1 /*
2 * Copyright (c) 2012, 2020, 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. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
10 *
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
16 *
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
20 *
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * questions.
24 */
25
26 package build.tools.generatelsrequivmaps;
27
28 import java.io.BufferedWriter;
29 import java.io.IOException;
30 import java.nio.charset.Charset;
31 import java.nio.file.Files;
32 import java.nio.file.Paths;
33 import java.time.ZoneId;
34 import java.time.ZonedDateTime;
35 import java.util.ArrayList;
36 import java.util.List;
37 import java.util.Locale;
38 import java.util.Map;
39 import java.util.TreeMap;
40
41 /**
42 * This tool reads the IANA Language Subtag Registry data file downloaded from
43 * http://www.iana.org/assignments/language-subtag-registry, which is specified
44 * in the command line and generates a .java source file as specified in
45 * command line. The generated .java source file contains equivalent language
46 * maps. These equivalent language maps are used by LocaleMatcher.java
47 * for the locale matching mechanism specified in RFC 4647 "Matching of Language
48 * Tags".
49 */
50 public class EquivMapsGenerator {
51
52 public static void main(String[] args) throws Exception {
53 if (args.length != 2) {
54 System.err.println("Usage: java EquivMapsGenerator"
55 + " language-subtag-registry.txt LocaleEquivalentMaps.java");
56 System.exit(1);
57 }
58 readLSRfile(args[0]);
59 generateEquivalentMap();
60 generateSourceCode(args[1]);
61 }
62
63 private static String LSRrevisionDate;
64 private static Map<String, StringBuilder> initialLanguageMap =
65 new TreeMap<>();
66 private static Map<String, StringBuilder> initialRegionVariantMap =
67 new TreeMap<>();
68
69 private static Map<String, String> sortedLanguageMap1 = new TreeMap<>();
70 private static Map<String, String[]> sortedLanguageMap2 = new TreeMap<>();
71 private static Map<String, String> sortedRegionVariantMap =
72 new TreeMap<>();
73
74 private static void readLSRfile(String filename) throws Exception {
75 String type = null;
76 String tag = null;
77 String preferred = null;
78
79 for (String line : Files.readAllLines(Paths.get(filename),
80 Charset.forName("UTF-8"))) {
81 line = line.toLowerCase(Locale.ROOT);
82 int index = line.indexOf(' ')+1;
83 if (line.startsWith("file-date:")) {
84 LSRrevisionDate = line.substring(index);
85 } else if (line.startsWith("type:")) {
86 type = line.substring(index);
87 } else if (line.startsWith("tag:") || line.startsWith("subtag:")) {
88 tag = line.substring(index);
89 } else if (line.startsWith("preferred-value:")
90 && !type.equals("extlang")) {
91 preferred = line.substring(index);
92 processDeprecatedData(type, tag, preferred);
93 } else if (line.equals("%%")) {
94 type = null;
95 tag = null;
96 }
97 }
98 }
99
100 private static void processDeprecatedData(String type,
101 String tag,
102 String preferred) {
103 StringBuilder sb;
104 if (type.equals("region") || type.equals("variant")) {
105 if (!initialRegionVariantMap.containsKey(preferred)) {
106 sb = new StringBuilder("-");
107 sb.append(preferred);
108 sb.append(",-");
109 sb.append(tag);
110 initialRegionVariantMap.put("-"+preferred, sb);
111 } else {
112 throw new RuntimeException("New case, need implementation."
113 + " A region/variant subtag \"" + preferred
114 + "\" is registered for more than one subtags.");
115 }
116 } else { // language, grandfahered, and redundant
117 if (!initialLanguageMap.containsKey(preferred)) {
118 sb = new StringBuilder(preferred);
119 sb.append(',');
120 sb.append(tag);
121 initialLanguageMap.put(preferred, sb);
122 } else {
123 sb = initialLanguageMap.get(preferred);
124 sb.append(',');
125 sb.append(tag);
126 initialLanguageMap.put(preferred, sb);
127 }
128 }
129 }
130
131 private static void generateEquivalentMap() {
132 String[] subtags;
133 for (String preferred : initialLanguageMap.keySet()) {
134 subtags = initialLanguageMap.get(preferred).toString().split(",");
135
136 if (subtags.length == 2) {
137 sortedLanguageMap1.put(subtags[0], subtags[1]);
138 sortedLanguageMap1.put(subtags[1], subtags[0]);
139 } else if (subtags.length > 2) {
140 for (int i = 0; i < subtags.length; i++) {
141 sortedLanguageMap2.put(subtags[i], createLangArray(i, subtags));
142 }
143 } else {
144 throw new RuntimeException("New case, need implementation."
