001/*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements.  See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License.  You may obtain a copy of the License at
008 *
009 *      http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017package org.apache.commons.io.filefilter;
018
019import java.io.File;
020import java.io.IOException;
021import java.io.RandomAccessFile;
022import java.io.Serializable;
023import java.nio.charset.Charset;
024import java.util.Arrays;
025
026/**
027 * <p>
028 * File filter for matching files containing a "magic number". A magic number
029 * is a unique series of bytes common to all files of a specific file format.
030 * For instance, all Java class files begin with the bytes
031 * <code>0xCAFEBABE</code>.
032 * </p>
033 *
034 * <pre>
035 * File dir = new File(".");
036 * MagicNumberFileFilter javaClassFileFilter =
037 *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
038 *       (byte) 0xBA, (byte) 0xBE});
039 * String[] javaClassFiles = dir.list(javaClassFileFilter);
040 * for (String javaClassFile : javaClassFiles) {
041 *     System.out.println(javaClassFile);
042 * }
043 * </pre>
044 *
045 * <p>
046 * Sometimes, such as in the case of TAR files, the
047 * magic number will be offset by a certain number of bytes in the file. In the
048 * case of TAR archive files, this offset is 257 bytes.
049 * </p>
050 *
051 * <pre>
052 * File dir = new File(".");
053 * MagicNumberFileFilter tarFileFilter =
054 *     MagicNumberFileFilter("ustar", 257);
055 * String[] tarFiles = dir.list(tarFileFilter);
056 * for (String tarFile : tarFiles) {
057 *     System.out.println(tarFile);
058 * }
059 * </pre>
060 *
061 * @since 2.0
062 * @see FileFilterUtils#magicNumberFileFilter(byte[])
063 * @see FileFilterUtils#magicNumberFileFilter(String)
064 * @see FileFilterUtils#magicNumberFileFilter(byte[], long)
065 * @see FileFilterUtils#magicNumberFileFilter(String, long)
066 */
067public class MagicNumberFileFilter extends AbstractFileFilter implements
068        Serializable {
069
070    /**
071     * The serialization version unique identifier.
072     */
073    private static final long serialVersionUID = -547733176983104172L;
074
075    /**
076     * The magic number to compare against the file's bytes at the provided
077     * offset.
078     */
079    private final byte[] magicNumbers;
080
081    /**
082     * The offset (in bytes) within the files that the magic number's bytes
083     * should appear.
084     */
085    private final long byteOffset;
086
087    /**
088     * <p>
089     * Constructs a new MagicNumberFileFilter and associates it with the magic
090     * number to test for in files. This constructor assumes a starting offset
091     * of <code>0</code>.
092     * </p>
093     *
094     * <p>
095     * It is important to note that <em>the array is not cloned</em> and that
096     * any changes to the magic number array after construction will affect the
097     * behavior of this file filter.
098     * </p>
099     *
100     * <pre>
101     * MagicNumberFileFilter javaClassFileFilter =
102     *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
103     *       (byte) 0xBA, (byte) 0xBE});
104     * </pre>
105     *
106     * @param magicNumber the magic number to look for in the file.
107     *
108     * @throws IllegalArgumentException if <code>magicNumber</code> is
109     *         {@code null}, or contains no bytes.
110     */
111    public MagicNumberFileFilter(final byte[] magicNumber) {
112        this(magicNumber, 0);
113    }
114
115    /**
116     * <p>
117     * Constructs a new MagicNumberFileFilter and associates it with the magic
118     * number to test for in files. This constructor assumes a starting offset
119     * of <code>0</code>.
120     * </p>
121     *
122     * Example usage:
123     * <pre>
124     * {@code
125     * MagicNumberFileFilter xmlFileFilter =
126     *     MagicNumberFileFilter("<?xml");
127     * }
128     * </pre>
129     *
130     * @param magicNumber the magic number to look for in the file.
131     *        The string is converted to bytes using the platform default charset.
132     *
133     * @throws IllegalArgumentException if <code>magicNumber</code> is
134     *         {@code null} or the empty String.
135     */
136    public MagicNumberFileFilter(final String magicNumber) {
137        this(magicNumber, 0);
138    }
139
140    /**
141     * <p>
142     * Constructs a new MagicNumberFileFilter and associates it with the magic
143     * number to test for in files and the byte offset location in the file to
144     * to look for that magic number.
