1 /*
2 * Copyright (c) 2016, 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 package jdk.tools.jlink.internal.plugins;
26
27 import java.io.ByteArrayInputStream;
28 import java.util.Arrays;
29 import java.util.Collections;
30 import java.util.IllformedLocaleException;
31 import java.util.Locale;
32 import java.util.List;
33 import java.util.Map;
34 import java.util.Optional;
35 import java.util.Set;
36 import java.util.function.Predicate;
37 import java.util.regex.Pattern;
38 import java.util.stream.Collectors;
39 import java.util.stream.IntStream;
40 import java.util.stream.Stream;
41 import jdk.internal.org.objectweb.asm.ClassReader;
42 import jdk.tools.jlink.internal.ResourcePrevisitor;
43 import jdk.tools.jlink.internal.StringTable;
44 import jdk.tools.jlink.plugin.LinkModule;
45 import jdk.tools.jlink.plugin.ModuleEntry;
46 import jdk.tools.jlink.plugin.PluginException;
47 import jdk.tools.jlink.plugin.ModulePool;
48 import jdk.tools.jlink.plugin.TransformerPlugin;
49
50 /**
51 * Plugin to explicitly specify the locale data included in jdk.localedata
52 * module. This plugin provides a jlink command line option "--include-locales"
53 * with an argument. The argument is a list of BCP 47 language tags separated
54 * by a comma. E.g.,
55 *
56 * "jlink --include-locales en,ja,*-IN"
57 *
58 * This option will include locale data for all available English and Japanese
59 * languages, and ones for the country of India. All other locale data are
60 * filtered out on the image creation.
61 *
62 * Here are a few assumptions:
63 *
64 * 0. All locale data in java.base are unconditionally included.
65 * 1. All the selective locale data are in jdk.localedata module
66 * 2. Their package names are constructed by appending ".ext" to
67 * the corresponding ones in java.base module.
68 * 3. Available locales string in LocaleDataMetaInfo class should
69 * start with at least one white space character, e.g., " ar ar-EG ..."
70 * ^
71 */
72 public final class IncludeLocalesPlugin implements TransformerPlugin, ResourcePrevisitor {
73
74 public static final String NAME = "include-locales";
75 private static final String MODULENAME = "jdk.localedata";
76 private static final Set<String> LOCALEDATA_PACKAGES = Set.of(
77 "sun.text.resources.cldr.ext",
78 "sun.text.resources.ext",
79 "sun.util.resources.cldr.ext",
80 "sun.util.resources.cldr.provider",
81 "sun.util.resources.ext",
82 "sun.util.resources.provider");
83 private static final String METAINFONAME = "LocaleDataMetaInfo";
84 private static final String META_FILES =
85 "*module-info.class," +
86 "*LocaleDataProvider.class," +
87 "*" + METAINFONAME + ".class,";
88 private static final String INCLUDE_LOCALE_FILES =
89 "*sun/text/resources/ext/[^\\/]+_%%.class," +
90 "*sun/util/resources/ext/[^\\/]+_%%.class," +
91 "*sun/text/resources/cldr/ext/[^\\/]+_%%.class," +
92 "*sun/util/resources/cldr/ext/[^\\/]+_%%.class,";
93 private Predicate<String> predicate;
94 private String userParam;
95 private List<Locale.LanguageRange> priorityList;
96 private List<Locale> available;
97 private List<String> filtered;
98
99 // Special COMPAT provider locales
100 private static final String jaJPJPTag = "ja-JP-JP";
101 private static final String noNONYTag = "no-NO-NY";
102 private static final String thTHTHTag = "th-TH-TH";
103 private static final Locale jaJPJP = new Locale("ja", "JP", "JP");
104 private static final Locale noNONY = new Locale("no", "NO", "NY");
