.4.7 // indirect
+	github.com/libp2p/go-libp2p-loggables v0.1.0 // indirect
+	github.com/libp2p/go-libp2p-peerstore v0.6.0 // indirect
+	github.com/libp2p/go-libp2p-pubsub v0.6.1 // indirect
+	github.com/libp2p/go-libp2p-pubsub-router v0.5.0 // indirect
+	github.com/libp2p/go-libp2p-record v0.1.3 // indirect
+	github.com/libp2p/go-libp2p-resource-manager v0.3.0 // indirect
+	github.com/libp2p/go-libp2p-routing-helpers v0.2.3 // indirect
+	github.com/libp2p/go-libp2p-swarm v0.11.0 // indirect
+	github.com/libp2p/go-libp2p-xor v0.1.0 // indirect
+	github.com/libp2p/go-mplex v0.7.0 // indirect
+	github.com/libp2p/go-msgio v0.2.0 // indirect
+	github.com/libp2p/go-nat v0.1.0 // indirect
+	github.com/libp2p/go-netroute v0.2.0 // indirect
+	github.com/libp2p/go-openssl v0.0.7 // indirect
+	github.com/libp2p/go-reuseport v0.2.0 // indirect
+	github.com/libp2p/go-yamux/v3 v3.1.2 // indirect
+	github.com/libp2p/zeroconf/v2 v2.1.1 // indirect
+	github.com/lucas-clemente/quic-go v0.27.1 // indirect
+	github.com/magiconair/properties v1.8.6 // indirect
+	github.com/marten-seemann/qtls-go1-16 v0.1.5 // indirect
+	github.com/marten-seemann/qtls-go1-17 v0.1.1 // indirect
+	github.com/marten-seemann/qtls-go1-18 v0.1.1 // indirect
+	github.com/marten-seemann/tcp v0.0.0-20210406111302-dfbc87cc63fd // indirect
+	github.com/mattn/go-isatty v0.0.14 // indirect
+	github.com/mattn/go-runewidth v0.0.9 // indirect
+	github.com/matttproud/golang_protobuf_extensions v1.0.1 // indirect
+	github.com/miekg/dns v1.1.48 // indirect
+	github.com/mikioh/tcpinfo v0.0.0-20190314235526-30a79bb1804b // indirect
+	github.com/mikioh/tcpopt v0.0.0-20190314235656-172688c1accc // indirect
+	github.com/minio/sha256-simd v1.0.0 // indirect
+	github.com/mitchellh/go-homedir v1.1.0 // indirect
+	github.com/mitchellh/mapstructure v1.4.3 // indirect
+	github.com/mr-tron/base58 v1.2.0 // indirect
+	github.com/multiformats/go-base32 v0.0.4 // indirect
+	github.com/multiformats/go-base36 v0.1.0 // indirect
+	github.com/multiformats/go-multiaddr v0.5.0 // indirect
+	github.com/multiformats/go-multiaddr-dns v0.3.1 // indirect
+	github.com/multiformats/go-multiaddr-fmt v0.1.0 // indirect
+	github.com/multiformats/go-multibase v0.1.0 // indirect
+	github.com/multiformats/go-multicodec v0.5.0 // indirect
+	github.com/multiformats/go-multistream v0.3.3 // indirect
+	github.com/multiformats/go-varint v0.0.6 // indirect
+	github.com/nxadm/tail v1.4.8 // indirect
+	github.com/olekukonko/tablewriter v0.0.5 // indirect
+	github.com/opencontainers/runtime-spec v1.0.2 // indirect
+	github.com/opentracing/opentracing-go v1.2.0 // indirect
+	github.com/pbnjay/memory v0.0.0-20210728143218-7b4eea64cf58 // indirect
+	github.com/pelletier/go-toml v1.9.4 // indirect
+	github.com/pelletier/go-toml/v2 v2.0.0-beta.8 // indirect
+	github.com/pkg/errors v0.9.1 // indirect
+	github.com/polydawn/refmt v0.0.0-20201211092308-30ac6d18308e // indirect