145     * </p>
146     *
147     * <pre>
148     * MagicNumberFileFilter tarFileFilter =
149     *     MagicNumberFileFilter("ustar", 257);
150     * </pre>
151     *
152     * @param magicNumber the magic number to look for in the file.
153     *        The string is converted to bytes using the platform default charset.
154     * @param offset the byte offset in the file to start comparing bytes.
155     *
156     * @throws IllegalArgumentException if <code>magicNumber</code> is
157     *         {@code null} or the empty String, or <code>offset</code> is
158     *         a negative number.
159     */
160    public MagicNumberFileFilter(final String magicNumber, final long offset) {
161        if (magicNumber == null) {
162            throw new IllegalArgumentException("The magic number cannot be null");
163        }
164        if (magicNumber.isEmpty()) {
165            throw new IllegalArgumentException("The magic number must contain at least one byte");
166        }
167        if (offset < 0) {
168            throw new IllegalArgumentException("The offset cannot be negative");
169        }
170
171        this.magicNumbers = magicNumber.getBytes(Charset.defaultCharset()); // explicitly uses the platform default
172                                                                            // charset
173        this.byteOffset = offset;
174    }
175
176    /**
177     * <p>
178     * Constructs a new MagicNumberFileFilter and associates it with the magic
179     * number to test for in files and the byte offset location in the file to
180     * to look for that magic number.
181     * </p>
182     *
183     * <pre>
184     * MagicNumberFileFilter tarFileFilter =
185     *     MagicNumberFileFilter(new byte[] {0x75, 0x73, 0x74, 0x61, 0x72}, 257);
186     * </pre>
187     *
188     * <pre>
189     * MagicNumberFileFilter javaClassFileFilter =
190     *     MagicNumberFileFilter(new byte[] {0xCA, 0xFE, 0xBA, 0xBE}, 0);
191     * </pre>
192     *
193     * @param magicNumber the magic number to look for in the file.
194     * @param offset the byte offset in the file to start comparing bytes.
195     *
196     * @throws IllegalArgumentException if <code>magicNumber</code> is
197     *         {@code null}, or contains no bytes, or <code>offset</code>
198     *         is a negative number.
199     */
200    public MagicNumberFileFilter(final byte[] magicNumber, final long offset) {
201        if (magicNumber == null) {
202            throw new IllegalArgumentException("The magic number cannot be null");
203        }
204        if (magicNumber.length == 0) {
205            throw new IllegalArgumentException("The magic number must contain at least one byte");
206        }
207        if (offset < 0) {
208            throw new IllegalArgumentException("The offset cannot be negative");
209        }
210
211        this.magicNumbers = new byte[magicNumber.length];
212        System.arraycopy(magicNumber, 0, this.magicNumbers, 0, magicNumber.length);
213        this.byteOffset = offset;
214    }
215
216    /**
217     * <p>
218     * Accepts the provided file if the file contains the file filter's magic
219     * number at the specified offset.
220     * </p>
221     *
222     * <p>
223     * If any {@link IOException}s occur while reading the file, the file will
224     * be rejected.
225     * </p>
226     *
227     * @param file the file to accept or reject.
228     *
229     * @return {@code true} if the file contains the filter's magic number
230     *         at the specified offset, {@code false} otherwise.
231     */
232    @Override
233    public boolean accept(final File file) {
234        if (file != null && file.isFile() && file.canRead()) {
235            try {
236                try (RandomAccessFile randomAccessFile = new RandomAccessFile(file, "r")) {
237                    final byte[] fileBytes = new byte[this.magicNumbers.length];
238                    randomAccessFile.seek(byteOffset);
239                    final int read = randomAccessFile.read(fileBytes);
240                    if (read != magicNumbers.length) {
241                        return false;
242                    }
243                    return Arrays.equals(this.magicNumbers, fileBytes);
244                }
245            }
246            catch (final IOException ioe) {
247                // Do nothing, fall through and do not accept file
248            }
249        }
250
251        return false;
252    }
253
254    /**
255     * Returns a String representation of the file filter, which includes the
256     * magic number bytes and byte offset.
257     *
258     * @return a String representation of the file filter.
259     */
260    @Override
261    public String toString() {
262        final StringBuilder builder = new StringBuilder(super.toString());
263        builder.append("(");
264        builder.append(new String(magicNumbers, Charset.defaultCharset()));// TODO perhaps use hex if value is not
265                                                                           // printable
266        builder.append(",");
267        builder.append(this.byteOffset);
268        builder.append(")");
269        return builder.toString();
270    }
271}