View Javadoc
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.io.filefilter;
18  
19  import java.io.File;
20  import java.io.IOException;
21  import java.io.RandomAccessFile;
22  import java.io.Serializable;
23  import java.nio.charset.Charset;
24  import java.util.Arrays;
25  
26  import org.apache.commons.io.IOUtils;
27  
28  /**
29   * <p>
30   * File filter for matching files containing a "magic number". A magic number
31   * is a unique series of bytes common to all files of a specific file format.
32   * For instance, all Java class files begin with the bytes
33   * <code>0xCAFEBABE</code>.
34   * </p>
35   *
36   * <pre>
37   * File dir = new File(".");
38   * MagicNumberFileFilter javaClassFileFilter =
39   *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
40   *       (byte) 0xBA, (byte) 0xBE});
41   * String[] javaClassFiles = dir.list(javaClassFileFilter);
42   * for (String javaClassFile : javaClassFiles) {
43   *     System.out.println(javaClassFile);
44   * }
45   * </pre>
46   *
47   * <p>
48   * Sometimes, such as in the case of TAR files, the
49   * magic number will be offset by a certain number of bytes in the file. In the
50   * case of TAR archive files, this offset is 257 bytes.
51   * </p>
52   *
53   * <pre>
54   * File dir = new File(".");
55   * MagicNumberFileFilter tarFileFilter =
56   *     MagicNumberFileFilter("ustar", 257);
57   * String[] tarFiles = dir.list(tarFileFilter);
58   * for (String tarFile : tarFiles) {
59   *     System.out.println(tarFile);
60   * }
61   * </pre>
62   *
63   * @since 2.0
64   * @see FileFilterUtils#magicNumberFileFilter(byte[])
65   * @see FileFilterUtils#magicNumberFileFilter(String)
66   * @see FileFilterUtils#magicNumberFileFilter(byte[], long)
67   * @see FileFilterUtils#magicNumberFileFilter(String, long)
68   */
69  public class MagicNumberFileFilter extends AbstractFileFilter implements
70          Serializable {
71  
72      /**
73       * The serialization version unique identifier.
74       */
75      private static final long serialVersionUID = -547733176983104172L;
76  
77      /**
78       * The magic number to compare against the file's bytes at the provided
79       * offset.
80       */
81      private final byte[] magicNumbers;
82  
83      /**
84       * The offset (in bytes) within the files that the magic number's bytes
85       * should appear.
86       */
87      private final long byteOffset;
88  
89      /**
90       * <p>
91       * Constructs a new MagicNumberFileFilter and associates it with the magic
92       * number to test for in files. This constructor assumes a starting offset
93       * of <code>0</code>.
94       * </p>
95       *
96       * <p>
97       * It is important to note that <em>the array is not cloned</em> and that
98       * any changes to the magic number array after construction will affect the
99       * behavior of this file filter.
100      * </p>
101      *
102      * <pre>
103      * MagicNumberFileFilter javaClassFileFilter =
104      *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
105      *       (byte) 0xBA, (byte) 0xBE});
106      * </pre>
107      *
108      * @param magicNumber the magic number to look for in the file.
109      *
110      * @throws IllegalArgumentException if <code>magicNumber</code> is
111      *         {@code null}, or contains no bytes.
112      */
113     public MagicNumberFileFilter(final byte[] magicNumber) {
114         this(magicNumber, 0);
115     }
116 
117     /**
118      * <p>
119      * Constructs a new MagicNumberFileFilter and associates it with the magic
120      * number to test for in files. This constructor assumes a starting offset
121      * of <code>0</code>.
122      * </p>
123      *
124      * Example usage:
125      * <pre>
126      * {@code
127      * MagicNumberFileFilter xmlFileFilter =
128      *     MagicNumberFileFilter("<?xml");
129      * }
130      * </pre>
131      *
132      * @param magicNumber the magic number to look for in the file.
133      *        The string is converted to bytes using the platform default charset.
134      *
135      * @throws IllegalArgumentException if <code>magicNumber</code> is
136      *         {@code null} or the empty String.
137      */
138     public MagicNumberFileFilter(final String magicNumber) {
139         this(magicNumber, 0);
140     }
141 
142     /**
143      * <p>
144      * Constructs a new MagicNumberFileFilter and associates it with the magic
145      * number to test for in files and the byte offset location in the file to
146      * to look for that magic number.
