1 /*
2 * Licensed to the Apache Software Foundation (ASF) under one
3 * or more contributor license agreements. See the NOTICE file
4 * distributed with this work for additional information
5 * regarding copyright ownership. The ASF licenses this file
6 * to you under the Apache License, Version 2.0 (the
7 * "License"); you may not use this file except in compliance
8 * with the License. You may obtain a copy of the License at
9 *
10 * https://www.apache.org/licenses/LICENSE-2.0
11 *
12 * Unless required by applicable law or agreed to in writing,
13 * software distributed under the License is distributed on an
14 * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
15 * KIND, either express or implied. See the License for the
16 * specific language governing permissions and limitations
17 * under the License.
18 */
19 package org.apache.commons.compress.compressors.gzip;
20
21 import java.nio.charset.Charset;
22 import java.nio.charset.StandardCharsets;
23 import java.util.LinkedHashMap;
24 import java.util.Map;
25
26 import org.apache.commons.compress.compressors.FileNameUtil;
27
28 /**
29 * Utility code for the GZIP compression format.
30 *
31 * @see <a href="https://datatracker.ietf.org/doc/html/rfc1952">RFC 1952 GZIP File Format Specification</a>
32 * @ThreadSafe
33 */
34 public class GzipUtils {
35
36 /** Header flag indicating a comment follows the header. */
37 static final int FCOMMENT = 0x10;
38
39 /** Header flag indicating an EXTRA subfields collection follows the header. */
40 static final int FEXTRA = 0x04;
41
42 /** Header flag indicating a header CRC follows the header. */
43 static final int FHCRC = 0x02;
44
45 private static final FileNameUtil fileNameUtil;
46
47 /** Header flag indicating a file name follows the header. */
48 static final int FNAME = 0x08;
49
50 static final int FRESERVED = 0xE0;
51
52 /**
53 * Charset for file name and comments per the <a href="https://tools.ietf.org/html/rfc1952">GZIP File Format Specification</a>.
54 */
55 static final Charset GZIP_ENCODING = StandardCharsets.ISO_8859_1;
56
57 /**
58 * Member header ID1 (IDentification 1).
59 *
60 * See <a href="https://datatracker.ietf.org/doc/html/rfc1952#page-5">RFC1952</a> 2.3.1. Member header and trailer.
61 */
62 static final int ID1 = 31;
63
64 /**
65 * Member header ID2 (IDentification 2).
66 *
67 * See <a href="https://datatracker.ietf.org/doc/html/rfc1952#page-5">RFC1952</a> 2.3.1. Member header and trailer.
68 */
69 static final int ID2 = 139;
70
71 /**
72 * Member header XFL (eXtra FLags) when the "deflate" method (CM = 8) is set, then XFL = 2 means the compressor used maximum compression (slowest
73 * algorithm).
74 *
75 * See <a href="https://datatracker.ietf.org/doc/html/rfc1952#page-5">RFC1952</a> 2.3.1. Member header and trailer.
76 */
77 static final byte XFL_MAX_COMPRESSION = 2;
78
79 /**
80 * Member header XFL (eXtra FLags) when the "deflate" method (CM = 8) is set, then XFL = 4 means the compressor used the fastest algorithm.
81 *
82 * See <a href="https://datatracker.ietf.org/doc/html/rfc1952#page-5">RFC1952</a> 2.3.1. Member header and trailer.
83 */
84 static final byte XFL_MAX_SPEED = 4;
85
86 static final byte XFL_UNKNOWN = 0;
87
88 /**
89 * Using {@link LinkedHashMap} so {@code .tgz} is preferred over {@code .taz} as compressed extension of {@code .tar} as FileNameUtil will use the first one
90 * found.
