1 /*
   2  * Copyright (c) 2018, Oracle and/or its affiliates. All rights reserved.
   3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   4  *
   5  * This code is free software; you can redistribute it and/or modify it
   6  * under the terms of the GNU General Public License version 2 only, as
   7  * published by the Free Software Foundation.
   8  *
   9  * This code is distributed in the hope that it will be useful, but WITHOUT
  10  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  11  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  12  * version 2 for more details (a copy is included in the LICENSE file that
  13  * accompanied this code).
  14  *
  15  * You should have received a copy of the GNU General Public License version
  16  * 2 along with this work; if not, write to the Free Software Foundation,
  17  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
  18  *
  19  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
  20  * or visit www.oracle.com if you need additional information or have any
  21  * questions.
  22  */
  23 
  24 import org.testng.Assert;
  25 import org.testng.annotations.AfterClass;
  26 import org.testng.annotations.BeforeClass;
  27 import org.testng.annotations.DataProvider;
  28 import org.testng.annotations.Test;
  29 
  30 import java.io.File;
  31 import java.io.IOException;
  32 import java.nio.charset.StandardCharsets;
  33 import java.nio.file.FileSystem;
  34 import java.nio.file.Files;
  35 import java.nio.file.Path;
  36 import java.nio.file.Paths;
  37 import java.nio.file.spi.FileSystemProvider;
  38 import java.util.Map;
  39 
  40 import static java.nio.file.StandardCopyOption.REPLACE_EXISTING;
  41 
  42 /* @test
  43  * @bug 8202285
  44  * @build Mismatch
  45  * @run testng Mismatch
  46  * @summary Unit test for the Files.mismatch method.
  47  */
  48 public class Mismatch {
  49     // the standard buffer size
  50     final static int BUFFER_SIZE = 8192;
  51 
  52     private static final int MISMATCH_NO = -1;
  53 
  54     // Map to be used for creating a ZIP archive
  55     private static final Map<String, String> ZIPFS_MAP = Map.of("create", "true");
  56 
  57     // temporary test directory where all test files will be created
  58     Path testDir;
  59 
  60     @BeforeClass
  61     void setup() throws IOException {
  62         testDir = Files.createTempDirectory("testMismatch");
  63     }
  64 
  65     @AfterClass
  66     void cleanup() throws IOException {
  67         // clean up files created under the test directory
  68         Files.walk(testDir).map(Path::toFile).forEach(File::delete);
  69         Files.deleteIfExists(testDir);
  70     }
  71 
  72     /*
  73      * DataProvider for mismatch test. Provides the following fields:
  74      * path1 -- the path to a file
  75      * path2 -- the path to another file
  76      * expected -- expected result of the mismatch method
  77      * note -- a note about the test
  78      */
  79     @DataProvider(name = "testMismatch")
  80     public Object[][] getDataForMismatch() throws IOException {
  81         // an non-existent file
  82         Path foo = Paths.get("nonexistentfile");
  83 
  84         /**
  85          * File path naming convention:
  86          * "test" + file size + [abm] [+ position of a modified char] + [ab]
  87          * where:
  88          * a or b -- is used to differentiate two files of the same size.
  89          * m -- indicates the file is modified at the position specified after it
  90          */
  91 
  92         // create empty files
  93         int size = 0;
  94         Path test0a = createASCIIFile(testDir, "test0a", 0, -1, ' ');
  95         Path test0b = createASCIIFile(testDir, "test0b", 0, -1, ' ');
  96 
  97         /**
  98          * Since the Impl uses a standard buffer of 8192, the test files are created
  99          * with sizes <= and > 8192, either multiples of the buffer size, or random.
 100          * The files are then altered at the begining (0), end (size), and a random
 101          * position.
