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.ByteBuffer;
024import java.nio.channels.FileChannel;
025import java.nio.charset.Charset;
026import java.nio.file.FileVisitResult;
027import java.nio.file.Files;
028import java.nio.file.Path;
029import java.nio.file.attribute.BasicFileAttributes;
030import java.util.Arrays;
031import java.util.Objects;
032
033import org.apache.commons.io.IOUtils;
034import org.apache.commons.io.RandomAccessFileMode;
035
036/**
037 * <p>
038 * File filter for matching files containing a "magic number". A magic number
039 * is a unique series of bytes common to all files of a specific file format.
040 * For instance, all Java class files begin with the bytes
041 * {@code 0xCAFEBABE}.
042 * </p>
043 * <h2>Using Classic IO</h2>
044 * <pre>
045 * File dir = FileUtils.current();
046 * MagicNumberFileFilter javaClassFileFilter =
047 *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
048 *       (byte) 0xBA, (byte) 0xBE});
049 * String[] javaClassFiles = dir.list(javaClassFileFilter);
050 * for (String javaClassFile : javaClassFiles) {
051 *     System.out.println(javaClassFile);
052 * }
053 * </pre>
054 *
055 * <p>
056 * Sometimes, such as in the case of TAR files, the
057 * magic number will be offset by a certain number of bytes in the file. In the
058 * case of TAR archive files, this offset is 257 bytes.
059 * </p>
060 *
061 * <pre>
062 * File dir = FileUtils.current();
063 * MagicNumberFileFilter tarFileFilter =
064 *     MagicNumberFileFilter("ustar", 257);
065 * String[] tarFiles = dir.list(tarFileFilter);
066 * for (String tarFile : tarFiles) {
067 *     System.out.println(tarFile);
068 * }
069 * </pre>
070 * <h2>Using NIO</h2>
071 * <pre>
072 * final Path dir = PathUtils.current();
073 * final AccumulatorPathVisitor visitor = AccumulatorPathVisitor.withLongCounters(MagicNumberFileFilter("ustar", 257));
074 * //
075 * // Walk one dir
076 * Files.<b>walkFileTree</b>(dir, Collections.emptySet(), 1, visitor);
077 * System.out.println(visitor.getPathCounters());
078 * System.out.println(visitor.getFileList());
079 * //
080 * visitor.getPathCounters().reset();
081 * //
082 * // Walk dir tree
083 * Files.<b>walkFileTree</b>(dir, visitor);
084 * System.out.println(visitor.getPathCounters());
085 * System.out.println(visitor.getDirList());
086 * System.out.println(visitor.getFileList());
087 * </pre>
088 * <h2>Deprecating Serialization</h2>
089 * <p>
090 * <em>Serialization is deprecated and will be removed in 3.0.</em>
091 * </p>
092 *
093 * @since 2.0
094 * @see FileFilterUtils#magicNumberFileFilter(byte[])
095 * @see FileFilterUtils#magicNumberFileFilter(String)
096 * @see FileFilterUtils#magicNumberFileFilter(byte[], long)
097 * @see FileFilterUtils#magicNumberFileFilter(String, long)
098 */
099public class MagicNumberFileFilter extends AbstractFileFilter implements
100        Serializable {
101
102    /**
103     * The serialization version unique identifier.
104     */
105    private static final long serialVersionUID = -547733176983104172L;
106
107    /**
108     * The magic number to compare against the file's bytes at the provided
109     * offset.
110     */
111    private final byte[] magicNumbers;
112
113    /**
114     * The offset (in bytes) within the files that the magic number's bytes
115     * should appear.
116     */
117    private final long byteOffset;
118
119    /**
120     * <p>
121     * Constructs a new MagicNumberFileFilter and associates it with the magic
122     * number to test for in files. This constructor assumes a starting offset
123     * of {@code 0}.
124     * </p>
125     *
126     * <p>
127     * It is important to note that <em>the array is not cloned</em> and that
128     * any changes to the magic number array after construction will affect the
129     * behavior of this file filter.
130     * </p>
131     *
132     * <pre>
133     * MagicNumberFileFilter javaClassFileFilter =
134     *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
135     *       (byte) 0xBA, (byte) 0xBE});
136     * </pre>
137     *
138     * @param magicNumber the magic number to look for in the file.
139     *
140     * @throws IllegalArgumentException if {@code magicNumber} is
141     *         {@code null}, or contains no bytes.
142     */
143    public MagicNumberFileFilter(final byte[] magicNumber) {
144        this(magicNumber, 0);
145    }
146
147    /**
148     * <p>
149     * Constructs a new MagicNumberFileFilter and associates it with the magic
150     * number to test for in files and the byte offset location in the file to
151     * to look for that magic number.
152     * </p>
153     *
154     * <pre>
155     * MagicNumberFileFilter tarFileFilter =
156     *     MagicNumberFileFilter(new byte[] {0x75, 0x73, 0x74, 0x61, 0x72}, 257);
157     * </pre>
158     *
159     * <pre>
160     * MagicNumberFileFilter javaClassFileFilter =
161     *     MagicNumberFileFilter(new byte[] {0xCA, 0xFE, 0xBA, 0xBE}, 0);
162     * </pre>
163     *
164     * @param magicNumbers the magic number to look for in the file.
165     * @param offset the byte offset in the file to start comparing bytes.
166     *
167     * @throws IllegalArgumentException if {@code magicNumber}
168     *         contains no bytes, or {@code offset}
169     *         is a negative number.
