001 /*
002 * Licensed to the Apache Software Foundation (ASF) under one
003 * or more contributor license agreements. See the NOTICE file
004 * distributed with this work for additional information
005 * regarding copyright ownership. The ASF licenses this file
006 * to you under the Apache License, Version 2.0 (the
007 * "License"); you may not use this file except in compliance
008 * with the License. You may obtain a copy of the License at
009 *
010 * http://www.apache.org/licenses/LICENSE-2.0
011 *
012 * Unless required by applicable law or agreed to in writing,
013 * software distributed under the License is distributed on an
014 * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
015 * KIND, either express or implied. See the License for the
016 * specific language governing permissions and limitations
017 * under the License.
018 */
019 package org.apache.commons.compress.compressors.gzip;
020
021 import java.util.HashMap;
022 import java.util.Locale;
023 import java.util.Map;
024
025 /**
026 * Utility code for the gzip compression format.
027 * @ThreadSafe
028 */
029 public class GzipUtils {
030
031 /**
032 * Map from common filename suffixes to the suffixes that identify gzipped
033 * versions of those file types. For example: from ".tar" to ".tgz".
034 */
035 private static final Map<String, String> compressSuffix =
036 new HashMap<String, String>();
037
038 /**
039 * Map from common filename suffixes of gzipped files to the corresponding
040 * suffixes of uncompressed files. For example: from ".tgz" to ".tar".
041 * <p>
042 * This map also contains gzip-specific suffixes like ".gz" and "-z".
043 * These suffixes are mapped to the empty string, as they should simply
044 * be removed from the filename when the file is uncompressed.
045 */
046 private static final Map<String, String> uncompressSuffix =
047 new HashMap<String, String>();
048
049 static {
050 compressSuffix.put(".tar", ".tgz");
051 compressSuffix.put(".svg", ".svgz");
052 compressSuffix.put(".cpio", ".cpgz");
053 compressSuffix.put(".wmf", ".wmz");
054 compressSuffix.put(".emf", ".emz");
055
056 uncompressSuffix.put(".tgz", ".tar");
057 uncompressSuffix.put(".taz", ".tar");
058 uncompressSuffix.put(".svgz", ".svg");
059 uncompressSuffix.put(".cpgz", ".cpio");
060 uncompressSuffix.put(".wmz", ".wmf");
061 uncompressSuffix.put(".emz", ".emf");
062 uncompressSuffix.put(".gz", "");
063 uncompressSuffix.put(".z", "");
064 uncompressSuffix.put("-gz", "");
065 uncompressSuffix.put("-z", "");
066 uncompressSuffix.put("_z", "");
067 }
068 // N.B. if any shorter or longer keys are added, ensure the for loop limits are changed
069
070 /** Private constructor to prevent instantiation of this utility class. */
071 private GzipUtils() {
072 }
073
074 /**
075 * Detects common gzip suffixes in the given filename.
076 *
077 * @param filename name of a file
078 * @return <code>true</code> if the filename has a common gzip suffix,
079 * <code>false</code> otherwise
080 */
081 public static boolean isCompressedFilename(String filename) {
082 String lower = filename.toLowerCase(Locale.ENGLISH);
083 int n = lower.length();
084 // Shortest suffix is two letters (_z), longest is five (.svgz)
085 for (int i = 2; i <= 5 && i < n; i++) {
086 if (uncompressSuffix.containsKey(lower.substring(n - i))) {
087 return true;
088 }
089 }
090 return false;
091 }
092
093 /**
094 * Maps the given name of a gzip-compressed file to the name that the
095 * file should have after uncompression. Commonly used file type specific
096 * suffixes like ".tgz" or ".svgz" are automatically detected and
097 * correctly mapped. For example the name "package.tgz" is mapped to
098 * "package.tar". And any filenames with the generic ".gz" suffix
099 * (or any other generic gzip suffix) is mapped to a name without that
100 * suffix. If no gzip suffix is detected, then the filename is returned
101 * unmapped.
102 *
103 * @param filename name of a file
104 * @return name of the corresponding uncompressed file
105 */
106 public static String getUncompressedFilename(String filename) {
107 String lower = filename.toLowerCase(Locale.ENGLISH);
108 int n = lower.length();
109 // Shortest suffix is two letters (_z), longest is five (.svgz)
110 for (int i = 2; i <= 5 && i < n; i++) {
111 Object suffix = uncompressSuffix.get(lower.substring(n - i));
112 if (suffix != null) {
113 return filename.substring(0, n - i) + suffix;
114 }
115 }
116 return filename;
117 }
118
119 /**
120 * Maps the given filename to the name that the file should have after
121 * compression with gzip. Common file types with custom suffixes for
122 * compressed versions are automatically detected and correctly mapped.
123 * For example the name "package.tar" is mapped to "package.tgz". If no
124 * custom mapping is applicable, then the default ".gz" suffix is appended
125 * to the filename.
126 *
127 * @param filename name of a file
128 * @return name of the corresponding compressed file
129 */
130 public static String getCompressedFilename(String filename) {
131 String lower = filename.toLowerCase(Locale.ENGLISH);
132 int n = lower.length();
133 // Shortest suffix is four letters (.svg), longest is five (.cpio)
134 for (int i = 4; i <= 5 && i < n; i++) {
135 Object suffix = compressSuffix.get(lower.substring(n - i));
136 if (suffix != null) {
137 return filename.substring(0, n - i) + suffix;
138 }
139 }
140 // No custom suffix found, just append the default .gz
141 return filename + ".gz";
142 }
143
144 }