From d5ab3c5825f0e367724f2baa17f20915732f4b55 Mon Sep 17 00:00:00 2001
From: Will <will.scott@protocol.ai>
Date: Sat, 16 Oct 2021 16:26:42 -0700
Subject: [PATCH] forEach iterates over index in stable order (#258)

* forEach iterates over index in stable order
---
 v2/index/index.go              |  2 +-
 v2/index/indexsorted.go        |  8 +++++++-
 v2/index/mhindexsorted.go      |  8 +++++++-
 v2/index/mhindexsorted_test.go | 29 +++++++++++++++++++++++++++++
 4 files changed, 44 insertions(+), 3 deletions(-)

diff --git a/v2/index/index.go b/v2/index/index.go
index 8e447d0f..998a17a0 100644
--- a/v2/index/index.go
+++ b/v2/index/index.go
@@ -87,7 +87,7 @@ type (
 		// An index may contain multiple offsets corresponding to the same multihash, e.g. via duplicate blocks.
 		// In such cases, the given function may be called multiple times with the same multhihash but different offset.
 		//
-		// The order of calls to the given function is entirely index-specific.
+		// The order of calls to the given function is deterministic, but entirely index-specific.
 		ForEach(func(multihash.Multihash, uint64) error) error
 	}
 )
diff --git a/v2/index/indexsorted.go b/v2/index/indexsorted.go
index 6b6c5a68..86994dd8 100644
--- a/v2/index/indexsorted.go
+++ b/v2/index/indexsorted.go
@@ -235,7 +235,13 @@ func (m *multiWidthIndex) Load(items []Record) error {
 }
 
 func (m *multiWidthIndex) forEachDigest(f func(digest []byte, offset uint64) error) error {
-	for _, swi := range *m {
+	sizes := make([]uint32, 0, len(*m))
+	for k := range *m {
+		sizes = append(sizes, k)
+	}
+	sort.Slice(sizes, func(i, j int) bool { return sizes[i] < sizes[j] })
+	for _, s := range sizes {
+		swi := (*m)[s]
 		if err := swi.forEachDigest(f); err != nil {
 			return err
 		}
diff --git a/v2/index/mhindexsorted.go b/v2/index/mhindexsorted.go
index f81e3a94..e3cae3d0 100644
--- a/v2/index/mhindexsorted.go
+++ b/v2/index/mhindexsorted.go
@@ -157,7 +157,13 @@ func (m *MultihashIndexSorted) GetAll(cid cid.Cid, f func(uint64) bool) error {
 
 // ForEach calls f for every multihash and its associated offset stored by this index.
 func (m *MultihashIndexSorted) ForEach(f func(mh multihash.Multihash, offset uint64) error) error {
-	for _, mwci := range *m {
+	sizes := make([]uint64, 0, len(*m))
+	for k := range *m {
+		sizes = append(sizes, k)
+	}
+	sort.Slice(sizes, func(i, j int) bool { return sizes[i] < sizes[j] })
+	for _, s := range sizes {
+		mwci := (*m)[s]
 		if err := mwci.forEach(f); err != nil {
 			return err
 		}
diff --git a/v2/index/mhindexsorted_test.go b/v2/index/mhindexsorted_test.go
index b5ef7b89..e02ba059 100644
--- a/v2/index/mhindexsorted_test.go
+++ b/v2/index/mhindexsorted_test.go
@@ -46,6 +46,35 @@ func TestMultiWidthCodedIndex_MarshalUnmarshal(t *testing.T) {
 	requireContainsAll(t, umSubject, records)
 }
 
+func TestMultiWidthCodedIndex_StableIterate(t *testing.T) {
+	rng := rand.New(rand.NewSource(1414))
+	records := generateIndexRecords(t, multihash.SHA2_256, rng)
+	records = append(records, generateIndexRecords(t, multihash.SHA2_512, rng)...)
+	records = append(records, generateIndexRecords(t, multihash.IDENTITY, rng)...)
+
+	// Create a new mh sorted index and load randomly generated records into it.
+	subject, err := index.New(multicodec.CarMultihashIndexSorted)
+	require.NoError(t, err)
+	err = subject.Load(records)
+	require.NoError(t, err)
+
+	iterable := subject.(index.IterableIndex)
+	mh := make([]multihash.Multihash, 0, len(records))
+	require.NoError(t, iterable.ForEach(func(m multihash.Multihash, _ uint64) error {
+		mh = append(mh, m)
+		return nil
+	}))
+
+	for i := 0; i < 10; i++ {
+		candidate := make([]multihash.Multihash, 0, len(records))
+		require.NoError(t, iterable.ForEach(func(m multihash.Multihash, _ uint64) error {
+			candidate = append(candidate, m)
+			return nil
+		}))
+		require.Equal(t, mh, candidate)
+	}
+}
+
 func generateIndexRecords(t *testing.T, hasherCode uint64, rng *rand.Rand) []index.Record {
 	var records []index.Record
 	recordCount := rng.Intn(99) + 1 // Up to 100 records