 102          */
 103         size = 147;
 104         Path test147a = createASCIIFile(testDir, "test147a", size, -1, ' ');
 105         Path test147b = createASCIIFile(testDir, "test147b", size, -1, ' ');
 106         Path test147m0 = createASCIIFile(testDir, "test147m0", size, 0, '!');
 107         Path test147m70 = createASCIIFile(testDir, "test147m70", size, 70, '@');
 108         Path test147m146 = createASCIIFile(testDir, "test147m146", size, size - 1, '$');
 109 
 110         size = 1024;
 111         Path test1024a = createASCIIFile(testDir, "test1024a", size, -1, ' ');
 112         Path test1024b = createASCIIFile(testDir, "test1024b", size, -1, ' ');
 113         Path test1024m512 = createASCIIFile(testDir, "test1024m512", size, size >> 1, '@');
 114         Path test1024m1023 = createASCIIFile(testDir, "test1024m1023", size, size - 1, '$');
 115 
 116         size = BUFFER_SIZE;
 117         Path test8192a = createASCIIFile(testDir, "test8192a", size, -1, ' ');
 118         Path test8192b = createASCIIFile(testDir, "test8192b", size, -1, ' ');
 119         Path test8192m4096 = createASCIIFile(testDir, "test8192m4096", size, size >> 1, '%');
 120         Path test8192m8191 = createASCIIFile(testDir, "test8192m8191", size, size - 1, '$');
 121 
 122 
 123         // create files with size several times > BUFFER_SIZE to be used for tests that verify
 124         // the situations where they are read into full buffers a few times
 125         size = BUFFER_SIZE << 3;
 126         Path test65536a = createASCIIFile(testDir, "test65536a", size, -1, ' ');
 127         Path test65536b = createASCIIFile(testDir, "test65536b", size, -1, ' ');
 128         Path test65536m0 = createASCIIFile(testDir, "test65536m0", size, 0, '!');
 129         Path test65536m32768 = createASCIIFile(testDir, "test65536m32768", size, size >> 1, '%');
 130         Path test65536m65535 = createASCIIFile(testDir, "test65536m65535", size, size - 1, '$');
 131 
 132         // create files with sizes that will be iterated several times with full buffers, and
 133         // then a partial one at the last
 134         size = 70025;
 135         Path test70025a = createASCIIFile(testDir, "test70025a", size, -1, ' ');
 136         Path test70025b = createASCIIFile(testDir, "test70025b", size, -1, ' ');
 137         Path test70025m8400 = createASCIIFile(testDir, "test70025m8400", size, 8400, '@');
 138         Path test70025m35000 = createASCIIFile(testDir, "test70025m35000", size, 35000, '%');
 139         Path test70025m70024 = createASCIIFile(testDir, "test70025m70024", size, 70024, '$');
 140 
 141         // create larger files with >= 1048576. The mismatching will be similar. These are just
 142         // tests to exercise the process with larger files
 143         size = 1048576;
 144         Path test1048576a = createASCIIFile(testDir, "test1048576a", size, -1, ' ');
 145 
 146         size = 1065000;
 147         Path test1065000m532500 = createASCIIFile(testDir, "test1065000m532500", size, size >> 1, '%');
 148         Path test1065000m1064999 = createASCIIFile(testDir, "test1065000m1064999", size, 1064999, '$');
 149 
 150         return new Object[][]{
 151             // Spec Case 1: the two paths locate the same file , even if one does not exist
 152             {foo, foo, MISMATCH_NO, "Same file, no mismatch"},
 153             {test1024a, test1024a, MISMATCH_NO, "Same file, no mismatch"},
 154 
 155             // Spec Case 2:  The two files are the same size, and every byte in the first file
 156             // is identical to the corresponding byte in the second file.
 157             {test0a, test0b, MISMATCH_NO, "Sizes == 0, no mismatch"},
 158             {test147a, test147b, MISMATCH_NO, "size = 147 < buffer = 8192, no mismatch"},
 159             {test1024a, test1024b, MISMATCH_NO, "size = 1024 < buffer = 8192, no mismatch"},
 160             {test8192a, test8192b, MISMATCH_NO, "size = 8192 = buffer = 8192, no mismatch"},
 161             {test65536a, test65536b, MISMATCH_NO, "read 8 * full buffer, no mismatch"},
 162             {test70025a, test70025b, MISMATCH_NO, "read 8 * full buffer plus a partial buffer, no mismatch"},
 163 
 164 
 165             /**
 166              * Spec Case 3: the value returned is the position of the first mismatched byte
 167              * Impl: the impl uses a buffer 8192. The testcases below covers a range of files
 168              * with sizes <= and > the buffer size. The last buffer is either full or partially full.