+	github.com/prometheus/client_golang v1.12.1 // indirect
+	github.com/prometheus/client_model v0.2.0 // indirect
+	github.com/prometheus/common v0.33.0 // indirect
+	github.com/prometheus/procfs v0.7.3 // indirect
+	github.com/prometheus/tsdb v0.10.0 // indirect
+	github.com/raulk/clock v1.1.0 // indirect
+	github.com/raulk/go-watchdog v1.2.0 // indirect
+	github.com/segmentio/fasthash v1.0.3 // indirect
+	github.com/shirou/gopsutil v3.21.11+incompatible // indirect
+	github.com/spacemonkeygo/spacelog v0.0.0-20180420211403-2296661a0572 // indirect
+	github.com/spaolacci/murmur3 v1.1.0 // indirect
+	github.com/spf13/afero v1.8.2 // indirect
+	github.com/spf13/cast v1.4.1 // indirect
+	github.com/spf13/jwalterweatherman v1.1.0 // indirect
+	github.com/spf13/pflag v1.0.5 // indirect
+	github.com/subosito/gotenv v1.2.0 // indirect
+	github.com/syndtr/goleveldb v1.0.1-0.20220614013038-64ee5596c38a // indirect
+	github.com/tidwall/gjson v1.14.0 // indirect
+	github.com/tidwall/match v1.1.1 // indirect
+	github.com/tidwall/pretty v1.2.0 // indirect
+	github.com/tklauser/go-sysconf v0.3.5 // indirect
+	github.com/tklauser/numcpus v0.2.2 // indirect
+	github.com/wI2L/jsondiff v0.2.0 // indirect
+	github.com/whyrusleeping/base32 v0.0.0-20170828182744-c30ac30633cc // indirect
+	github.com/whyrusleeping/cbor-gen v0.0.0-20210219115102-f37d292932f2 // indirect
+	github.com/whyrusleeping/chunker v0.0.0-20181014151217-fe64bd25879f // indirect
+	github.com/whyrusleeping/go-keyspace v0.0.0-20160322163242-5b898ac5add1 // indirect
+	github.com/whyrusleeping/multiaddr-filter v0.0.0-20160516205228-e903e4adabd7 // indirect
+	github.com/whyrusleeping/timecache v0.0.0-20160911033111-cfcb2f1abfee // indirect
+	github.com/yusufpapurcu/wmi v1.2.2 // indirect
+	go.opencensus.io v0.23.0 // indirect
+	go.opentelemetry.io/otel v1.7.0 // indirect
+	go.opentelemetry.io/otel/trace v1.7.0 // indirect
+	go.uber.org/atomic v1.9.0 // indirect
+	go.uber.org/dig v1.14.0 // indirect
+	go.uber.org/fx v1.16.0 // indirect
+	go.uber.org/multierr v1.8.0 // indirect
+	go.uber.org/zap v1.21.0 // indirect
+	go4.org v0.0.0-20200411211856-f5505b9728dd // indirect
+	golang.org/x/crypto v0.0.0-20220829220503-c86fa9a7ed90 // indirect
+	golang.org/x/mod v0.6.0-dev.0.20220106191415-9b9b3d81d5e3 // indirect
+	golang.org/x/net v0.0.0-20220607020251-c690dde0001d // indirect
+	golang.org/x/sys v0.0.0-20220715151400-c0bba94af5f8 // indirect
+	golang.org/x/text v0.3.7 // indirect
+	golang.org/x/tools v0.1.10 // indirect
+	golang.org/x/xerrors v0.0.0-20220517211312-f3a8303e98df // indirect
+	google.golang.org/protobuf v1.28.0 // indirect
+	gopkg.in/ini.v1 v1.66.4 // indirect
+	gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7 // indirect
+	gopkg.in/yaml.v2 v2.4.0 // indirect
+	gopkg.in/yaml.v3 v3.0.1 // indirect
+	lukechampine.com/blake3 v1.1.7 // indirect
+)
+
+replace github.com/ethereum/go-ethereum v1.10.23 => github.com/cerc-io/go-ethereum v1.10.23-statediff-4.2.0-alpha

