1 /* 2 * Copyright (c) 2008, 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 /* @test 25 @bug 4153987 26 @summary Malformed surrogates should be handled by the converter in 27 substitution mode. 28 */ 29 30 import java.io.*; 31 32 public class MalformedSurrogates { 33 34 public static void main(String[] args) throws Exception { 35 36 String fe = System.getProperty("file.encoding"); 37 if ( fe.equalsIgnoreCase("UTF8") 38 || fe.equalsIgnoreCase("UTF-8") 39 || fe.equalsIgnoreCase("UTF_8")) 40 // This test is meaningless if the default charset 41 // does handle surrogates 42 return; 43 44 System.out.println("Testing string conversion..."); 45 /* Example with malformed surrogate, and an offset */ 46 String t = "abc\uD800\uDB00efgh"; 47 String t2 = t.substring(2); 48 byte[] b = t2.getBytes(); 49 System.err.println(b.length); 50 for (int i = 0; i < b.length; i++) 51 System.err.println("[" + i + "]" + "=" + (char) b[i] 52 + "=" + (int) b[i]); 53 if (b.length != 7) { 54 throw new Exception("Bad string conversion for bad surrogate"); 55 } 56 57 /* Example with a proper surrogate, no offset. Always worked */ 58 String t3 = "abc\uD800\uDC00efgh"; 59 byte[] b2 = t3.getBytes(); 60 System.out.println(b2.length); 61 for(int i = 0; i < b2.length; i++) 62 System.err.println("[" + i + "]" + "=" + (char) b2[i]); 63 if (b2.length != 8) { 64 throw new Exception("Bad string conversion for good surrogate"); 65 } 66 67 OutputStream os = new ByteArrayOutputStream(); 68 OutputStreamWriter osw = new OutputStreamWriter(os); 69 System.out.println("Testing flush...."); 70 /* Check for the case where the converter has a left over 71 high surrogate when flush is called on the converter */ 72 osw.flush(); 73 String s = "abc\uD800"; // High surrogate 74 char[] c = s.toCharArray(); 75 osw.write(s, 0, 4); 76 osw.flush(); 77 78 System.out.println("Testing convert..."); 79 /* Verify that all other characters go through */ 80 for (int k = 1; k < 65535 ; k++) { 81 osw.write("Char[" + k + "]=\"" + ((char) k) + "\""); 82 } 83 84 } 85 } | 1 /* 2 * Copyright (c) 2002, 2013, 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 /* @test 25 @bug 4153987 26 @summary Malformed surrogates should be handled by the converter in 27 substitution mode. 28 */ 29 30 import java.io.*; 31 import java.nio.charset.Charset; 32 import java.nio.charset.CharsetDecoder; 33 import java.nio.charset.CharsetEncoder; 34 import java.nio.CharBuffer; 35 import java.nio.ByteBuffer; 36 import java.nio.charset.CodingErrorAction; 37 import java.nio.charset.MalformedInputException; 38 import java.nio.charset.UnmappableCharacterException; 39 import java.util.SortedMap; 40 41 public class MalformedSurrogates { 42 private static final String PREFIX = "abc"; 43 private static final String SUFFIX = "efgh"; 44 private static final String MALFORMED_SURROGATE = PREFIX + "\uD800\uDB00" + SUFFIX; 45 private static final String NORMAL_SURROGATE = PREFIX + "\uD800\uDC00" + SUFFIX; 46 public static void main(String[] args) throws IOException { 47 SortedMap<String, Charset> map = Charset.availableCharsets(); 48 for (String name : map.keySet()) { 49 Charset charset = map.get(name); 50 boolean canEncode = charset.canEncode(); 51 CharsetEncoder en = null; 52 CharsetDecoder de = null; 53 try { 54 en = charset.newEncoder(); 55 } catch (UnsupportedOperationException ex) { 56 if (canEncode) { 57 throw ex; 58 } 59 } 60 try { 61 de = charset.newDecoder(); 62 } catch (UnsupportedOperationException ex) { 63 if (canEncode) { 64 throw ex; 65 } 66 } 67 68 if (!canEncode) { 69 continue; 70 } 71 72 boolean supported =true; 73 try { 74 supported = en.canEncode(NORMAL_SURROGATE); 75 } catch (UnsupportedOperationException ex) { 76 supported = false; 77 } finally { 78 en.reset(); 79 } 80 String[] arr = new String[] {MALFORMED_SURROGATE, NORMAL_SURROGATE}; 81 for (String surrogate : arr) { 82 canEncode = false; 83 try { 84 canEncode = en.canEncode(surrogate); 85 en.encode(CharBuffer.wrap(surrogate)); 86 } catch (MalformedInputException | UnmappableCharacterException ex) { 87 if (canEncode) { 88 throw ex; 89 } 90 } catch (UnsupportedOperationException ex) { 91 if (supported) { 92 throw ex; 93 } 94 } 95 finally { 96 en.reset(); 97 } 98 99 try (OutputStreamWriter osw = new OutputStreamWriter(new ByteArrayOutputStream(), en)) { 100 osw.write(surrogate); 101 } catch (MalformedInputException | UnmappableCharacterException ex) { 102 if (canEncode) { 103 throw ex; 104 } 105 } catch (IllegalStateException ex) { 106 if (supported) { 107 throw ex; 108 } 109 }finally { 110 en.reset(); 111 } 112 113 // Continue if charset doesn't support surrogate 114 if (!supported) { 115 continue; 116 } 117 118 String expected = NORMAL_SURROGATE; 119 120 try { 121 if (!canEncode) { 122 en.onMalformedInput(CodingErrorAction.REPLACE); 123 en.onUnmappableCharacter(CodingErrorAction.REPLACE); 124 expected = PREFIX + new String(en.replacement(), charset) + new String(en.replacement(), charset) + SUFFIX; 125 } 126 127 ByteBuffer bbuf = en.encode(CharBuffer.wrap(surrogate)); 128 CharBuffer cbuf = de.decode(bbuf); 129 if (!cbuf.toString().equals(expected)) { 130 throw new RuntimeException("charset " + charset.name() + " (en)decoded the surrogate " + surrogate + " to " + cbuf.toString() + " which is not same as the expected " + expected); 131 } 132 } finally { 133 en.reset(); 134 de.reset(); 135 } 136 137 try (ByteArrayOutputStream bos = new ByteArrayOutputStream(); 138 OutputStreamWriter osw = new OutputStreamWriter(bos, en);) { 139 osw.write(surrogate); 140 osw.flush(); 141 try (InputStreamReader isr = new InputStreamReader(new ByteArrayInputStream(bos.toByteArray()), de)) { 142 CharBuffer cbuf = CharBuffer.allocate(expected.length()); 143 isr.read(cbuf); 144 cbuf.rewind(); 145 if (!cbuf.toString().equals(expected)) { 146 throw new RuntimeException("charset " + charset.name() + " (en)decoded the surrogate " + surrogate + " to " + cbuf.toString() + " which is not same as the expected " + expected); 147 } 148 } 149 } finally { 150 en.reset(); 151 de.reset(); 152 } 153 } 154 } 155 } 156 } |