From 63f3f4372d1d8ec9deb2180561ed9e9f3edf1468 Mon Sep 17 00:00:00 2001
From: Andy Balholm <andy@balholm.com>
Date: Sat, 30 Dec 2023 15:56:13 -0800
Subject: [PATCH] matchfinder.M4: add LimitedSearch option

Using LimitedSearch, it only checks for overlapping matches in one
place instead of checking at each byte.
This gains about 50% in compression speed while only losing about
2% in compression ratio.
---
 brotli_test.go    |  8 ++++++++
 matchfinder/m4.go | 11 ++++++++++-
 2 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/brotli_test.go b/brotli_test.go
index 4dd8b54..4ad417b 100644
--- a/brotli_test.go
+++ b/brotli_test.go
@@ -657,3 +657,11 @@ func TestEncodeM4(t *testing.T) {
 func BenchmarkEncodeM4(b *testing.B) {
 	benchmark(b, "testdata/Isaac.Newton-Opticks.txt", &matchfinder.M4{MaxDistance: 1 << 20}, 1<<16)
 }
+
+func TestEncodeM4Limited(t *testing.T) {
+	test(t, "testdata/Isaac.Newton-Opticks.txt", &matchfinder.M4{MaxDistance: 1 << 18, LimitedSearch: true}, 1<<16)
+}
+
+func BenchmarkEncodeM4Limited(b *testing.B) {
+	benchmark(b, "testdata/Isaac.Newton-Opticks.txt", &matchfinder.M4{MaxDistance: 1 << 20, LimitedSearch: true}, 1<<16)
+}
diff --git a/matchfinder/m4.go b/matchfinder/m4.go
index fb5c83d..d16f892 100644
--- a/matchfinder/m4.go
+++ b/matchfinder/m4.go
@@ -10,7 +10,7 @@ import (
 // interface that uses a simple hash table to find matches,
 // but the advanced parsing technique from
 // https://fastcompression.blogspot.com/2011/12/advanced-parsing-strategies.html,
-// except that it looks for matches at every input position.
+// except that it normally looks for matches at every input position.
 type M4 struct {
 	// MaxDistance is the maximum distance (in bytes) to look back for
 	// a match. The default is 65535.
@@ -28,6 +28,11 @@ type M4 struct {
 	// The default is 17 (128K entries).
 	TableBits int
 
+	// When LimitedSearch is true, it only looks for matches at certain
+	// points in the input rather than at every bite.
+	// (This makes compression faster, but hurts the compression ratio.)
+	LimitedSearch bool
+
 	table []uint32
 
 	history []byte
@@ -97,6 +102,10 @@ func (q *M4) FindMatches(dst []Match, src []byte) []Match {
 		candidate := int(q.table[h])
 		q.table[h] = uint32(i)
 
+		if q.LimitedSearch && i < matches[0].End && i != matches[0].End+2-q.HashLen {
+			continue
+		}
+
 		if candidate == 0 || i-candidate > q.MaxDistance || i-candidate == matches[0].Start-matches[0].Match {
 			continue
 		}