91 */
92 static {
93 final Map<String, String> uncompressSuffix = new LinkedHashMap<>();
94 uncompressSuffix.put(".tgz", ".tar");
95 uncompressSuffix.put(".taz", ".tar");
96 uncompressSuffix.put(".svgz", ".svg");
97 uncompressSuffix.put(".cpgz", ".cpio");
98 uncompressSuffix.put(".wmz", ".wmf");
99 uncompressSuffix.put(".emz", ".emf");
100 uncompressSuffix.put(".gz", "");
101 uncompressSuffix.put(".z", "");
102 uncompressSuffix.put("-gz", "");
103 uncompressSuffix.put("-z", "");
104 uncompressSuffix.put("_z", "");
105 fileNameUtil = new FileNameUtil(uncompressSuffix, ".gz");
106 }
107 /**
108 * Maps the given file name to the name that the file should have after compression with gzip. Common file types with custom suffixes for compressed
109 * versions are automatically detected and correctly mapped. For example the name "package.tar" is mapped to "package.tgz". If no custom mapping is
110 * applicable, then the default ".gz" suffix is appended to the file name.
111 *
112 * @param fileName name of a file
113 * @return name of the corresponding compressed file
114 * @deprecated Use {@link #getCompressedFileName(String)}.
115 */
116 @Deprecated
117 public static String getCompressedFilename(final String fileName) {
118 return fileNameUtil.getCompressedFileName(fileName);
119 }
120
121 /**
122 * Maps the given file name to the name that the file should have after compression with gzip. Common file types with custom suffixes for compressed
123 * versions are automatically detected and correctly mapped. For example the name "package.tar" is mapped to "package.tgz". If no custom mapping is
124 * applicable, then the default ".gz" suffix is appended to the file name.
125 *
126 * @param fileName name of a file
127 * @return name of the corresponding compressed file
128 * @since 1.25.0
129 */
130 public static String getCompressedFileName(final String fileName) {
131 return fileNameUtil.getCompressedFileName(fileName);
132 }
133
134 /**
135 * Maps the given name of a gzip-compressed file to the name that the file should have after uncompression. Commonly used file type specific suffixes like
136 * ".tgz" or ".svgz" are automatically detected and correctly mapped. For example the name "package.tgz" is mapped to "package.tar". And any file names with
137 * the generic ".gz" suffix (or any other generic gzip suffix) is mapped to a name without that suffix. If no gzip suffix is detected, then the file name is
138 * returned unmapped.
139 *
140 * @param fileName name of a file
141 * @return name of the corresponding uncompressed file
142 * @deprecated Use {@link #getUncompressedFileName(String)}.
143 */
144 @Deprecated
145 public static String getUncompressedFilename(final String fileName) {
146 return fileNameUtil.getUncompressedFileName(fileName);
147 }
148
149 /**
150 * Maps the given name of a gzip-compressed file to the name that the file should have after uncompression. Commonly used file type specific suffixes like
151 * ".tgz" or ".svgz" are automatically detected and correctly mapped. For example the name "package.tgz" is mapped to "package.tar". And any file names with
152 * the generic ".gz" suffix (or any other generic gzip suffix) is mapped to a name without that suffix. If no gzip suffix is detected, then the file name is
153 * returned unmapped.
154 *
155 * @param fileName name of a file
156 * @return name of the corresponding uncompressed file
157 * @since 1.25.0
158 */
159 public static String getUncompressedFileName(final String fileName) {
160 return fileNameUtil.getUncompressedFileName(fileName);
161 }
162
163 /**
164 * Detects common gzip suffixes in the given file name.
165 *
166 * @param fileName name of a file
167 * @return {@code true} if the file name has a common gzip suffix, {@code false} otherwise
168 * @deprecated Use {@link #isCompressedFileName(String)}.
169 */
170 @Deprecated
171 public static boolean isCompressedFilename(final String fileName) {
172 return fileNameUtil.isCompressedFileName(fileName);
173 }
174
175 /**
176 * Detects common gzip suffixes in the given file name.
177 *
178 * @param fileName name of a file
179 * @return {@code true} if the file name has a common gzip suffix, {@code false} otherwise
180 * @since 1.25.0
181 */
182 public static boolean isCompressedFileName(final String fileName) {
183 return fileNameUtil.isCompressedFileName(fileName);
184 }
185
186 /** Private constructor to prevent instantiation of this utility class. */
187 private GzipUtils() {
188 }
189
190 }