From 6e7ae4a4a3a2b6d367b945d15ddbb065dae41b0d Mon Sep 17 00:00:00 2001
From: kvii <13721952202@163.com>
Date: Wed, 8 Oct 2025 10:45:22 +0800
Subject: [PATCH 1/3] feat: Add support for "UniGB-UCS2-H" encoding. #55

---
 page.go      | 12 ++++++++++++
 page_test.go | 37 +++++++++++++++++++++++++++++++++++++
 2 files changed, 49 insertions(+)
 create mode 100644 page_test.go

diff --git a/page.go b/page.go
index 31db8ab9..3f691599 100644
--- a/page.go
+++ b/page.go
@@ -6,11 +6,13 @@ package pdf
 
 import (
 	"bytes"
+	"encoding/binary"
 	"errors"
 	"fmt"
 	"io"
 	"sort"
 	"strings"
+	"unicode/utf16"
 )
 
 // A Page represent a single page in a PDF file.
@@ -210,6 +212,8 @@ func (f Font) getEncoder() TextEncoding {
 			return &byteEncoder{&macRomanEncoding}
 		case "Identity-H":
 			return f.charmapEncoding()
+		case "UniGB-UCS2-H":
+			return &ucs2Encoder{}
 		default:
 			if DebugOn {
 				println("unknown encoding", enc.Name())
@@ -287,6 +291,14 @@ func (e *nopEncoder) Decode(raw string) (text string) {
 	return raw
 }
 
+type ucs2Encoder struct{}
+
+func (e *ucs2Encoder) Decode(raw string) (text string) {
+	u16s := make([]uint16, len(raw)/2)
+	_ = binary.Read(strings.NewReader(raw), binary.BigEndian, &u16s)
+	return string(utf16.Decode(u16s))
+}
+
 type byteEncoder struct {
 	table *[256]rune
 }
diff --git a/page_test.go b/page_test.go
new file mode 100644
index 00000000..04da5c65
--- /dev/null
+++ b/page_test.go
@@ -0,0 +1,37 @@
+// Copyright 2014 The Go Authors.  All rights reserved.
+// Use of this source code is governed by a BSD-style
+// license that can be found in the LICENSE file.
+
+package pdf
+
+import (
+	"bytes"
+	"testing"
+)
+
+func TestUcs2Encoder(t *testing.T) {
+	// Extract form the raw file. 0x5c is an escape character in pdf.
+	data := []byte{
+		0x6C, 0x5F, // 江
+		0x82, 0xCF, // 苏
+		0x94, 0xF6, // 银
+		0x88, 0x4C, // 行
+		0x00, 0x5C, 0x28, // (
+		0x4E, 0xA4, // 交
+		0x66, 0x13, // 易
+		0x62, 0x63, // 扣
+		0x6B, 0x3E, // 款
+		0x56, 0xDE, // 回
+		0x53, 0x55, // 单
+		0x00, 0x5C, 0x29, // )
+	}
+	data = bytes.ReplaceAll(data, []byte{0x5C}, []byte{})
+
+	var e ucs2Encoder
+	text := e.Decode(string(data))
+
+	const want = "江苏银行(交易扣款回单)"
+	if text != want {
+		t.Errorf("got %q, want %q", text, want)
+	}
+}

From 5f555d437550afc222d98e960ea3e32a71da04f3 Mon Sep 17 00:00:00 2001
From: kvii <13721952202@163.com>
Date: Sat, 5 Sep 2026 19:00:54 +0800
Subject: [PATCH 2/3] feat: reuse utf16Decode; only test the encoder; remove
 copyright.

---
 page.go      |  6 +-----
 page_test.go | 16 +++++++---------
 2 files changed, 8 insertions(+), 14 deletions(-)

diff --git a/page.go b/page.go
index 3f691599..93cd746e 100644
--- a/page.go
+++ b/page.go
@@ -6,13 +6,11 @@ package pdf
 
 import (
 	"bytes"
-	"encoding/binary"
 	"errors"
 	"fmt"
 	"io"
 	"sort"
 	"strings"
-	"unicode/utf16"
 )
 
 // A Page represent a single page in a PDF file.
@@ -294,9 +292,7 @@ func (e *nopEncoder) Decode(raw string) (text string) {
 type ucs2Encoder struct{}
 
 func (e *ucs2Encoder) Decode(raw string) (text string) {
-	u16s := make([]uint16, len(raw)/2)
-	_ = binary.Read(strings.NewReader(raw), binary.BigEndian, &u16s)
-	return string(utf16.Decode(u16s))
+	return utf16Decode(raw)
 }
 
 type byteEncoder struct {
diff --git a/page_test.go b/page_test.go
index 04da5c65..4e87bcd9 100644
--- a/page_test.go
+++ b/page_test.go
@@ -1,31 +1,29 @@
-// Copyright 2014 The Go Authors.  All rights reserved.
-// Use of this source code is governed by a BSD-style
-// license that can be found in the LICENSE file.
-
 package pdf
 
 import (
-	"bytes"
 	"testing"
 )
 
 func TestUcs2Encoder(t *testing.T) {
-	// Extract form the raw file. 0x5c is an escape character in pdf.
+	// raw content in PDF (hexed for pretty)
+	// 6c5f82cf94f6884c005c284ea4661362636b3e56de5355005c29
+
+	// raw bytes in showEncodedText (hexed for pretty)
+	// 6c5f82cf94f6884c00284ea4661362636b3e56de53550029
 	data := []byte{
 		0x6C, 0x5F, // 江
 		0x82, 0xCF, // 苏
 		0x94, 0xF6, // 银
 		0x88, 0x4C, // 行
-		0x00, 0x5C, 0x28, // (
+		0x00, 0x28, // (
 		0x4E, 0xA4, // 交
 		0x66, 0x13, // 易
 		0x62, 0x63, // 扣
 		0x6B, 0x3E, // 款
 		0x56, 0xDE, // 回
 		0x53, 0x55, // 单
-		0x00, 0x5C, 0x29, // )
+		0x00, 0x29, // )
 	}
-	data = bytes.ReplaceAll(data, []byte{0x5C}, []byte{})
 
 	var e ucs2Encoder
 	text := e.Decode(string(data))

From 57ce244ec9daa5ad7ca55d2b9bf793e9f5c6c03e Mon Sep 17 00:00:00 2001
From: kvii <13721952202@163.com>
Date: Mon, 7 Sep 2026 20:03:27 +0800
Subject: [PATCH 3/3] feat: fix test code

---
 page_test.go | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/page_test.go b/page_test.go
index 9a569226..00c03849 100644
--- a/page_test.go
+++ b/page_test.go
@@ -1,6 +1,8 @@
 package pdf
 
 import (
+	"reflect"
+	"sort"
 	"testing"
 )
 
@@ -31,10 +33,8 @@ func TestUcs2Encoder(t *testing.T) {
 	const want = "江苏银行(交易扣款回单)"
 	if text != want {
 		t.Errorf("got %q, want %q", text, want)
-	"reflect"
-	"sort"
-	"testing"
-)
+	}
+}
 
 func TestNopEncoder(t *testing.T) {
 	e := &nopEncoder{}