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.ArrayList;
29 import java.util.Arrays;
30 import java.util.Collections;
31 import java.util.IllformedLocaleException;
32 import java.util.Locale;
33 import java.util.List;
34 import java.util.Map;
35 import java.util.Optional;
36 import java.util.Set;
37 import java.util.function.Predicate;
38 import java.util.regex.Pattern;
39 import java.util.stream.Collectors;
40 import java.util.stream.IntStream;
41 import java.util.stream.Stream;
42 import jdk.internal.org.objectweb.asm.ClassReader;
43 import jdk.tools.jlink.internal.ResourcePrevisitor;
44 import jdk.tools.jlink.internal.StringTable;
45 import jdk.tools.jlink.plugin.LinkModule;
46 import jdk.tools.jlink.plugin.ModuleEntry;
47 import jdk.tools.jlink.plugin.PluginException;
48 import jdk.tools.jlink.plugin.ModulePool;
49 import jdk.tools.jlink.plugin.TransformerPlugin;
50
51 /**
52 * Plugin to explicitly specify the locale data included in jdk.localedata
53 * module. This plugin provides a jlink command line option "--include-locales"
54 * with an argument. The argument is a list of BCP 47 language tags separated
55 * by a comma. E.g.,
56 *
57 * "jlink --include-locales en,ja,*-IN"
58 *
59 * This option will include locale data for all available English and Japanese
60 * languages, and ones for the country of India. All other locale data are
61 * filtered out on the image creation.
62 *
63 * Here are a few assumptions:
64 *
65 * 0. All locale data in java.base are unconditionally included.
66 * 1. All the selective locale data are in jdk.localedata module
67 * 2. Their package names are constructed by appending ".ext" to
68 * the corresponding ones in java.base module.
69 * 3. Available locales string in LocaleDataMetaInfo class should
70 * start with at least one white space character, e.g., " ar ar-EG ..."
71 * ^
72 */
73 public final class IncludeLocalesPlugin implements TransformerPlugin, ResourcePrevisitor {
74
75 public static final String NAME = "include-locales";
76 private static final String MODULENAME = "jdk.localedata";
77 private static final Set<String> LOCALEDATA_PACKAGES = Set.of(
78 "sun.text.resources.cldr.ext",
79 "sun.text.resources.ext",
80 "sun.util.resources.cldr.ext",
81 "sun.util.resources.cldr.provider",
82 "sun.util.resources.ext",
83 "sun.util.resources.provider");
84 private static final String METAINFONAME = "LocaleDataMetaInfo";
85 private static final List<String> META_FILES = List.of(
86 ".+module-info.class",
87 ".+LocaleDataProvider.class",
88 ".+" + METAINFONAME + ".class");
89 private static final List<String> INCLUDE_LOCALE_FILES = List.of(
90 ".+sun/text/resources/ext/[^_]+_",
91 ".+sun/util/resources/ext/[^_]+_",
92 ".+sun/text/resources/cldr/ext/[^_]+_",
93 ".+sun/util/resources/cldr/ext/[^_]+_");
94 private Predicate<String> predicate;
95 private String userParam;
96 private List<Locale.LanguageRange> priorityList;
97 private List<Locale> available;
98 private List<String> filtered;
99
100 // Special COMPAT provider locales
101 private static final String jaJPJPTag = "ja-JP-JP";
102 private static final String noNONYTag = "no-NO-NY";
103 private static final String thTHTHTag = "th-TH-TH";
104 private static final Locale jaJPJP = new Locale("ja", "JP", "JP");
105 private static final Locale noNONY = new Locale("no", "NO", "NY");
106 private static final Locale thTHTH = new Locale("th", "TH", "TH");
107
108 @Override
109 public String getName() {
110 return NAME;
111 }
112
113 @Override
114 public void visit(ModulePool in, ModulePool out) {
115 in.transformAndCopy((resource) -> {
116 if (resource.getModule().equals(MODULENAME)) {
117 String path = resource.getPath();
118 resource = predicate.test(path) ? resource: null;
119 if (resource != null &&
120 resource.getType().equals(ModuleEntry.Type.CLASS_OR_RESOURCE)) {
