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
026import org.apache.commons.io.IOUtils;
027
028/**
029 * <p>
030 * File filter for matching files containing a "magic number". A magic number
031 * is a unique series of bytes common to all files of a specific file format.
032 * For instance, all Java class files begin with the bytes
033 * <code>0xCAFEBABE</code>.
034 * </p>
035 *
036 * <pre>
037 * File dir = new File(".");
038 * MagicNumberFileFilter javaClassFileFilter =
039 *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
040 *       (byte) 0xBA, (byte) 0xBE});
041 * String[] javaClassFiles = dir.list(javaClassFileFilter);
042 * for (String javaClassFile : javaClassFiles) {
043 *     System.out.println(javaClassFile);
044 * }
045 * </pre>
046 *
047 * <p>
048 * Sometimes, such as in the case of TAR files, the
049 * magic number will be offset by a certain number of bytes in the file. In the
050 * case of TAR archive files, this offset is 257 bytes.
051 * </p>
052 *
053 * <pre>
054 * File dir = new File(".");
055 * MagicNumberFileFilter tarFileFilter =
056 *     MagicNumberFileFilter("ustar", 257);
057 * String[] tarFiles = dir.list(tarFileFilter);
058 * for (String tarFile : tarFiles) {
059 *     System.out.println(tarFile);
060 * }
061 * </pre>
062 *
063 * @since 2.0
064 * @see FileFilterUtils#magicNumberFileFilter(byte[])
065 * @see FileFilterUtils#magicNumberFileFilter(String)
066 * @see FileFilterUtils#magicNumberFileFilter(byte[], long)
067 * @see FileFilterUtils#magicNumberFileFilter(String, long)
068 */
069public class MagicNumberFileFilter extends AbstractFileFilter implements
070        Serializable {
071
072    /**
073     * The serialization version unique identifier.
074     */
075    private static final long serialVersionUID = -547733176983104172L;
076
077    /**
078     * The magic number to compare against the file's bytes at the provided
079     * offset.
080     */
081    private final byte[] magicNumbers;
082
083    /**
084     * The offset (in bytes) within the files that the magic number's bytes
085     * should appear.
086     */
087    private final long byteOffset;
088
089    /**
090     * <p>
091     * Constructs a new MagicNumberFileFilter and associates it with the magic
092     * number to test for in files. This constructor assumes a starting offset
093     * of <code>0</code>.
094     * </p>
095     *
096     * <p>
097     * It is important to note that <em>the array is not cloned</em> and that
098     * any changes to the magic number array after construction will affect the
099     * 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.isEmpty()) {
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
174                                                                            // charset
175        this.byteOffset = offset;
176    }
177
178    /**
179     * <p>
180     * Constructs a new MagicNumberFileFilter and associates it with the magic
181     * number to test for in files and the byte offset location in the file to
182     * to look for that magic number.
183     * </p>
184     *
185     * <pre>
186     * MagicNumberFileFilter tarFileFilter =
187     *     MagicNumberFileFilter(new byte[] {0x75, 0x73, 0x74, 0x61, 0x72}, 257);
188     * </pre>
189     *
190     * <pre>
191     * MagicNumberFileFilter javaClassFileFilter =
192     *     MagicNumberFileFilter(new byte[] {0xCA, 0xFE, 0xBA, 0xBE}, 0);
193     * </pre>
194     *
195     * @param magicNumber the magic number to look for in the file.
196     * @param offset the byte offset in the file to start comparing bytes.
197     *
198     * @throws IllegalArgumentException if <code>magicNumber</code> is
199     *         {@code null}, or contains no bytes, or <code>offset</code>
200     *         is a negative number.
201     */
202    public MagicNumberFileFilter(final byte[] magicNumber, final long offset) {
203        if (magicNumber == null) {
204            throw new IllegalArgumentException("The magic number cannot be null");
205        }
206        if (magicNumber.length == 0) {
207            throw new IllegalArgumentException("The magic number must contain at least one byte");
208        }
209        if (offset < 0) {
210            throw new IllegalArgumentException("The offset cannot be negative");
211        }
212
213        this.magicNumbers = new byte[magicNumber.length];
214        System.arraycopy(magicNumber, 0, this.magicNumbers, 0, magicNumber.length);
215        this.byteOffset = offset;
216    }
217
218    /**
219     * <p>
220     * Accepts the provided file if the file contains the file filter's magic
221     * number at the specified offset.
222     * </p>
223     *
224     * <p>
225     * If any {@link IOException}s occur while reading the file, the file will
226     * be rejected.
227     * </p>
228     *
229     * @param file the file to accept or reject.
230     *
231     * @return {@code true} if the file contains the filter's magic number
232     *         at the specified offset, {@code false} otherwise.
233     */
234    @Override
235    public boolean accept(final File file) {
236        if (file != null && file.isFile() && file.canRead()) {
237            RandomAccessFile randomAccessFile = null;
238            try {
239                final byte[] fileBytes = new byte[this.magicNumbers.length];
240                randomAccessFile = new RandomAccessFile(file, "r");
241                randomAccessFile.seek(byteOffset);
242                final int read = randomAccessFile.read(fileBytes);
243                if (read != magicNumbers.length) {
244                    return false;
245                }
246                return Arrays.equals(this.magicNumbers, fileBytes);
247            } catch (final IOException ioe) {
248                // Do nothing, fall through and do not accept file
249            } finally {
250                IOUtils.closeQuietly(randomAccessFile);
251            }
252        }
253
254        return false;
255    }
256
257    /**
258     * Returns a String representation of the file filter, which includes the
259     * magic number bytes and byte offset.
260     *
261     * @return a String representation of the file filter.
262     */
263    @Override
264    public String toString() {
265        final StringBuilder builder = new StringBuilder(super.toString());
266        builder.append("(");
267        builder.append(new String(magicNumbers, Charset.defaultCharset()));// TODO perhaps use hex if value is not
268                                                                           // printable
269        builder.append(",");
270        builder.append(this.byteOffset);
271        builder.append(")");
272        return builder.toString();
273    }
274}