1 /*
2 * Copyright (c) 2007, 2012, 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 /*
25 * @test
26 * @bug 5030265
27 * @compile -XDignore.symbol.file UnicodeTest.java
28 * @run main/othervm UnicodeTest
29 * @summary Verify that the J2RE can handle all legal Unicode characters
30 * in class names unless limited by the file system encoding
31 * or the encoding used for command line arguments.
32 * @author Norbert Lindenberg, ksrini
33 */
34
35 /*
36 * This class creates Java source files using Unicode characters
37 * that test the limits of what's possible
38 * - in situations where the platform encoding imposes limits
39 * (command line arguments, non-Unicode file system)
40 * - in situations where full Unicode is supported
41 * (file system access in UTF-8 locales and on Windows 2000++,
42 * jar file contents)
43 *
44 * This test needs to be run in othervm as the locale is reset.
45 */
46
47 import java.io.File;
48 import java.io.FileOutputStream;
49 import java.io.OutputStreamWriter;
50 import java.nio.charset.Charset;
51 import java.util.Locale;
52
53 public class UnicodeTest extends TestHelper {
54 static final File UnicodeTestSrc = new File("UnicodeTest-src");
55 static final File UnicodeTestClasses = new File("UnicodeTest-classes");
56 static final String UnicodeTestJarName = "UnicodeTest" + JAR_FILE_EXT;
57 static final File UnicodeTestJar = new File(UnicodeTestJarName);
58 static final File SolarisUnicodeTestJar = new File(TEST_SOURCES_DIR,
59 UnicodeTestJarName);
60
61 /*
62 * the main method is a port of the shell based test to a java, this
63 * eliminates the need for MKS on windows, thus we can rely on consistent
64 * results regardless of the shell being used.
65 */
66 public static void main(String... args) throws Exception {
67 System.out.println("creating test source files");
68 UnicodeTestSrc.mkdirs();
69 UnicodeTestClasses.mkdirs();
70 String classname = generateSources();
71 File javaFile = new File(UnicodeTestSrc, classname + JAVA_FILE_EXT);
72 System.out.println("building test apps");
73 compile("-encoding", "UTF-8",
74 "-sourcepath", UnicodeTestSrc.getAbsolutePath(),
75 "-d", UnicodeTestClasses.getAbsolutePath(),
76 javaFile.getAbsolutePath());
77
78 createJar("-cvfm", UnicodeTestJar.getAbsolutePath(),
79 new File(UnicodeTestSrc, "MANIFEST.MF").getAbsolutePath(),
80 "-C", UnicodeTestClasses.getAbsolutePath(), ".");
81
82 if (!UnicodeTestJar.exists()) {
83 throw new Error("failed to create " + UnicodeTestJar.getAbsolutePath());
84 }
85
86 System.out.println("running test app using class file");
87 TestResult tr = doExec(javaCmd,
88 "-cp", UnicodeTestClasses.getAbsolutePath(), classname);
89 if (!tr.isOK()) {
90 System.out.println(tr);
91 throw new RuntimeException("test fails");
92 }
93
94 System.out.println("delete generated files with non-ASCII names");
95 recursiveDelete(UnicodeTestSrc);
96 recursiveDelete(UnicodeTestClasses);
97
98 /*
99 * test in whatever the default locale is
100 */
101 runJarTests();
102
103 /*
104 * if the Japanese locale is available, test in that locale as well
105 */
106 if (setLocale(Locale.JAPANESE)) {
107 runJarTests();
108 }
109
110 /*
111 * if we can switch to a C locale, then test whether jar files with
112 * non-ASCII characters in the manifest still work in this crippled
113 * environment
114 */
115 if (setLocale(Locale.ENGLISH)) {
116 runJarTests();
117 }
118 // thats it we are outta here
119 }
120
121 static void runJarTests() {
122 System.out.println("running test app using newly built jar file in " +
123 Locale.getDefault());
124 runTest(UnicodeTestJar);
125
126 System.out.println("running test app using jar file " +
127 "(built with Solaris UTF-8 locale) in " + Locale.getDefault());
128 runTest(SolarisUnicodeTestJar);
129 }
130
131 static void runTest(File testJar) {
132 TestResult tr = doExec(javaCmd, "-jar", testJar.getAbsolutePath());
133 if (!tr.isOK()) {
134 System.out.println(tr);
135 throw new RuntimeException("test fails");
136 }
137 }
138
139 static boolean setLocale(Locale desired) {
140 if (Locale.getDefault().equals(desired)) {
141 return true; // already set nothing more
142 }
143 for (Locale l : Locale.getAvailableLocales()) {
144 if (l == desired) {
145 Locale.setDefault(l);
146 return true;
147 }
148 }
149 return false;
150 }
151
152 static String generateSources() throws Exception {
153 String commandLineClassNameSuffix = commandLineClassNameSuffix();
154 String commandLineClassName = "ClassA" + commandLineClassNameSuffix;
155 String manifestClassName = "ClassB" +
156 (hasUnicodeFileSystem() ? unicode : commandLineClassNameSuffix);
157
158 generateSource(commandLineClassName, manifestClassName);
159 generateSource(manifestClassName, commandLineClassName);
160 generateManifest(manifestClassName);
161 return commandLineClassName;
162 }
163
164 private static final String defaultEncoding = Charset.defaultCharset().name();
165
166 // language names taken from java.util.Locale.getDisplayLanguage for the respective language
167 private static final String arabic = "\u0627\u0644\u0639\u0631\u0628\u064a\u0629";
168 private static final String s_chinese = "\u4e2d\u6587";
169 private static final String t_chinese = "\u4e2d\u6587";
170 private static final String russian = "\u0440\u0443\u0441\u0441\u043A\u0438\u0439";
171 private static final String hindi = "\u0939\u093f\u0902\u0926\u0940";
172 private static final String greek = "\u03b5\u03bb\u03bb\u03b7\u03bd\u03b9\u03ba\u03ac";
173 private static final String hebrew = "\u05e2\u05d1\u05e8\u05d9\u05ea";
174 private static final String japanese = "\u65e5\u672c\u8a9e";
175 private static final String korean = "\ud55c\uad6d\uc5b4";
176 private static final String lithuanian = "Lietuvi\u0173";
177 private static final String czech = "\u010de\u0161tina";
178 private static final String turkish = "T\u00fcrk\u00e7e";
179 private static final String spanish = "espa\u00f1ol";
180 private static final String thai = "\u0e44\u0e17\u0e22";
181 private static final String unicode = arabic + s_chinese + t_chinese
182 + russian + hindi + greek + hebrew + japanese + korean
183 + lithuanian + czech + turkish + spanish + thai;
184
185 private static String commandLineClassNameSuffix() {
186
187 // Mapping from main platform encodings to language names
188 // for Unix and Windows, respectively. Use empty suffix
189 // for Windows encodings where OEM encoding differs.
