Optimize string accumulation in ExtractTopology to avoid intermediate allocations

Co-authored-by: xkilldash9x <223238109+xkilldash9x@users.noreply.github.com>

Author: google-labs-jules[bot]

pkg/analysis/topology/topology.go

@@ -253,16 +253,7 @@ func ExtractTopology(fn *ssa.Function) *FunctionTopology {
 		return t.StringLiterals[i] < t.StringLiterals[j]
 	})
 
-	totalSize := 0
-	for _, s := range t.StringLiterals {
-		// With StringVal, we no longer need to strip quotes manually, but we keep logic simple
-		totalSize += len(s)
-	}
-
-	dataAccumulator := make([]byte, 0, totalSize)
-	for _, s := range t.StringLiterals {
-		dataAccumulator = append(dataAccumulator, []byte(s)...)
-	}
+	dataAccumulator := flattenStringLiterals(t.StringLiterals)
 
 	if len(dataAccumulator) > 0 {
 		t.EntropyScore = CalculateEntropy(dataAccumulator)
@@ -571,3 +562,17 @@ func TopologyFingerprint(t *FunctionTopology) string {
 
 	return fmt.Sprintf("L%dB%dI%d[%s]", t.LoopCount, t.BranchCount, t.InstrCount, callStr)
 }
+
+func flattenStringLiterals(literals []string) []byte {
+	totalSize := 0
+	for _, s := range literals {
+		// With StringVal, we no longer need to strip quotes manually, but we keep logic simple
+		totalSize += len(s)
+	}
+
+	dataAccumulator := make([]byte, 0, totalSize)
+	for _, s := range literals {
+		dataAccumulator = append(dataAccumulator, s...)
+	}
+	return dataAccumulator
+}

pkg/analysis/topology/topology_internal_test.go

@@ -0,0 +1,43 @@
+package topology
+
+import (
+	"bytes"
+	"testing"
+)
+
+func TestFlattenStringLiterals(t *testing.T) {
+	literals := []string{
+		"hello",
+		"world",
+		"",
+		"foo",
+		"bar",
+		"baz",
+	}
+	expected := []byte("helloworldfoobarbaz")
+
+	result := flattenStringLiterals(literals)
+
+	if !bytes.Equal(result, expected) {
+		t.Errorf("expected %q, got %q", expected, result)
+	}
+}
+
+func BenchmarkFlattenStringLiterals(b *testing.B) {
+	baseLiterals := []string{
+		"some string",
+		"another string",
+		"yet another string",
+		"long string data here to make it worthwile",
+		"short",
+	}
+	var literals []string
+	for i := 0; i < 2000; i++ {
+		literals = append(literals, baseLiterals...)
+	}
+
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		_ = flattenStringLiterals(literals)
+	}
+}