1 /*
2 * Licensed to the Apache Software Foundation (ASF) under one or more
3 * contributor license agreements. See the NOTICE file distributed with
4 * this work for additional information regarding copyright ownership.
5 * The ASF licenses this file to You under the Apache License, Version 2.0
6 * (the "License"); you may not use this file except in compliance with
7 * the License. You may obtain a copy of the License at
8 *
9 * http://www.apache.org/licenses/LICENSE-2.0
10 *
11 * Unless required by applicable law or agreed to in writing, software
12 * distributed under the License is distributed on an "AS IS" BASIS,
13 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 * See the License for the specific language governing permissions and
15 * limitations under the License.
16 */
17 package org.apache.commons.text.translate;
18
19 import java.io.IOException;
20 import java.io.Writer;
21
22 /**
23 * Translates code points to their Unicode escaped value.
24 *
25 * @since 1.0
26 */
27 public class UnicodeEscaper extends CodePointTranslator {
28
29 /**
30 * Constructs a {@code UnicodeEscaper} above the specified value (exclusive).
31 *
32 * @param codePoint above which to escape
33 * @return The newly created {@code UnicodeEscaper} instance
34 */
35 public static UnicodeEscaper above(final int codePoint) {
36 return outsideOf(0, codePoint);
37 }
38 /**
39 * Constructs a {@code UnicodeEscaper} below the specified value (exclusive).
40 *
41 * @param codePoint below which to escape
42 * @return The newly created {@code UnicodeEscaper} instance
43 */
44 public static UnicodeEscaper below(final int codePoint) {
45 return outsideOf(codePoint, Integer.MAX_VALUE);
46 }
47 /**
48 * Constructs a {@code UnicodeEscaper} between the specified values (inclusive).
49 *
50 * @param codePointLow above which to escape
51 * @param codePointHigh below which to escape
52 * @return The newly created {@code UnicodeEscaper} instance
53 */
54 public static UnicodeEscaper between(final int codePointLow, final int codePointHigh) {
55 return new UnicodeEscaper(codePointLow, codePointHigh, true);
56 }
57
58 /**
59 * Constructs a {@code UnicodeEscaper} outside of the specified values (exclusive).
60 *
61 * @param codePointLow below which to escape
62 * @param codePointHigh above which to escape
63 * @return The newly created {@code UnicodeEscaper} instance
64 */
65 public static UnicodeEscaper outsideOf(final int codePointLow, final int codePointHigh) {
66 return new UnicodeEscaper(codePointLow, codePointHigh, false);
67 }
68
69 /** The lowest code point boundary. */
70 private final int below;
71
72 /** The highest code point boundary. */
73 private final int above;
74
75 /** Whether to escape between the boundaries or outside them. */
76 private final boolean between;
77
78 /**
79 * Constructs a {@code UnicodeEscaper} for all characters.
80 */
81 public UnicodeEscaper() {
82 this(0, Integer.MAX_VALUE, true);
83 }
84
85 /**
86 * Constructs a {@code UnicodeEscaper} for the specified range. This is
87 * the underlying method for the other constructors/builders. The {@code below}
88 * and {@code above} boundaries are inclusive when {@code between} is
89 * {@code true} and exclusive when it is {@code false}.
90 *
91 * @param below int value representing the lowest code point boundary
92 * @param above int value representing the highest code point boundary
93 * @param between whether to escape between the boundaries or outside them
94 */
95 protected UnicodeEscaper(final int below, final int above, final boolean between) {
96 this.below = below;
97 this.above = above;
98 this.between = between;
99 }
100
101 /**
102 * Converts the given code point to a hexadecimal string of the form {@code "\\uXXXX"}.
103 *
104 * @param codePoint
105 * a Unicode code point
106 * @return The hexadecimal string for the given code point
107 */
108 protected String toUtf16Escape(final int codePoint) {
109 return "\\u" + hex(codePoint);
110 }
111
112 /**
113 * {@inheritDoc}
114 */
115 @Override
116 public boolean translate(final int codePoint, final Writer writer) throws IOException {
117 if (between) {
118 if (codePoint < below || codePoint > above) {
119 return false;
120 }
121 } else if (codePoint >= below && codePoint <= above) {
122 return false;
123 }
124
125 if (codePoint > 0xffff) {
126 writer.write(toUtf16Escape(codePoint));
127 } else {
128 writer.write("\\u");
129 writer.write(HEX_DIGITS[codePoint >> 12 & 15]);
130 writer.write(HEX_DIGITS[codePoint >> 8 & 15]);
131 writer.write(HEX_DIGITS[codePoint >> 4 & 15]);
132 writer.write(HEX_DIGITS[codePoint & 15]);
133 }
134 return true;
135 }
136 }