170     */
171    public MagicNumberFileFilter(final byte[] magicNumbers, final long offset) {
172        Objects.requireNonNull(magicNumbers, "magicNumbers");
173        if (magicNumbers.length == 0) {
174            throw new IllegalArgumentException("The magic number must contain at least one byte");
175        }
176        if (offset < 0) {
177            throw new IllegalArgumentException("The offset cannot be negative");
178        }
179
180        this.magicNumbers = magicNumbers.clone();
181        this.byteOffset = offset;
182    }
183
184    /**
185     * <p>
186     * Constructs a new MagicNumberFileFilter and associates it with the magic
187     * number to test for in files. This constructor assumes a starting offset
188     * of {@code 0}.
189     * </p>
190     *
191     * Example usage:
192     * <pre>
193     * {@code
194     * MagicNumberFileFilter xmlFileFilter =
195     *     MagicNumberFileFilter("<?xml");
196     * }
197     * </pre>
198     *
199     * @param magicNumber the magic number to look for in the file.
200     *        The string is converted to bytes using the platform default charset.
201     *
202     * @throws IllegalArgumentException if {@code magicNumber} is
203     *         {@code null} or the empty String.
204     */
205    public MagicNumberFileFilter(final String magicNumber) {
206        this(magicNumber, 0);
207    }
208
209    /**
210     * <p>
211     * Constructs a new MagicNumberFileFilter and associates it with the magic
212     * number to test for in files and the byte offset location in the file to
213     * to look for that magic number.
214     * </p>
215     *
216     * <pre>
217     * MagicNumberFileFilter tarFileFilter =
218     *     MagicNumberFileFilter("ustar", 257);
219     * </pre>
220     *
221     * @param magicNumber the magic number to look for in the file.
222     *        The string is converted to bytes using the platform default charset.
223     * @param offset the byte offset in the file to start comparing bytes.
224     *
225     * @throws IllegalArgumentException if {@code magicNumber} is
226     *         the empty String, or {@code offset} is
227     *         a negative number.
228     */
229    public MagicNumberFileFilter(final String magicNumber, final long offset) {
230        Objects.requireNonNull(magicNumber, "magicNumber");
231        if (magicNumber.isEmpty()) {
232            throw new IllegalArgumentException("The magic number must contain at least one byte");
233        }
234        if (offset < 0) {
235            throw new IllegalArgumentException("The offset cannot be negative");
236        }
237
238        this.magicNumbers = magicNumber.getBytes(Charset.defaultCharset()); // explicitly uses the platform default charset
239        this.byteOffset = offset;
240    }
241
242    /**
243     * <p>
244     * Accepts the provided file if the file contains the file filter's magic
245     * number at the specified offset.
246     * </p>
247     *
248     * <p>
249     * If any {@link IOException}s occur while reading the file, the file will
250     * be rejected.
251     * </p>
252     *
253     * @param file the file to accept or reject.
254     *
255     * @return {@code true} if the file contains the filter's magic number
256     *         at the specified offset, {@code false} otherwise.
257     */
258    @Override
259    public boolean accept(final File file) {
260        if (file != null && file.isFile() && file.canRead()) {
261            try {
262                try (RandomAccessFile randomAccessFile = RandomAccessFileMode.READ_ONLY.create(file)) {
263                    final byte[] fileBytes = IOUtils.byteArray(this.magicNumbers.length);
264                    randomAccessFile.seek(byteOffset);
265                    final int read = randomAccessFile.read(fileBytes);
266                    if (read != magicNumbers.length) {
267                        return false;
268                    }
269                    return Arrays.equals(this.magicNumbers, fileBytes);
270                }
271            }
272            catch (final IOException ignored) {
273                // Do nothing, fall through and do not accept file
274            }
275        }
276
277        return false;
278    }
279
280    /**
281     * <p>
282     * Accepts the provided file if the file contains the file filter's magic
283     * number at the specified offset.
284     * </p>
285     *
286     * <p>
287     * If any {@link IOException}s occur while reading the file, the file will
288     * be rejected.
289     * </p>
290     * @param file the file to accept or reject.
291     *
292     * @return {@code true} if the file contains the filter's magic number
293     *         at the specified offset, {@code false} otherwise.
294     * @since 2.9.0
295     */
296    @Override
297    public FileVisitResult accept(final Path file, final BasicFileAttributes attributes) {
298        if (file != null && Files.isRegularFile(file) && Files.isReadable(file)) {
299            try {
300                try (FileChannel fileChannel = FileChannel.open(file)) {
301                    final ByteBuffer byteBuffer = ByteBuffer.allocate(this.magicNumbers.length);
302                    final int read = fileChannel.read(byteBuffer);
303                    if (read != magicNumbers.length) {
304                        return FileVisitResult.TERMINATE;
305                    }
306                    return toFileVisitResult(Arrays.equals(this.magicNumbers, byteBuffer.array()));
307                }
308            }
309            catch (final IOException ignored) {
310                // Do nothing, fall through and do not accept file
311            }
312        }
313        return FileVisitResult.TERMINATE;
314    }
315
316    /**
317     * Returns a String representation of the file filter, which includes the
318     * magic number bytes and byte offset.
319     *
320     * @return a String representation of the file filter.
321     */
322    @Override
323    public String toString() {
324        final StringBuilder builder = new StringBuilder(super.toString());
325        builder.append("(");
326        builder.append(new String(magicNumbers, Charset.defaultCharset()));// TODO perhaps use hex if value is not
327                                                                           // printable
328        builder.append(",");
329        builder.append(this.byteOffset);
330        builder.append(")");
331        return builder.toString();
332    }
333}