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.Serializable;
022import java.nio.ByteBuffer;
023import java.nio.channels.FileChannel;
024import java.nio.charset.Charset;
025import java.nio.file.FileVisitResult;
026import java.nio.file.Files;
027import java.nio.file.Path;
028import java.nio.file.attribute.BasicFileAttributes;
029import java.util.Arrays;
030import java.util.Objects;
031
032import org.apache.commons.io.RandomAccessFileMode;
033import org.apache.commons.io.RandomAccessFiles;
034
035/**
036 * <p>
037 * File filter for matching files containing a "magic number". A magic number
038 * is a unique series of bytes common to all files of a specific file format.
039 * For instance, all Java class files begin with the bytes
040 * {@code 0xCAFEBABE}.
041 * </p>
042 * <h2>Using Classic IO</h2>
043 * <pre>
044 * File dir = FileUtils.current();
045 * MagicNumberFileFilter javaClassFileFilter =
046 *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
047 *       (byte) 0xBA, (byte) 0xBE});
048 * String[] javaClassFiles = dir.list(javaClassFileFilter);
049 * for (String javaClassFile : javaClassFiles) {
050 *     System.out.println(javaClassFile);
051 * }
052 * </pre>
053 *
054 * <p>
055 * Sometimes, such as in the case of TAR files, the
056 * magic number will be offset by a certain number of bytes in the file. In the
057 * case of TAR archive files, this offset is 257 bytes.
058 * </p>
059 *
060 * <pre>
061 * File dir = FileUtils.current();
062 * MagicNumberFileFilter tarFileFilter =
063 *     MagicNumberFileFilter("ustar", 257);
064 * String[] tarFiles = dir.list(tarFileFilter);
065 * for (String tarFile : tarFiles) {
066 *     System.out.println(tarFile);
067 * }
068 * </pre>
069 * <h2>Using NIO</h2>
070 * <pre>
071 * final Path dir = PathUtils.current();
072 * final AccumulatorPathVisitor visitor = AccumulatorPathVisitor.withLongCounters(MagicNumberFileFilter("ustar", 257));
073 * //
074 * // Walk one directory
075 * Files.<strong>walkFileTree</strong>(dir, Collections.emptySet(), 1, visitor);
076 * System.out.println(visitor.getPathCounters());
077 * System.out.println(visitor.getFileList());
078 * //
079 * visitor.getPathCounters().reset();
080 * //
081 * // Walk directory tree
082 * Files.<strong>walkFileTree</strong>(dir, visitor);
083 * System.out.println(visitor.getPathCounters());
084 * System.out.println(visitor.getDirList());
085 * System.out.println(visitor.getFileList());
086 * </pre>
087 * <h2>Deprecating Serialization</h2>
088 * <p>
089 * <em>Serialization is deprecated and will be removed in 3.0.</em>
090 * </p>
091 *
092 * <h2>Deprecating Serialization</h2>
093 * <p>
094 * <em>Serialization is deprecated and will be removed in 3.0.</em>
095 * </p>
096 *
097 * @since 2.0
098 * @see FileFilterUtils#magicNumberFileFilter(byte[])
099 * @see FileFilterUtils#magicNumberFileFilter(String)
100 * @see FileFilterUtils#magicNumberFileFilter(byte[], long)
101 * @see FileFilterUtils#magicNumberFileFilter(String, long)
102 */
103public class MagicNumberFileFilter extends AbstractFileFilter implements Serializable {
104
105    /**
106     * The serialization version unique identifier.
107     */
108    private static final long serialVersionUID = -547733176983104172L;
109
110    /**
111     * The magic number to compare against the file's bytes at the provided
112     * offset.
113     */
114    private final byte[] magicNumbers;
115
116    /**
117     * The offset (in bytes) within the files that the magic number's bytes
118     * should appear.
119     */
120    private final long byteOffset;
121
122    /**
123     * <p>
124     * Constructs a new MagicNumberFileFilter and associates it with the magic
125     * number to test for in files. This constructor assumes a starting offset
126     * of {@code 0}.
127     * </p>
128     *
129     * <p>
130     * It is important to note that <em>the array is not cloned</em> and that
131     * any changes to the magic number array after construction will affect the
132     * behavior of this file filter.
133     * </p>
134     *
135     * <pre>
136     * MagicNumberFileFilter javaClassFileFilter =
137     *     MagicNumberFileFilter(new byte[] {(byte) 0xCA, (byte) 0xFE,
138     *       (byte) 0xBA, (byte) 0xBE});
139     * </pre>
140     *
141     * @param magicNumber the magic number to look for in the file.
142     *
143     * @throws IllegalArgumentException if {@code magicNumber} is
144     *         {@code null}, or contains no bytes.
145     */
146    public MagicNumberFileFilter(final byte[] magicNumber) {
147        this(magicNumber, 0);
148    }
149
150    /**
151     * <p>
152     * Constructs a new MagicNumberFileFilter and associates it with the magic
153     * number to test for in files and the byte offset location in the file to
154     * to look for that magic number.
155     * </p>
156     *
157     * <pre>
158     * MagicNumberFileFilter tarFileFilter =
159     *     MagicNumberFileFilter(new byte[] {0x75, 0x73, 0x74, 0x61, 0x72}, 257);
160     * </pre>
161     *
162     * <pre>
163     * MagicNumberFileFilter javaClassFileFilter =
164     *     MagicNumberFileFilter(new byte[] {0xCA, 0xFE, 0xBA, 0xBE}, 0);
165     * </pre>
166     *
167     * @param magicNumbers the magic number to look for in the file.
168     * @param offset the byte offset in the file to start comparing bytes.