 169             */
 170 
 171             // edge case, one of the file sizes is zero
 172             // also covers Spec Case 4 and 6
 173             {test147a, test147m0, 0, "mismatch = 0 (at the beginning)"},
 174             {test65536m0, test65536a, 0, "mismatch = 0 (at the beginning)"},
 175 
 176             /**
 177              * Compares files of equal sizes
 178             */
 179             // small files
 180             {test147a, test147m70, 70, "read one partial buffer, mismatch = 70"},
 181             {test147a, test147m146, 146, "read one partial buffer, mismatch = 146 (end)"},
 182             {test1024a, test1024m512, 512, "read one partial buffer, mismatch = 512"},
 183             {test1024a, test1024m1023, 1023, "read one partial buffer, mismatch = 1023 (end)"},
 184 
 185             // file size >= Impl's Buffer Size
 186             {test8192a, test8192m4096, 4096, "read one buffer, mismatch = 4096 "},
 187             {test8192a, test8192m8191, 8191, "read one buffer, mismatch = 8191 (at the end)"},
 188 
 189             // file size = n * Impl's Buffer Size
 190             {test65536a, test65536m32768, 32768, "read through half of the file, mismatch = 32768"},
 191             {test65536a, test65536m65535, 65535, "read through the whole file, mismatch = 65535 (at the end)"},
 192 
 193             // file size = n * Impl's Buffer Size + x
 194             {test70025a, test70025m8400, 8400, "mismatch in the 2nd buffer, mismatch = 8400"},
 195             {test70025a, test70025m35000, 35000, "read about half of the file, mismatch = 35000"},
 196             {test70025a, test70025m70024, 70024, "read through the whole file, mismatch = 70024 (at the end)"},
 197 
 198             /**
 199              * Compares files of unequal sizes
 200             */
 201             {test8192m8191, test70025m35000, 8191, "mismatch at the end of the 1st file/buffer, mismatch = 8191"},
 202             {test65536m32768, test70025m8400, 8400, "mismatch in the 2nd buffer, mismatch = 8400"},
 203             {test70025m70024, test1065000m532500, 70024, "mismatch at the end of the 1st file, mismatch = 70024"},
 204 
 205             /**
 206              * Spec Case 4:  returns the size of the smaller file (in bytes) when the files are
 207              * different sizes and every byte of the smaller file is identical to the corresponding
 208              * byte of the larger file.
 209              * Impl: similar to case 3, covers a range of file sizes
 210             */
 211             {test147a, test1024a, 147, "mismatch is the length of the smaller file: 147"},
 212             {test1024a, test8192a, 1024, "mismatch is the length of the smaller file: 1024"},
 213             {test1024a, test65536a, 1024, "mismatch is the length of the smaller file: 1024"},
 214             {test8192a, test65536a, 8192, "mismatch is the length of the smaller file: 8192"},
 215             {test70025a, test65536a, 65536, "mismatch is the length of the smaller file: 65536"},
 216             {test1048576a, test1065000m1064999, 1048576, "mismatch is the length of the smaller file: 1048576"},
 217 
 218             // Spec Case 5: This method is always reflexive (for Path f , mismatch(f,f) returns -1L)
 219             // See tests for Spec Case 1.
 220 
 221             // Spec Case 6: If the file system and files remain static, then this method is symmetric
 222             // (for two Paths f and g, mismatch(f,g) will return the same value as mismatch(g,f)).
