Merge pull request #154 from saljam/arabic

add arabic light stemmer

analysis/language/ar/stemmer_ar.go

@@ -0,0 +1,80 @@
+//  Copyright (c) 2014 Couchbase, Inc.
+//  Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file
+//  except in compliance with the License. You may obtain a copy of the License at
+//    http://www.apache.org/licenses/LICENSE-2.0
+//  Unless required by applicable law or agreed to in writing, software distributed under the
+//  License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,
+//  either express or implied. See the License for the specific language governing permissions
+//  and limitations under the License.
+
+package ar
+
+import (
+	"bytes"
+
+	"github.com/blevesearch/bleve/analysis"
+	"github.com/blevesearch/bleve/registry"
+)
+
+const StemmerName = "stemmer_ar"
+
+// These were obtained from org.apache.lucene.analysis.ar.ArabicStemmer
+var prefixes = [][]byte{
+	[]byte("ال"),
+	[]byte("وال"),
+	[]byte("بال"),
+	[]byte("كال"),
+	[]byte("فال"),
+	[]byte("لل"),
+	[]byte("و"),
+}
+var suffixes = [][]byte{
+	[]byte("ه"),
+	[]byte("ة"),
+	[]byte("يه"),
+	[]byte("ية"),
+	[]byte("ها"),
+	[]byte("ي"),
+	[]byte("ان"),
+	[]byte("ين"),
+	[]byte("ون"),
+	[]byte("ات"),
+}
+
+type ArabicStemmerFilter struct{}
+
+func NewArabicStemmerFilter() *ArabicStemmerFilter {
+	return &ArabicStemmerFilter{}
+}
+
+func (s *ArabicStemmerFilter) Filter(input analysis.TokenStream) analysis.TokenStream {
+	for _, token := range input {
+		term := stem(token.Term)
+		token.Term = term
+	}
+	return input
+}
+
+func stem(input []byte) []byte {
+	for _, p := range prefixes {
+		if bytes.HasPrefix(input, p) {
+			input = input[len(p):]
+			break
+		}
+	}
+	for _, s := range suffixes {
+		if bytes.HasSuffix(input, s) {
+			input = input[:len(input)-len(s)]
+			break
+		}
+	}
+	return input
+}
+
+func StemmerFilterConstructor(config map[string]interface{}, cache *registry.Cache) (analysis.TokenFilter, error) {
+	return NewArabicStemmerFilter(), nil
+}
+
+func init() {
+	registry.RegisterTokenFilter(StemmerName, StemmerFilterConstructor)
+}

analysis/language/ar/stemmer_ar_test.go

@@ -0,0 +1,119 @@
+//  Copyright (c) 2014 Couchbase, Inc.
+//  Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file
+//  except in compliance with the License. You may obtain a copy of the License at
+//    http://www.apache.org/licenses/LICENSE-2.0
+//  Unless required by applicable law or agreed to in writing, software distributed under the
+//  License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,
+//  either express or implied. See the License for the specific language governing permissions
+//  and limitations under the License.
+
+package ar
+
+import (
+	"reflect"
+	"testing"
+
+	"github.com/blevesearch/bleve/analysis"
+)
+
+func TestArabicStemmerFilter(t *testing.T) {
+	tests := []struct {
+		input  analysis.TokenStream
+		output analysis.TokenStream
+	}{
+		{
+			input: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("سلام"),
+				},
+			},
+			output: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("سلام"),
+				},
+			},
+		},
+		{
+			input: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("السلام"),
+				},
+			},
+			output: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("سلام"),
+				},
+			},
+		},
+		{
+			input: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("سلامة"),
+				},
+			},
+			output: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("سلام"),
+				},
+			},
+		},
+		{
+			input: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("السلامة"),
+				},
+			},
+			output: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("سلام"),
+				},
+			},
+		},
+		{
+			input: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("الوصل"),
+				},
+			},
+			output: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("وصل"),
+				},
+			},
+		},
+		{
+			input: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("والصل"),
+				},
+			},
+			output: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte("صل"),
+				},
+			},
+		},
+		// empty
+		{
+			input: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte(""),
+				},
+			},
+			output: analysis.TokenStream{
+				&analysis.Token{
+					Term: []byte(""),
+				},
+			},
+		},
+	}
+
+	arabicStemmerFilter := NewArabicStemmerFilter()
+	for _, test := range tests {
+		actual := arabicStemmerFilter.Filter(test.input)
+		if !reflect.DeepEqual(actual, test.output) {
+			t.Errorf("expected %#v, got %#v", test.output, actual)
+			t.Errorf("expected % x, got % x", test.output[0].Term, actual[0].Term)
+		}
+	}
+}