169     *
170     * @throws IllegalArgumentException if {@code magicNumber}
171     *         contains no bytes, or {@code offset}
172     *         is a negative number.
173     */
174    public MagicNumberFileFilter(final byte[] magicNumbers, final long offset) {
175        Objects.requireNonNull(magicNumbers, "magicNumbers");
176        if (magicNumbers.length == 0) {
177            throw new IllegalArgumentException("The magic number must contain at least one byte");
178        }
179        if (offset < 0) {
180            throw new IllegalArgumentException("The offset cannot be negative");
181        }
182
183        this.magicNumbers = magicNumbers.clone();
184        this.byteOffset = offset;
185    }
186
187    /**
188     * <p>
189     * Constructs a new MagicNumberFileFilter and associates it with the magic
190     * number to test for in files. This constructor assumes a starting offset
191     * of {@code 0}.
192     * </p>
193     *
194     * Example usage:
195     * <pre>
196     * {@code
197     * MagicNumberFileFilter xmlFileFilter =
198     *     MagicNumberFileFilter("<?xml");
199     * }
200     * </pre>
201     *
202     * @param magicNumber the magic number to look for in the file.
203     *        The string is converted to bytes using the platform default charset.
204     *
205     * @throws IllegalArgumentException if {@code magicNumber} is
206     *         {@code null} or the empty String.
207     */
208    public MagicNumberFileFilter(final String magicNumber) {
209        this(magicNumber, 0);
210    }
211
212    /**
213     * <p>
214     * Constructs a new MagicNumberFileFilter and associates it with the magic
215     * number to test for in files and the byte offset location in the file to
216     * to look for that magic number.
217     * </p>
218     *
219     * <pre>
220     * MagicNumberFileFilter tarFileFilter =
221     *     MagicNumberFileFilter("ustar", 257);
222     * </pre>
223     *
224     * @param magicNumber the magic number to look for in the file.
225     *        The string is converted to bytes using the platform default charset.
226     * @param offset the byte offset in the file to start comparing bytes.
227     *
228     * @throws IllegalArgumentException if {@code magicNumber} is
229     *         the empty String, or {@code offset} is
230     *         a negative number.
231     */
232    public MagicNumberFileFilter(final String magicNumber, final long offset) {
233        Objects.requireNonNull(magicNumber, "magicNumber");
234        if (magicNumber.isEmpty()) {
235            throw new IllegalArgumentException("The magic number must contain at least one byte");
236        }
237        if (offset < 0) {
238            throw new IllegalArgumentException("The offset cannot be negative");
239        }
240
241        this.magicNumbers = magicNumber.getBytes(Charset.defaultCharset()); // explicitly uses the platform default charset
242        this.byteOffset = offset;
243    }
244
245    /**
246     * <p>
247     * Accepts the provided file if the file contains the file filter's magic
248     * number at the specified offset.
249     * </p>
250     *
251     * <p>
252     * If any {@link IOException}s occur while reading the file, the file will
253     * be rejected.
254     * </p>
255     *
256     * @param file the file to accept or reject.
257     *
258     * @return {@code true} if the file contains the filter's magic number
259     *         at the specified offset, {@code false} otherwise.
260     */
261    @Override
262    public boolean accept(final File file) {
263        if (file != null && file.isFile() && file.canRead()) {
264            try {
265                return RandomAccessFileMode.READ_ONLY.apply(file.toPath(),
266                        raf -> Arrays.equals(magicNumbers, RandomAccessFiles.read(raf, byteOffset, magicNumbers.length)));
267            } catch (final IOException ignored) {
268                // Do nothing, fall through and do not accept file
269            }
270        }
271        return false;
272    }
273
274    /**
275     * <p>
276     * Accepts the provided file if the file contains the file filter's magic
277     * number at the specified offset.
278     * </p>
279     * <p>
280     * If any {@link IOException}s occur while reading the file, the file will
281     * be rejected.
282     *
283     * </p>
284     * @param file the file to accept or reject.
285     * @param attributes the path's basic attributes (may be null).
286     * @return {@code true} if the file contains the filter's magic number
287     *         at the specified offset, {@code false} otherwise.
288     * @since 2.9.0
289     */
290    @Override
291    public FileVisitResult accept(final Path file, final BasicFileAttributes attributes) {
292        if (file != null && Files.isRegularFile(file) && Files.isReadable(file)) {
293            try {
294                try (FileChannel fileChannel = FileChannel.open(file)) {
295                    final ByteBuffer byteBuffer = ByteBuffer.allocate(this.magicNumbers.length);
296                    fileChannel.position(byteOffset);
297                    final int read = fileChannel.read(byteBuffer);
298                    if (read != magicNumbers.length) {
299                        return FileVisitResult.TERMINATE;
300                    }
301                    return toFileVisitResult(Arrays.equals(this.magicNumbers, byteBuffer.array()));
302                }
303            }
304            catch (final IOException ignored) {
305                // Do nothing, fall through and do not accept file
306            }
307        }
308        return FileVisitResult.TERMINATE;
309    }
310
311    /**
312     * Returns a String representation of the file filter, which includes the
313     * magic number bytes and byte offset.
314     *
315     * @return a String representation of the file filter.
316     */
317    @Override
318    public String toString() {
319        final StringBuilder builder = new StringBuilder(super.toString());
320        builder.append("(");
321        // TODO perhaps use hex if value is not printable
322        builder.append(new String(magicNumbers, Charset.defaultCharset()));
323        builder.append(",");
324        builder.append(this.byteOffset);
325        builder.append(")");
326        return builder.toString();
327    }
328}