145 + " A language subtag \"" + preferred
146 + "\" is registered for more than two subtags. ");
147 }
148 }
149
150 for (String preferred : initialRegionVariantMap.keySet()) {
151 subtags =
152 initialRegionVariantMap.get(preferred).toString().split(",");
153
154 sortedRegionVariantMap.put(subtags[0], subtags[1]);
155 sortedRegionVariantMap.put(subtags[1], subtags[0]);
156 }
157
158 }
159
160 /* create the array of subtags excluding the subtag at index location */
161 private static String[] createLangArray(int index, String[] subtags) {
162 List<String> list = new ArrayList<>();
163 for (int i = 0; i < subtags.length; i++) {
164 if (i != index) {
165 list.add(subtags[i]);
166 }
167 }
168 return list.toArray(new String[list.size()]);
169 }
170
171 private static String generateValuesString(String[] values) {
172 String outputStr = "";
173 for (int i = 0; i < values.length; i++) {
174 if (i != values.length - 1) {
175 outputStr = outputStr + "\"" + values[i] + "\", ";
176 } else {
177 outputStr = outputStr + "\"" + values[i] + "\"";
178 }
179
180 }
181 return outputStr;
182 }
183
184 private static final String COPYRIGHT = "/*\n"
185 + " * Copyright (c) 2012, %d, Oracle and/or its affiliates. All rights reserved.\n"
186 + " * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.\n"
187 + " *\n"
188 + " * This code is free software; you can redistribute it and/or modify it\n"
189 + " * under the terms of the GNU General Public License version 2 only, as\n"
190 + " * published by the Free Software Foundation. Oracle designates this\n"
191 + " * particular file as subject to the \"Classpath\" exception as provided\n"
192 + " * by Oracle in the LICENSE file that accompanied this code.\n"
193 + " *\n"
194 + " * This code is distributed in the hope that it will be useful, but WITHOUT\n"
195 + " * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or\n"
196 + " * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License\n"
197 + " * version 2 for more details (a copy is included in the LICENSE file that\n"
198 + " * accompanied this code).\n"
199 + " *\n"
200 + " * You should have received a copy of the GNU General Public License version\n"
201 + " * 2 along with this work; if not, write to the Free Software Foundation,\n"
202 + " * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.\n"
203 + " *\n"
204 + " * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA\n"
205 + " * or visit www.oracle.com if you need additional information or have any\n"
206 + " * questions.\n"
207 + "*/\n\n";
208
209 private static final String headerText =
210 "package sun.util.locale;\n\n"
211 + "import java.util.HashMap;\n"
212 + "import java.util.Map;\n\n"
213 + "final class LocaleEquivalentMaps {\n\n"
214 + " static final Map<String, String> singleEquivMap;\n"
215 + " static final Map<String, String[]> multiEquivsMap;\n"
216 + " static final Map<String, String> regionVariantEquivMap;\n\n"
217 + " static {\n"
218 + " singleEquivMap = new HashMap<>(";
219
220 private static final String footerText =
221 " }\n\n"
222 + "}";
223
224 private static String getOpenJDKCopyright() {
225 int year = ZonedDateTime.now(ZoneId
226 .of("America/Los_Angeles")).getYear();
227 return String.format(Locale.US, COPYRIGHT, year);
228 }
229
230 /**
231 * The input lsr data file is in UTF-8, so theoretically for the characters
232 * beyond US-ASCII, the generated Java String literals need to be Unicode
233 * escaped (\\uXXXX) while writing to a file. But as of now, there is not
234 * the case since we don't use "description", "comment" or alike.
235 */
236 private static void generateSourceCode(String fileName) {
237
238 try (BufferedWriter writer = Files.newBufferedWriter(
239 Paths.get(fileName))) {
240 writer.write(getOpenJDKCopyright());
241 writer.write(headerText
242 + (int)(sortedLanguageMap1.size() / 0.75f + 1) + ");\n"
243 + " multiEquivsMap = new HashMap<>("
244 + (int)(sortedLanguageMap2.size() / 0.75f + 1) + ");\n"
245 + " regionVariantEquivMap = new HashMap<>("
246 + (int)(sortedRegionVariantMap.size() / 0.75f + 1) + ");\n\n"
247 + " // This is an auto-generated file and should not be manually edited.\n"
248 + " // LSR Revision: " + LSRrevisionDate);
249 writer.newLine();
250
251 for (String key : sortedLanguageMap1.keySet()) {
252 String value = sortedLanguageMap1.get(key);
253 writer.write(" singleEquivMap.put(\""
254 + key + "\", \"" + value + "\");");
255 writer.newLine();
256 }
257
258 writer.newLine();
259 for (String key : sortedLanguageMap2.keySet()) {
260 String[] values = sortedLanguageMap2.get(key);
261
262 if (values.length >= 2) {
263 writer.write(" multiEquivsMap.put(\""
264 + key + "\", new String[] {"
265 + generateValuesString(values) + "});");
266 writer.newLine();
267 }
268 }
269
270 writer.newLine();
271 for (String key : sortedRegionVariantMap.keySet()) {
272 String value = sortedRegionVariantMap.get(key);
273 writer.write(" regionVariantEquivMap.put(\""
274 + key + "\", \"" + value + "\");");
275 writer.newLine();
276 }
277
278 writer.write(footerText);
279 } catch (IOException ex) {
280 ex.printStackTrace(System.err);
281 System.exit(1);
282 }
283
284 }
285
286 }