main.go

@@ -18,7 +18,7 @@ package main
 import (
 	"github.com/sirupsen/logrus"
 
-	"github.com/vulcanize/eth-ipfs-state-validator/v4/cmd"
+	"github.com/cerc-io/eth-ipfs-state-validator/v4/cmd"
 )
 
 func main() {

pkg/types.go

@@ -0,0 +1,9 @@
+package validator
+
+type TraversalType = string
+
+const (
+	fullTraversal    = "full"
+	stateTraversal   = "state"
+	storageTraversal = "storage"
+)

pkg/util.go

@@ -23,8 +23,8 @@ import (
 	"github.com/ipfs/go-cid"
 	blockstore "github.com/ipfs/go-ipfs-blockstore"
 	dshelp "github.com/ipfs/go-ipfs-ds-help"
-	"github.com/ipfs/go-ipfs/core"
-	"github.com/ipfs/go-ipfs/repo/fsrepo"
+	"github.com/ipfs/kubo/core"
+	"github.com/ipfs/kubo/repo/fsrepo"
 	"github.com/jmoiron/sqlx"
 )
 

pkg/validator.go

@@ -17,19 +17,28 @@
 package validator
 
 import (
+	"bytes"
+	"context"
+	"fmt"
 	"time"
 
 	"github.com/ethereum/go-ethereum/common"
 	"github.com/ethereum/go-ethereum/core/state"
+	"github.com/ethereum/go-ethereum/core/types"
 	"github.com/ethereum/go-ethereum/crypto"
 	"github.com/ethereum/go-ethereum/ethdb"
+	"github.com/ethereum/go-ethereum/rlp"
 	"github.com/ethereum/go-ethereum/trie"
 	"github.com/ipfs/go-blockservice"
 	"github.com/jmoiron/sqlx"
 	"github.com/mailgun/groupcache/v2"
+	log "github.com/sirupsen/logrus"
+	"golang.org/x/sync/errgroup"
 
-	ipfsethdb "github.com/vulcanize/ipfs-ethdb/v4"
-	pgipfsethdb "github.com/vulcanize/ipfs-ethdb/v4/postgres"
+	nodeiter "github.com/cerc-io/go-eth-state-node-iterator"
+	"github.com/cerc-io/go-eth-state-node-iterator/tracker"
+	ipfsethdb "github.com/cerc-io/ipfs-ethdb/v4"
+	pgipfsethdb "github.com/cerc-io/ipfs-ethdb/v4/postgres"
 )
 
 // Validator is used for validating Ethereum state and storage tries on PG-IPFS
@@ -38,10 +47,22 @@ type Validator struct {
 	trieDB        *trie.Database
 	stateDatabase state.Database
 	db            *pgipfsethdb.Database
+
+	params Params
 }
 
+type Params struct {
+	Workers        uint
+	RecoveryFormat string // %s substituted with traversal type
+}
+
+var (
+	DefaultRecoveryFormat = "./recover_validate_%s"
+	emptyCodeHash         = crypto.Keccak256(nil)
+)
+
 // NewPGIPFSValidator returns a new trie validator ontop of a connection pool for an IPFS backing Postgres database
-func NewPGIPFSValidator(db *sqlx.DB) *Validator {
+func NewPGIPFSValidator(db *sqlx.DB, par Params) *Validator {
 	kvs := pgipfsethdb.NewKeyValueStore(db, pgipfsethdb.CacheConfig{
 		Name:           "kv",
 		Size:           16 * 1000 * 1000, // 16MB
@@ -54,11 +75,13 @@ func NewPGIPFSValidator(db *sqlx.DB) *Validator {
 		ExpiryDuration: time.Hour * 8,    // 8 hours
 	})
 
+	normalizeParams(&par)
 	return &Validator{
 		kvs:           kvs,
 		trieDB:        trie.NewDatabase(kvs),
 		stateDatabase: state.NewDatabase(database),
 		db:            database.(*pgipfsethdb.Database),
+		params:        par,
 	}
 }
 
@@ -67,13 +90,15 @@ func (v *Validator) GetCacheStats() groupcache.Stats {
 }
 
 // NewIPFSValidator returns a new trie validator ontop of an IPFS blockservice
-func NewIPFSValidator(bs blockservice.BlockService) *Validator {
+func NewIPFSValidator(bs blockservice.BlockService, par Params) *Validator {
 	kvs := ipfsethdb.NewKeyValueStore(bs)
 	database := ipfsethdb.NewDatabase(bs)
+	normalizeParams(&par)
 	return &Validator{
 		kvs:           kvs,
 		trieDB:        trie.NewDatabase(kvs),
 		stateDatabase: state.NewDatabase(database),
+		params:        par,
 	}
 }
 