 223             // The following tests are selected from tests for Spec Case 3 with the order of
 224             // file paths switched, the returned values are the same as those for Case 3:
 225             {test147m70, test147a, 70, "read one partial buffer, mismatch = 70"},
 226             {test147m146, test147a, 146, "read one partial buffer, mismatch = 146 (end)"},
 227             {test1024m512, test1024a, 512, "read one partial buffer, mismatch = 512"},
 228             {test1024m1023, test1024a, 1023, "read one partial buffer, mismatch = 1023 (end)"},
 229 
 230             {test70025m35000, test8192m8191, 8191, "mismatch at the end of the 1st file/buffer, mismatch = 8191"},
 231             {test70025m8400, test65536m32768, 8400, "mismatch in the 2nd buffer, mismatch = 8400"},
 232             {test1065000m532500, test70025m70024, 70024, "mismatch at the end of the 1st file, mismatch = 70024"},
 233         };
 234     }
 235 
 236     /*
 237      * DataProvider for mismatch tests involving ZipFS using a few test cases selected
 238      * from those of the original mismatch tests.
 239      */
 240     @DataProvider(name = "testMismatchZipfs")
 241     public Object[][] getDataForMismatchZipfs() throws IOException {
 242         Path test1200 = createASCIIFile(testDir, "test1200", 1200, -1, ' ');
 243         Path test9500 = createASCIIFile(testDir, "test9500", 9500, -1, ' ');
 244         Path test9500m4200 = createASCIIFile(testDir, "test9500m4200", 9500, 4200, '!');
 245         Path test80025 = createASCIIFile(testDir, "test80025", 80025, -1, ' ');
 246         Path test1028500 = createASCIIFile(testDir, "test1028500", 1028500, -1, ' ');
 247         return new Object[][]{
 248             {test1200, test1200, MISMATCH_NO, "Compares the file and its copy in zip, no mismatch"},
 249             {test9500, test9500m4200, 4200,
 250                 "Compares a copy of test9500m4200 in zip with test9500, shall return 4200"},
 251             {test80025, test1028500, 80025, "mismatch is the length of the smaller file: 80025"},
 252         };
 253     }
 254 
 255     /*
 256      * DataProvider for verifying null handling.
 257      */
 258     @DataProvider(name = "testFileNull")
 259     public Object[][] getDataForNull() throws IOException {
 260         Path test = createASCIIFile(testDir, "testNonNull", 2200, -1, ' ');
 261         return new Object[][]{
 262             {(Path)null, (Path)null},
 263             {(Path)null, test},
 264             {test, (Path)null},
 265         };
 266     }
 267 
 268     /*
 269      * DataProvider for verifying how the mismatch method handles the situation
 270      * when one or both files do not exist.
 271      */
 272     @DataProvider(name = "testFileNotExist")
 273     public Object[][] getDataForFileNotExist() throws IOException {
 274         Path test = createASCIIFile(testDir, "testFileNotExist", 3200, -1, ' ');
 275         return new Object[][]{
 276             {Paths.get("foo"), Paths.get("bar")},
 277             {Paths.get("foo"), test},
 278             {test, Paths.get("bar")},
 279         };
 280     }
 281 
 282     /**
 283      * Tests the mismatch method. Refer to the dataProvider testMismatch for more
 284      * details about the cases.
 285      * @param path the path to a file
 286      * @param path2 the path to another file
 287      * @param expected the expected result
 288      * @param msg the message about the test
 289      * @throws IOException if the test fails
 290      */
 291     @Test(dataProvider = "testMismatch", priority = 0)
 292     public void testMismatch(Path path, Path path2, long expected, String msg)
 293         throws IOException {
 294         Assert.assertEquals(Files.mismatch(path, path2), expected, msg);
 295     }
 296 
 297     /**
 298      * Tests the mismatch method by comparing files with those in a ZIP file.
 299      * @param path the path to a file
 300      * @param path2 the path to another file to be added into a ZIP file
 301      * @param expected the expected result
 302      * @param msg the message about the test
 303      * @throws IOException if the test fails
 304      */
 305     @Test(dataProvider = "testMismatchZipfs", priority = 1)
 306     public void testMismatchZipfs(Path path, Path path2, long expected, String msg)
 307         throws IOException {
 308         Path zipPath = Paths.get(testDir.toString(), "TestWithFSZip.zip");
 309         try (FileSystem fs = getZipFSProvider().newFileSystem(zipPath, ZIPFS_MAP)) {
 310             Path copy = fs.getPath(path.getFileName().toString());
 311             Files.copy(path, copy, REPLACE_EXISTING);
 312 
 313             if (path2 == null) {
 314                 Assert.assertEquals(Files.mismatch(copy, path), expected, msg);
 315             } else {
 316                 Assert.assertEquals(Files.mismatch(copy, path2), expected, msg);
 317             }
 318         }
 319     }
 320 
 321     /**
 322      * Verifies that NullPointerException is thrown when one or both files are null.