190 // Use null if encoding isn't used.
191 String[][] names = {
192 { "UTF-8", unicode, "" },
193 { "windows-1256", null, "" },
194 { "iso-8859-6", arabic, null },
195 { "GBK", s_chinese, s_chinese },
196 { "GB18030", s_chinese, s_chinese },
197 { "GB2312", s_chinese, null },
198 { "x-windows-950", null, t_chinese },
199 { "x-MS950-HKSCS", null, t_chinese },
200 { "x-euc-tw", t_chinese, null },
201 { "Big5", t_chinese, null },
202 { "Big5-HKSCS", t_chinese, null },
203 { "windows-1251", null, "" },
204 { "iso-8859-5", russian, null },
205 { "koi8-r", russian, null },
206 { "windows-1253", null, "" },
207 { "iso-8859-7", greek, null },
208 { "windows-1255", null, "" },
209 { "iso8859-8", hebrew, null },
210 { "windows-31j", null, japanese },
211 { "x-eucJP-Open", japanese, null },
212 { "x-EUC-JP-LINUX", japanese, null },
213 { "x-pck", japanese, null },
214 { "x-windows-949", null, korean },
215 { "euc-kr", korean, null },
216 { "windows-1257", null, "" },
217 { "iso-8859-13", lithuanian, null },
218 { "windows-1250", null, "" },
219 { "iso-8859-2", czech, null },
220 { "windows-1254", null, "" },
221 { "iso-8859-9", turkish, null },
222 { "windows-1252", null, "" },
223 { "iso-8859-1", spanish, null },
224 { "iso-8859-15", spanish, null },
225 { "x-windows-874", null, thai },
226 { "tis-620", thai, null },
227 };
228
229 int column = isWindows ? 2 : 1;
230 for (int i = 0; i < names.length; i++) {
231 if (names[i][0].equalsIgnoreCase(defaultEncoding)) {
232 return names[i][column];
233 }
234 }
235 return "";
236 }
237
238 private static boolean hasUnicodeFileSystem() {
239 return (isWindows) ? true : defaultEncoding.equalsIgnoreCase("UTF-8");
240 }
241
242 private static void generateSource(String thisClass, String otherClass) throws Exception {
243 File file = new File(UnicodeTestSrc, thisClass + JAVA_FILE_EXT);
244 OutputStreamWriter out = new OutputStreamWriter(new FileOutputStream(file), "UTF-8");
245 out.write("public class " + thisClass + " {\n");
246 out.write(" public static void main(String[] args) {\n");
247 out.write(" if (!" + otherClass + "." + otherClass.toLowerCase() + "().equals(\"" + otherClass + "\")) {\n");
248 out.write(" throw new RuntimeException();\n");
249 out.write(" }\n");
250 out.write(" }\n");
251 out.write(" public static String " + thisClass.toLowerCase() + "() {\n");
252 out.write(" return \"" + thisClass + "\";\n");
253 out.write(" }\n");
254 out.write("}\n");
255 out.close();
256 }
257
258 private static void generateManifest(String mainClass) throws Exception {
259 File file = new File(UnicodeTestSrc, "MANIFEST.MF");
260 FileOutputStream out = new FileOutputStream(file);
261 out.write("Manifest-Version: 1.0\n".getBytes("UTF-8"));
262 // Header lines are limited to 72 bytes.
263 // The manifest spec doesn't say we have to break at character boundaries,
264 // so we rudely break at byte boundaries.
265 byte[] headerBytes = ("Main-Class: " + mainClass + "\n").getBytes("UTF-8");
266 if (headerBytes.length <= 72) {
267 out.write(headerBytes);
268 } else {
269 out.write(headerBytes, 0, 72);
270 int start = 72;
271 while (headerBytes.length > start) {
272 out.write((byte) '\n');
273 out.write((byte) ' ');
274 int count = Math.min(71, headerBytes.length - start);
275 out.write(headerBytes, start, count);
276 start += count;
277 }
278 }
279 out.close();
280 }
281 }