1 /* 2 * Copyright (C) 2009 The Guava Authors 3 * 4 * Licensed under the Apache License, Version 2.0 (the "License"); 5 * you may not use this file except in compliance with the License. 6 * You may obtain a copy of the License at 7 * 8 * http://www.apache.org/licenses/LICENSE-2.0 9 * 10 * Unless required by applicable law or agreed to in writing, software 11 * distributed under the License is distributed on an "AS IS" BASIS, 12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13 * See the License for the specific language governing permissions and 14 * limitations under the License. 15 */ 16 17 package com.google.common.escape; 18 19 import java.util.HashMap; 20 import java.util.Map; 21 22 /** 23 * A factory for Escaper instances used to escape strings for safe use in Java. 24 * 25 * <p>This is a subset of source code escapers that are in the process of being open-sources as part 26 * of guava, see: https://github.com/google/guava/issues/1620 27 */ 28 // TODO(cushon): migrate to the guava version once it is open-sourced, and delete this 29 public final class SourceCodeEscapers { SourceCodeEscapers()30 private SourceCodeEscapers() {} 31 32 // For each xxxEscaper() method, please add links to external reference pages 33 // that are considered authoritative for the behavior of that escaper. 34 35 // From: http://en.wikipedia.org/wiki/ASCII#ASCII_printable_characters 36 private static final char PRINTABLE_ASCII_MIN = 0x20; // ' ' 37 private static final char PRINTABLE_ASCII_MAX = 0x7E; // '~' 38 39 private static final char[] HEX_DIGITS = "0123456789abcdef".toCharArray(); 40 41 /** 42 * Returns an {@link Escaper} instance that escapes special characters in a string so it can 43 * safely be included in either a Java character literal or string literal. This is the preferred 44 * way to escape Java characters for use in String or character literals. 45 * 46 * <p>See: <a href= "http://java.sun.com/docs/books/jls/third_edition/html/lexical.html#101089" 47 * >The Java Language Specification</a> for more details. 48 */ javaCharEscaper()49 public static CharEscaper javaCharEscaper() { 50 return JAVA_CHAR_ESCAPER; 51 } 52 53 private static final CharEscaper JAVA_CHAR_ESCAPER; 54 55 static { 56 Map<Character, String> javaMap = new HashMap<>(); 57 javaMap.put('\b', "\\b"); 58 javaMap.put('\f', "\\f"); 59 javaMap.put('\n', "\\n"); 60 javaMap.put('\r', "\\r"); 61 javaMap.put('\t', "\\t"); 62 javaMap.put('\"', "\\\""); 63 javaMap.put('\\', "\\\\"); 64 javaMap.put('\'', "\\'"); 65 JAVA_CHAR_ESCAPER = new JavaCharEscaper(javaMap); 66 } 67 68 // This escaper does not produce octal escape sequences. See: 69 // http://java.sun.com/docs/books/jls/third_edition/html/lexical.html#101089 70 // "Octal escapes are provided for compatibility with C, but can express 71 // only Unicode values \u0000 through \u00FF, so Unicode escapes are 72 // usually preferred." 73 private static class JavaCharEscaper extends ArrayBasedCharEscaper { JavaCharEscaper(Map<Character, String> replacements)74 JavaCharEscaper(Map<Character, String> replacements) { 75 super(replacements, PRINTABLE_ASCII_MIN, PRINTABLE_ASCII_MAX); 76 } 77 78 @Override escapeUnsafe(char c)79 protected char[] escapeUnsafe(char c) { 80 return asUnicodeHexEscape(c); 81 } 82 } 83 84 // Helper for common case of escaping a single char. asUnicodeHexEscape(char c)85 private static char[] asUnicodeHexEscape(char c) { 86 // Equivalent to String.format("\\u%04x", (int)c); 87 char[] r = new char[6]; 88 r[0] = '\\'; 89 r[1] = 'u'; 90 r[5] = HEX_DIGITS[c & 0xF]; 91 c >>>= 4; 92 r[4] = HEX_DIGITS[c & 0xF]; 93 c >>>= 4; 94 r[3] = HEX_DIGITS[c & 0xF]; 95 c >>>= 4; 96 r[2] = HEX_DIGITS[c & 0xF]; 97 return r; 98 } 99 } 100