| Classes in this File | Line Coverage | Branch Coverage | Complexity | ||||
| CharSequenceTranslator |
|
| 3.4;3.4 |
| 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.lang3.text.translate; | |
| 18 | ||
| 19 | import java.io.IOException; | |
| 20 | import java.io.StringWriter; | |
| 21 | import java.io.Writer; | |
| 22 | import java.util.Locale; | |
| 23 | ||
| 24 | /** | |
| 25 | * An API for translating text. | |
| 26 | * Its core use is to escape and unescape text. Because escaping and unescaping | |
| 27 | * is completely contextual, the API does not present two separate signatures. | |
| 28 | * | |
| 29 | * @since 3.0 | |
| 30 | * @version $Id: CharSequenceTranslator.java 1436770 2013-01-22 07:09:45Z ggregory $ | |
| 31 | */ | |
| 32 | 65 | public abstract class CharSequenceTranslator { |
| 33 | ||
| 34 | /** | |
| 35 | * Translate a set of codepoints, represented by an int index into a CharSequence, | |
| 36 | * into another set of codepoints. The number of codepoints consumed must be returned, | |
| 37 | * and the only IOExceptions thrown must be from interacting with the Writer so that | |
| 38 | * the top level API may reliable ignore StringWriter IOExceptions. | |
| 39 | * | |
| 40 | * @param input CharSequence that is being translated | |
| 41 | * @param index int representing the current point of translation | |
| 42 | * @param out Writer to translate the text to | |
| 43 | * @return int count of codepoints consumed | |
| 44 | * @throws IOException if and only if the Writer produces an IOException | |
| 45 | */ | |
| 46 | public abstract int translate(CharSequence input, int index, Writer out) throws IOException; | |
| 47 | ||
| 48 | /** | |
| 49 | * Helper for non-Writer usage. | |
| 50 | * @param input CharSequence to be translated | |
| 51 | * @return String output of translation | |
| 52 | */ | |
| 53 | public final String translate(final CharSequence input) { | |
| 54 | 65897 | if (input == null) { |
| 55 | 9 | return null; |
| 56 | } | |
| 57 | try { | |
| 58 | 65888 | final StringWriter writer = new StringWriter(input.length() * 2); |
| 59 | 65888 | translate(input, writer); |
| 60 | 65885 | return writer.toString(); |
| 61 | 0 | } catch (final IOException ioe) { |
| 62 | // this should never ever happen while writing to a StringWriter | |
| 63 | 0 | throw new RuntimeException(ioe); |
| 64 | } | |
| 65 | } | |
| 66 | ||
| 67 | /** | |
| 68 | * Translate an input onto a Writer. This is intentionally final as its algorithm is | |
| 69 | * tightly coupled with the abstract method of this class. | |
| 70 | * | |
| 71 | * @param input CharSequence that is being translated | |
| 72 | * @param out Writer to translate the text to | |
| 73 | * @throws IOException if and only if the Writer produces an IOException | |
| 74 | */ | |
| 75 | public final void translate(final CharSequence input, final Writer out) throws IOException { | |
| 76 | 65956 | if (out == null) { |
| 77 | 6 | throw new IllegalArgumentException("The Writer must not be null"); |
| 78 | } | |
| 79 | 65950 | if (input == null) { |
| 80 | 4 | return; |
| 81 | } | |
| 82 | 65946 | int pos = 0; |
| 83 | 65946 | final int len = input.length(); |
| 84 | 210086 | while (pos < len) { |
| 85 | 144143 | final int consumed = translate(input, pos, out); |
| 86 | 144140 | if (consumed == 0) { |
| 87 | 10670 | final char[] c = Character.toChars(Character.codePointAt(input, pos)); |
| 88 | 10670 | out.write(c); |
| 89 | 10670 | pos+= c.length; |
| 90 | 10670 | continue; |
| 91 | } | |
| 92 | // // contract with translators is that they have to understand codepoints | |
| 93 | // // and they just took care of a surrogate pair | |
| 94 | 1176117 | for (int pt = 0; pt < consumed; pt++) { |
| 95 | 1042647 | pos += Character.charCount(Character.codePointAt(input, pt)); |
| 96 | } | |
| 97 | 133470 | } |
| 98 | 65943 | } |
| 99 | ||
| 100 | /** | |
| 101 | * Helper method to create a merger of this translator with another set of | |
| 102 | * translators. Useful in customizing the standard functionality. | |
| 103 | * | |
| 104 | * @param translators CharSequenceTranslator array of translators to merge with this one | |
| 105 | * @return CharSequenceTranslator merging this translator with the others | |
| 106 | */ | |
| 107 | public final CharSequenceTranslator with(final CharSequenceTranslator... translators) { | |
| 108 | 4 | final CharSequenceTranslator[] newArray = new CharSequenceTranslator[translators.length + 1]; |
| 109 | 4 | newArray[0] = this; |
| 110 | 4 | System.arraycopy(translators, 0, newArray, 1, translators.length); |
| 111 | 4 | return new AggregateTranslator(newArray); |
| 112 | } | |
| 113 | ||
| 114 | /** | |
| 115 | * <p>Returns an upper case hexadecimal <code>String</code> for the given | |
| 116 | * character.</p> | |
| 117 | * | |
| 118 | * @param codepoint The codepoint to convert. | |
| 119 | * @return An upper case hexadecimal <code>String</code> | |
| 120 | */ | |
| 121 | public static String hex(final int codepoint) { | |
| 122 | 608 | return Integer.toHexString(codepoint).toUpperCase(Locale.ENGLISH); |
| 123 | } | |
| 124 | ||
| 125 | } |