 323      * @param path the path to a file
 324      * @param path2 the path to another file
 325      * @throws NullPointerException as expected
 326      */
 327     @Test(dataProvider = "testFileNull", priority = 2, expectedExceptions = NullPointerException.class)
 328     public void testMismatchNull(Path path, Path path2) throws Exception {
 329         long result = Files.mismatch(path, path2);
 330     }
 331 
 332     /**
 333      * Verifies that IOException is thrown when one or both files do not exist.
 334      * @param path the path to a file
 335      * @param path2 the path to another file
 336      * @throws IOException as expected
 337      */
 338     @Test(dataProvider = "testFileNotExist", priority = 2, expectedExceptions = IOException.class)
 339     public void testMismatchNotExist(Path path, Path path2) throws IOException {
 340         long result = Files.mismatch(path, path2);
 341     }
 342 
 343     /**
 344      * Creates a file with ASCII content with one character altered
 345      * at the specified position.
 346      *
 347      * Note: Files.mismatch method does a byte-by-byte comparison. ASCII files
 348      * are sufficient for verifying the feature.
 349      *
 350      * @param dir the directory in which the file is to be created
 351      * @param purpose the purpose of the file
 352      * @param size the size of the file
 353      * @param pos the position where the alternative char is to be added. If it
 354      *            is smaller than zero, no alternation shall be made.
 355      * @param c the character
 356      * @return path of the created file
 357      * @throws IOException
 358      */
 359     private static Path createASCIIFile(Path dir, String purpose, int size, int pos,
 360                                         char c) throws IOException {
 361         Path path = Files.createFile(Paths.get(dir.toString(), purpose + ".txt"));
 362         if (size > 0) {
 363             writeASCIIFile(path, size, pos, c);
 364         }
 365         return path;
 366     }
 367 
 368     private static void writeASCIIFile(Path path, int size, int pos, char c)
 369         throws IOException {
 370         byte[] a = createASCIIArray(size);
 371         if (pos >= 0) a[pos] = (byte)(c & 0xFF); // US_ASCII char only, may cast directly
 372         Files.write(path, a);
 373     }
 374 
 375     private static byte[] createASCIIArray(int length) {
 376         byte[] bytes = "ABCDEFGHIJKLMNOPQRSTUVWXYZ abcdefghijklmnopqrstuvwxyz 0123456789 \n"
 377             .getBytes(StandardCharsets.US_ASCII);
 378         byte[] a = new byte[length];
 379         fillArray(bytes, a);
 380         return a;
 381     }
 382 
 383     private static FileSystemProvider getZipFSProvider() {
 384         for (FileSystemProvider provider : FileSystemProvider.installedProviders()) {
 385             if ("jar".equals(provider.getScheme())) {
 386                 return provider;
 387             }
 388         }
 389         return null;
 390     }
 391 
 392     /**
 393      * Fills the destination array by copying the source array repeatedly until 
 394      * it is completely filled.
 395      *
 396      * @param src the source array
 397      * @param dest the destination array
 398      */
 399     public static void fillArray(byte[] src, byte[] dest) {
 400         int bLen = src.length;
 401         int space = dest.length;
 402         int iteration = 0;
 403 
 404         while (space > 0) {
 405             if (space >= bLen) {
 406                 System.arraycopy(src, 0, dest, iteration++ * bLen, bLen);
 407                 space -= bLen;
 408             } else {
 409                 System.arraycopy(src, 0, dest, iteration++ * bLen, space);
 410                 break;
 411             }
 412         }
 413     }
 414 }