@@ -88,21 +113,25 @@ func NewValidator(kvs ethdb.KeyValueStore, database ethdb.Database) *Validator {
 	}
 }
 
+// Ensure params are valid
+func normalizeParams(p *Params) {
+	if p.Workers == 0 {
+		p.Workers = 1
+	}
+	if len(p.RecoveryFormat) == 0 {
+		p.RecoveryFormat = DefaultRecoveryFormat
+	}
+}
+
 // ValidateTrie returns an error if the state and storage tries for the provided state root cannot be confirmed as complete
 // This does consider child storage tries
 func (v *Validator) ValidateTrie(stateRoot common.Hash) error {
-	// Generate the state.NodeIterator for this root
-	stateDB, err := state.New(stateRoot, v.stateDatabase, nil)
+	t, err := v.stateDatabase.OpenTrie(stateRoot)
 	if err != nil {
 		return err
 	}
-	it := state.NewNodeIterator(stateDB)
-	for it.Next() {
-		// iterate through entire state trie and descendent storage tries
-		// it.Next() will return false when we have either completed iteration of the entire trie or have ran into an error (e.g. a missing node)
-		// if we are able to iterate through the entire trie without error then the trie is complete
-	}
-	return it.Error
+	iterate := func(it trie.NodeIterator) error { return v.iterate(it, true) }
+	return iterateTracked(t, fmt.Sprintf(v.params.RecoveryFormat, fullTraversal), v.params.Workers, iterate)
 }
 
 // ValidateStateTrie returns an error if the state trie for the provided state root cannot be confirmed as complete
@@ -113,13 +142,8 @@ func (v *Validator) ValidateStateTrie(stateRoot common.Hash) error {
 	if err != nil {
 		return err
 	}
-	it := t.NodeIterator(nil)
-	for it.Next(true) {
-		// iterate through entire state trie
-		// it.Next() will return false when we have either completed iteration of the entire trie or have ran into an error (e.g. a missing node)
-		// if we are able to iterate through the entire trie without error then the trie is complete
-	}
-	return it.Error()
+	iterate := func(it trie.NodeIterator) error { return v.iterate(it, false) }
+	return iterateTracked(t, fmt.Sprintf(v.params.RecoveryFormat, stateTraversal), v.params.Workers, iterate)
 }
 
 // ValidateStorageTrie returns an error if the storage trie for the provided storage root and contract address cannot be confirmed as complete
@@ -130,13 +154,8 @@ func (v *Validator) ValidateStorageTrie(address common.Address, storageRoot comm
 	if err != nil {
 		return err
 	}
-	it := t.NodeIterator(nil)
-	for it.Next(true) {
-		// iterate through entire storage trie
-		// it.Next() will return false when we have either completed iteration of the entire trie or have ran into an error (e.g. a missing node)
-		// if we are able to iterate through the entire trie without error then the trie is complete
-	}
-	return it.Error()
+	iterate := func(it trie.NodeIterator) error { return v.iterate(it, false) }
+	return iterateTracked(t, fmt.Sprintf(v.params.RecoveryFormat, storageTraversal), v.params.Workers, iterate)
 }
 