105 private static final Locale thTHTH = new Locale("th", "TH", "TH");
106
107 @Override
108 public String getName() {
109 return NAME;
110 }
111
112 @Override
113 public void visit(ModulePool in, ModulePool out) {
114 in.transformAndCopy((resource) -> {
115 if (resource.getModule().equals(MODULENAME)) {
116 String path = resource.getPath();
117 resource = predicate.test(path) ? resource: null;
118 if (resource != null &&
119 resource.getType().equals(ModuleEntry.Type.CLASS_OR_RESOURCE)) {
120 byte[] bytes = resource.getBytes();
121 ClassReader cr = new ClassReader(bytes);
122 if (Arrays.stream(cr.getInterfaces())
123 .anyMatch(i -> i.contains(METAINFONAME)) &&
124 stripUnsupportedLocales(bytes, cr)) {
125 resource = ModuleEntry.create(MODULENAME, path,
126 resource.getType(),
127 new ByteArrayInputStream(bytes), bytes.length);
128 }
129 }
130 }
131 return resource;
132 }, out);
133 }
134
135 @Override
136 public Category getType() {
137 return Category.FILTER;
138 }
139
140 @Override
141 public String getDescription() {
142 return PluginsResourceBundle.getDescription(NAME);
143 }
144
145 @Override
146 public boolean hasArguments() {
147 return true;
148 }
149
150 @Override
151 public String getArgumentsDescription() {
152 return PluginsResourceBundle.getArgument(NAME);
153 }
154
155 @Override
156 public void configure(Map<String, String> config) {
157 userParam = config.get(NAME);
158 priorityList = Arrays.stream(userParam.split(","))
159 .map(s -> {
160 try {
161 return new Locale.LanguageRange(s);
162 } catch (IllegalArgumentException iae) {
163 throw new IllegalArgumentException(String.format(
164 PluginsResourceBundle.getMessage(NAME + ".invalidtag"), s));
165 }
166 })
167 .collect(Collectors.toList());
168 }
169
170 @Override
171 public void previsit(ModulePool resources, StringTable strings) {
172 final Pattern p = Pattern.compile(".*((Data_)|(Names_))(?<tag>.*)\\.class");
173 Optional<LinkModule> optMod = resources.findModule(MODULENAME);
174
175 // jdk.localedata module validation
176 if (optMod.isPresent()) {
177 LinkModule module = optMod.get();
178 Set<String> packages = module.getAllPackages();
179 if (!packages.containsAll(LOCALEDATA_PACKAGES)) {
180 throw new PluginException(PluginsResourceBundle.getMessage(NAME + ".missingpackages") +
181 LOCALEDATA_PACKAGES.stream()
182 .filter(pn -> !packages.contains(pn))
183 .collect(Collectors.joining(",\n\t")));
184 }
185
186 available = Stream.concat(module.entries()
187 .map(md -> p.matcher(md.getPath()))
188 .filter(m -> m.matches())
189 .map(m -> m.group("tag").replaceAll("_", "-")),
190 Stream.concat(Stream.of(jaJPJPTag), Stream.of(thTHTHTag)))
191 .distinct()
192 .sorted()
193 .map(IncludeLocalesPlugin::tagToLocale)
194 .collect(Collectors.toList());
195 } else {
196 // jdk.localedata is not added.
197 throw new PluginException(PluginsResourceBundle.getMessage(NAME + ".localedatanotfound"));
198 }
199 filtered = filterLocales(available);
200
201 if (filtered.isEmpty()) {
202 throw new PluginException(
203 String.format(PluginsResourceBundle.getMessage(NAME + ".nomatchinglocales"), userParam));
204 }
205
206 String value = META_FILES + filtered.stream()
207 .map(s -> includeLocaleFilePatterns(s))
208 .collect(Collectors.joining(","));
209 predicate = ResourceFilter.includeFilter(value);
210 }
211
212 private String includeLocaleFilePatterns(String tag) {
213 String pTag = tag.replaceAll("-", "_");
214 String files = "";
215 int lastDelimiter = tag.length();