147      * </p>
148      *
149      * <pre>
150      * MagicNumberFileFilter tarFileFilter =
151      *     MagicNumberFileFilter("ustar", 257);
152      * </pre>
153      *
154      * @param magicNumber the magic number to look for in the file.
155      *        The string is converted to bytes using the platform default charset.
156      * @param offset the byte offset in the file to start comparing bytes.
157      *
158      * @throws IllegalArgumentException if <code>magicNumber</code> is
159      *         {@code null} or the empty String, or <code>offset</code> is
160      *         a negative number.
161      */
162     public MagicNumberFileFilter(final String magicNumber, final long offset) {
163         if (magicNumber == null) {
164             throw new IllegalArgumentException("The magic number cannot be null");
165         }
166         if (magicNumber.length() == 0) {
167             throw new IllegalArgumentException("The magic number must contain at least one byte");
168         }
169         if (offset < 0) {
170             throw new IllegalArgumentException("The offset cannot be negative");
171         }
172 
173         this.magicNumbers = magicNumber.getBytes(Charset.defaultCharset()); // explicitly uses the platform default charset
174         this.byteOffset = offset;
175     }
176 
177     /**
178      * <p>
179      * Constructs a new MagicNumberFileFilter and associates it with the magic
180      * number to test for in files and the byte offset location in the file to
181      * to look for that magic number.
182      * </p>
183      *
184      * <pre>
185      * MagicNumberFileFilter tarFileFilter =
186      *     MagicNumberFileFilter(new byte[] {0x75, 0x73, 0x74, 0x61, 0x72}, 257);
187      * </pre>
188      *
189      * <pre>
190      * MagicNumberFileFilter javaClassFileFilter =
191      *     MagicNumberFileFilter(new byte[] {0xCA, 0xFE, 0xBA, 0xBE}, 0);
192      * </pre>
193      *
194      * @param magicNumber the magic number to look for in the file.
195      * @param offset the byte offset in the file to start comparing bytes.
196      *
197      * @throws IllegalArgumentException if <code>magicNumber</code> is
198      *         {@code null}, or contains no bytes, or <code>offset</code>
199      *         is a negative number.
200      */
201     public MagicNumberFileFilter(final byte[] magicNumber, final long offset) {
202         if (magicNumber == null) {
203             throw new IllegalArgumentException("The magic number cannot be null");
204         }
205         if (magicNumber.length == 0) {
206             throw new IllegalArgumentException("The magic number must contain at least one byte");
207         }
208         if (offset < 0) {
209             throw new IllegalArgumentException("The offset cannot be negative");
210         }
211 
212         this.magicNumbers = new byte[magicNumber.length];
213         System.arraycopy(magicNumber, 0, this.magicNumbers, 0, magicNumber.length);
214         this.byteOffset = offset;
215     }
216 
217     /**
218      * <p>
219      * Accepts the provided file if the file contains the file filter's magic
220      * number at the specified offset.
221      * </p>
222      *
223      * <p>
224      * If any {@link IOException}s occur while reading the file, the file will
225      * be rejected.
226      * </p>
227      *
228      * @param file the file to accept or reject.
229      *
230      * @return {@code true} if the file contains the filter's magic number
231      *         at the specified offset, {@code false} otherwise.
232      */
233     @Override
234     public boolean accept(final File file) {
235         if (file != null && file.isFile() && file.canRead()) {
236             RandomAccessFile randomAccessFile = null;
237             try {
238                 final byte[] fileBytes = new byte[this.magicNumbers.length];
239                 randomAccessFile = new RandomAccessFile(file, "r");
240                 randomAccessFile.seek(byteOffset);
241                 final int read = randomAccessFile.read(fileBytes);
242                 if (read != magicNumbers.length) {
243                     return false;
244                 }
245                 return Arrays.equals(this.magicNumbers, fileBytes);
246             } catch (final IOException ioe) {
247                 // Do nothing, fall through and do not accept file
248             } finally {
249                 IOUtils.closeQuietly(randomAccessFile);
250             }
251         }
252 
253         return false;
254     }
255 
256     /**
257      * Returns a String representation of the file filter, which includes the
258      * magic number bytes and byte offset.
259      *
260      * @return a String representation of the file filter.
261      */
262     @Override
263     public String toString() {
264         final StringBuilder builder = new StringBuilder(super.toString());
265         builder.append("(");
266         builder.append(new String(magicNumbers, Charset.defaultCharset()));// TODO perhaps use hex if value is not printable
267         builder.append(",");
268         builder.append(this.byteOffset);
269         builder.append(")");
270         return builder.toString();
271     }
272 }