 // Close implements io.Closer
@@ -146,3 +165,79 @@ func (v *Validator) Close() error {
 	groupcache.DeregisterGroup("db")
 	return nil
 }
+
+// Traverses each iterator in a separate goroutine.
+// If storage = true, also traverse storage tries for each leaf.
+func (v *Validator) iterate(it trie.NodeIterator, storage bool) error {
+	// Iterate through entire state trie. it.Next() will return false when we have
+	// either completed iteration of the entire trie or run into an error (e.g. a
+	// missing node). If we are able to iterate through the entire trie without error
+	// then the trie is complete.
+	for it.Next(true) {
+		// This block adapted from geth - core/state/iterator.go
+		// If storage is not requested, or the state trie node is an internal entry, skip
+		if !storage || !it.Leaf() {
+			continue
+		}
+		// Otherwise we've reached an account node, initiate data iteration
+		var account types.StateAccount
+		if err := rlp.Decode(bytes.NewReader(it.LeafBlob()), &account); err != nil {
+			return err
+		}
+		dataTrie, err := v.stateDatabase.OpenStorageTrie(common.BytesToHash(it.LeafKey()), account.Root)
+		if err != nil {
+			return err
+		}
+		dataIt := dataTrie.NodeIterator(nil)
+		if !bytes.Equal(account.CodeHash, emptyCodeHash) {
+			addrHash := common.BytesToHash(it.LeafKey())
+			_, err := v.stateDatabase.ContractCode(addrHash, common.BytesToHash(account.CodeHash))
+			if err != nil {
+				return fmt.Errorf("code %x: %v", account.CodeHash, err)
+			}
+		}
+		for dataIt.Next(true) {
+		}
+		if dataIt.Error() != nil {
+			return dataIt.Error()
+		}
+	}
+	return it.Error()
+}
+
+func iterateTracked(tree state.Trie, recoveryFile string, iterCount uint, fn func(trie.NodeIterator) error) error {
+	ctx, cancelCtx := context.WithCancel(context.Background())
+	tracker := tracker.New(recoveryFile, iterCount)
+	tracker.CaptureSignal(cancelCtx)
+	halt := func() {
+		if err := tracker.HaltAndDump(); err != nil {
+			log.Errorf("failed to write recovery file: %v", err)
+		}
+	}
+
+	// attempt to restore from recovery file if it exists
+	iters, err := tracker.Restore(tree)
+	if err != nil {
+		return err
+	}
+	if iterCount < uint(len(iters)) {
+		return fmt.Errorf("recovered too many iterators: got %d, expected %d", len(iters), iterCount)
+	}
+
+	if iters == nil { // nothing restored
+		iters = nodeiter.SubtrieIterators(tree, iterCount)
+		for i, it := range iters {
+			iters[i] = tracker.Tracked(it, nil)
+		}
+	}
+
+	g, ctx := errgroup.WithContext(ctx)
+	defer halt()
+
+	for _, it := range iters {
+		func(it trie.NodeIterator) {
+			g.Go(func() error { return fn(it) })
+		}(it)
+	}
+	return g.Wait()
+}

pkg/validator_test.go

@@ -19,6 +19,8 @@ package validator_test
 import (
 	"fmt"
 	"math/big"
+	"os"
+	"path/filepath"
 
 	"github.com/ethereum/go-ethereum/common"
 	"github.com/ethereum/go-ethereum/core/types"
@@ -30,8 +32,8 @@ import (
 	. "github.com/onsi/ginkgo"
 	. "github.com/onsi/gomega"
 
-	validator "github.com/vulcanize/eth-ipfs-state-validator/v4/pkg"
-	pgipfsethdb "github.com/vulcanize/ipfs-ethdb/v4/postgres"
+	validator "github.com/cerc-io/eth-ipfs-state-validator/v4/pkg"
+	pgipfsethdb "github.com/cerc-io/ipfs-ethdb/v4/postgres"
 )
 
 var (
@@ -194,15 +196,20 @@ var (
 	v   *validator.Validator
 	db  *sqlx.DB
 	err error
+	tmp string
 )
 
 var _ = Describe("PG-IPFS Validator", func() {
 	BeforeEach(func() {
 		db, err = pgipfsethdb.TestDB()
 		Expect(err).ToNot(HaveOccurred())
-		v = validator.NewPGIPFSValidator(db)
+		tmp, err = os.MkdirTemp("", "test_validator")
+		Expect(err).ToNot(HaveOccurred())
+		params := validator.Params{Workers: 4, RecoveryFormat: filepath.Join(tmp, "recover_%s")}
+		v = validator.NewPGIPFSValidator(db, params)
 	})
 	AfterEach(func() {
+		os.RemoveAll(tmp)
 		v.Close()
 	})
 	Describe("ValidateTrie", func() {
@@ -210,7 +217,7 @@ var _ = Describe("PG-IPFS Validator", func() {
 			err = validator.ResetTestDB(db)
 			Expect(err).ToNot(HaveOccurred())
 		})
-		It("Returns an error the state root node is missing", func() {
+		It("Returns an error if the state root node is missing", func() {
 			// we write code to ethdb, there should probably be an EthCode IPLD codec
 			// but there isn't, and we don't need one here since blockstore keys are mh-derived
 			loadTrie(append(missingRootStateNodes, mockCode), trieStorageNodes)