improve perf of porter stemmer

1.  porter stemmer offers method to NOT do lowercasing, however
to use this we must convert to runes first ourself, so we did this

2.  now we can invoke the version that skips lowercasing, we
already do this ourselves before stemming through separate filter

due to the fact that the stemmer modifies the runes in place
we have no way to know if there were changes, thus we must
always encode back into the term byte slice

added unit test which catches the problem found

NOTE this uses analysis.BuildTermFromRunes so perf gain is
only visible with other PR also merged

future gains are possible if we udpate the stemmer to let us
know if changes were made, thus skipping re-encoding to
[]byte when no changes were actually made

analysis/token_filters/porter/porter.go

@@ -10,6 +10,8 @@
 package porter
 
 import (
+	"bytes"
+
 	"github.com/blevesearch/bleve/analysis"
 	"github.com/blevesearch/bleve/registry"
 
@@ -29,8 +31,9 @@ func (s *PorterStemmer) Filter(input analysis.TokenStream) analysis.TokenStream
 	for _, token := range input {
 		// if it is not a protected keyword, stem it
 		if !token.KeyWord {
-			stemmed := porterstemmer.StemString(string(token.Term))
-			token.Term = []byte(stemmed)
+			termRunes := bytes.Runes(token.Term)
+			stemmedRunes := porterstemmer.StemWithoutLowerCasing(termRunes)
+			token.Term = analysis.BuildTermFromRunes(stemmedRunes)
 		}
 	}
 	return input

analysis/token_filters/porter/porter_test.go

@@ -32,6 +32,16 @@ func TestPorterStemmer(t *testing.T) {
 			Term:    []byte("protected"),
 			KeyWord: true,
 		},
+		&analysis.Token{
+			Term: []byte("cat"),
+		},
+		&analysis.Token{
+			Term: []byte("done"),
+		},
+		// a term which does stem, but does not change length
+		&analysis.Token{
+			Term: []byte("marty"),
+		},
 	}
 
 	expectedTokenStream := analysis.TokenStream{
@@ -48,6 +58,15 @@ func TestPorterStemmer(t *testing.T) {
 			Term:    []byte("protected"),
 			KeyWord: true,
 		},
+		&analysis.Token{
+			Term: []byte("cat"),
+		},
+		&analysis.Token{
+			Term: []byte("done"),
+		},
+		&analysis.Token{
+			Term: []byte("marti"),
+		},
 	}
 
 	filter := NewPorterStemmer()
@@ -56,3 +75,36 @@ func TestPorterStemmer(t *testing.T) {
 		t.Errorf("expected %#v got %#v", expectedTokenStream[3], ouputTokenStream[3])
 	}
 }
+
+func BenchmarkPorterStemmer(b *testing.B) {
+
+	inputTokenStream := analysis.TokenStream{
+		&analysis.Token{
+			Term: []byte("walking"),
+		},
+		&analysis.Token{
+			Term: []byte("talked"),
+		},
+		&analysis.Token{
+			Term: []byte("business"),
+		},
+		&analysis.Token{
+			Term:    []byte("protected"),
+			KeyWord: true,
+		},
+		&analysis.Token{
+			Term: []byte("cat"),
+		},
+		&analysis.Token{
+			Term: []byte("done"),
+		},
+	}
+
+	filter := NewPorterStemmer()
+	b.ResetTimer()
+
+	for i := 0; i < b.N; i++ {
+		filter.Filter(inputTokenStream)
+	}
+
+}