216 String isoSpecial = pTag.matches("^(he|yi|id).*") ?
217 pTag.replaceFirst("he", "iw")
218 .replaceFirst("yi", "ji")
219 .replaceFirst("id", "in") : "";
220
221 // Add tag patterns including parents
222 while (true) {
223 pTag = pTag.substring(0, lastDelimiter);
224 files += INCLUDE_LOCALE_FILES.replaceAll("%%", pTag);
225
226 if (!isoSpecial.isEmpty()) {
227 isoSpecial = isoSpecial.substring(0, lastDelimiter);
228 files += INCLUDE_LOCALE_FILES.replaceAll("%%", isoSpecial);
229 }
230
231 lastDelimiter = pTag.lastIndexOf('_');
232 if (lastDelimiter == -1) {
233 break;
234 }
235 }
236
237 final String lang = pTag;
238
239 // Add possible special locales of the COMPAT provider
240 files += Set.of(jaJPJPTag, noNONYTag, thTHTHTag).stream()
241 .filter(stag -> lang.equals(stag.substring(0,2)))
242 .map(t -> INCLUDE_LOCALE_FILES.replaceAll("%%", t.replaceAll("-", "_")))
243 .collect(Collectors.joining(","));
244
245 // Add possible UN.M49 files (unconditional for now) for each language
246 files += INCLUDE_LOCALE_FILES.replaceAll("%%", lang + "_[0-9]{3}");
247 if (!isoSpecial.isEmpty()) {
248 files += INCLUDE_LOCALE_FILES.replaceAll("%%", isoSpecial + "_[0-9]{3}");
249 }
250
251 // Add Thai BreakIterator related data files
252 if (lang.equals("th")) {
253 files += "*sun/text/resources/thai_dict," +
254 "*sun/text/resources/[^\\/]+BreakIteratorData_th,";
255 }
256
257 // Add Taiwan resource bundles for Hong Kong
258 if (tag.startsWith("zh-HK")) {
259 files += INCLUDE_LOCALE_FILES.replaceAll("%%", "zh_TW");
260 }
261
262 return files;
263 }
264
265 private boolean stripUnsupportedLocales(byte[] bytes, ClassReader cr) {
266 char[] buf = new char[cr.getMaxStringLength()];
267 boolean[] modified = new boolean[1];
268
269 IntStream.range(1, cr.getItemCount())
270 .map(item -> cr.getItem(item))
271 .forEach(itemIndex -> {
272 if (bytes[itemIndex - 1] == 1 && // UTF-8
273 bytes[itemIndex + 2] == (byte)' ') { // fast check for leading space
274 int length = cr.readUnsignedShort(itemIndex);
275 byte[] b = new byte[length];
276 System.arraycopy(bytes, itemIndex + 2, b, 0, length);
277 if (filterOutUnsupportedTags(b)) {
278 // copy back
279 System.arraycopy(b, 0, bytes, itemIndex + 2, length);
280 modified[0] = true;
281 }
282 }
283 });
284
285 return modified[0];
286 }
287
288 private boolean filterOutUnsupportedTags(byte[] b) {
289 List<Locale> locales;
290
291 try {
292 locales = Arrays.asList(new String(b).split(" ")).stream()
293 .filter(tag -> !tag.isEmpty())
294 .map(IncludeLocalesPlugin::tagToLocale)
295 .collect(Collectors.toList());
296 } catch (IllformedLocaleException ile) {
297 // Seems not an available locales string literal.
298 return false;
299 }
300
301 byte[] filteredBytes = filterLocales(locales).stream()
302 .collect(Collectors.joining(" "))
303 .getBytes();
304 System.arraycopy(filteredBytes, 0, b, 0, filteredBytes.length);
305 Arrays.fill(b, filteredBytes.length, b.length, (byte)' ');
306 return true;
307 }
308
309 private List<String> filterLocales(List<Locale> locales) {
310 List<String> ret =
311 Locale.filter(priorityList, locales, Locale.FilteringMode.EXTENDED_FILTERING).stream()
312 .map(loc ->
313 // Locale.filter() does not preserve the case, which is
314 // significant for "variant" equality. Retrieve the original
315 // locales from the pre-filtered list.
316 locales.stream()
317 .filter(l -> l.toString().equalsIgnoreCase(loc.toString()))
318 .findAny()
319 .orElse(Locale.ROOT)
320 .toLanguageTag())
321 .collect(Collectors.toList());
322
323 // no-NO-NY.toLanguageTag() returns "nn-NO", so specially handle it here
324 if (ret.contains("no-NO")) {
325 ret.add(noNONYTag);
326 }
327
328 return ret;
329 }
330
331 private static final Locale.Builder LOCALE_BUILDER = new Locale.Builder();
332 private static Locale tagToLocale(String tag) {
333 // ISO3166 compatibility
334 tag = tag.replaceFirst("^iw", "he").replaceFirst("^ji", "yi").replaceFirst("^in", "id");
335
336 switch (tag) {
337 case jaJPJPTag:
338 return jaJPJP;
339 case noNONYTag:
340 return noNONY;
341 case thTHTHTag:
342 return thTHTH;
343 default:
344 LOCALE_BUILDER.clear();
345 LOCALE_BUILDER.setLanguageTag(tag);
346 return LOCALE_BUILDER.build();
347 }
348 }
349 }