121 byte[] bytes = resource.getBytes();
122 ClassReader cr = new ClassReader(bytes);
123 if (Arrays.stream(cr.getInterfaces())
124 .anyMatch(i -> i.contains(METAINFONAME)) &&
125 stripUnsupportedLocales(bytes, cr)) {
126 resource = ModuleEntry.create(MODULENAME, path,
127 resource.getType(),
128 new ByteArrayInputStream(bytes), bytes.length);
129 }
130 }
131 }
132 return resource;
133 }, out);
134 }
135
136 @Override
137 public Category getType() {
138 return Category.FILTER;
139 }
140
141 @Override
142 public String getDescription() {
143 return PluginsResourceBundle.getDescription(NAME);
144 }
145
146 @Override
147 public boolean hasArguments() {
148 return true;
149 }
150
151 @Override
152 public String getArgumentsDescription() {
153 return PluginsResourceBundle.getArgument(NAME);
154 }
155
156 @Override
157 public void configure(Map<String, String> config) {
158 userParam = config.get(NAME);
159 priorityList = Arrays.stream(userParam.split(","))
160 .map(s -> {
161 try {
162 return new Locale.LanguageRange(s);
163 } catch (IllegalArgumentException iae) {
164 throw new IllegalArgumentException(String.format(
165 PluginsResourceBundle.getMessage(NAME + ".invalidtag"), s));
166 }
167 })
168 .collect(Collectors.toList());
169 }
170
171 @Override
172 public void previsit(ModulePool resources, StringTable strings) {
173 final Pattern p = Pattern.compile(".*((Data_)|(Names_))(?<tag>.*)\\.class");
174 Optional<LinkModule> optMod = resources.findModule(MODULENAME);
175
176 // jdk.localedata module validation
177 if (optMod.isPresent()) {
178 LinkModule module = optMod.get();
179 Set<String> packages = module.getAllPackages();
180 if (!packages.containsAll(LOCALEDATA_PACKAGES)) {
181 throw new PluginException(PluginsResourceBundle.getMessage(NAME + ".missingpackages") +
182 LOCALEDATA_PACKAGES.stream()
183 .filter(pn -> !packages.contains(pn))
184 .collect(Collectors.joining(",\n\t")));
185 }
186
187 available = Stream.concat(module.entries()
188 .map(md -> p.matcher(md.getPath()))
189 .filter(m -> m.matches())
190 .map(m -> m.group("tag").replaceAll("_", "-")),
191 Stream.concat(Stream.of(jaJPJPTag), Stream.of(thTHTHTag)))
192 .distinct()
193 .sorted()
194 .map(IncludeLocalesPlugin::tagToLocale)
195 .collect(Collectors.toList());
196 } else {
197 // jdk.localedata is not added.
198 throw new PluginException(PluginsResourceBundle.getMessage(NAME + ".localedatanotfound"));
199 }
200 filtered = filterLocales(available);
201
202 if (filtered.isEmpty()) {
203 throw new PluginException(
204 String.format(PluginsResourceBundle.getMessage(NAME + ".nomatchinglocales"), userParam));
205 }
206
207 List<String> value = Stream.concat(
208 META_FILES.stream(),
209 filtered.stream().flatMap(s -> includeLocaleFilePatterns(s).stream()))
210 .map(s -> "regex:" + s)
211 .collect(Collectors.toList());
212 predicate = ResourceFilter.includeFilter(value);
213 }
214
215 private List<String> includeLocaleFilePatterns(String tag) {
216 List<String> files = new ArrayList<>();
217 String pTag = tag.replaceAll("-", "_");
218 int lastDelimiter = tag.length();
219 String isoSpecial = pTag.matches("^(he|yi|id).*") ?
220 pTag.replaceFirst("he", "iw")
221 .replaceFirst("yi", "ji")
222 .replaceFirst("id", "in") : "";
223
224 // Add tag patterns including parents
225 while (true) {
226 pTag = pTag.substring(0, lastDelimiter);
227 files.addAll(includeLocaleFiles(pTag));
228
229 if (!isoSpecial.isEmpty()) {
230 isoSpecial = isoSpecial.substring(0, lastDelimiter);
231 files.addAll(includeLocaleFiles(isoSpecial));
232 }
233
234 lastDelimiter = pTag.lastIndexOf('_');
235 if (lastDelimiter == -1) {
236 break;
237 }
238 }
239
240 final String lang = pTag;
241
242 // Add possible special locales of the COMPAT provider
243 files.addAll(Set.of(jaJPJPTag, noNONYTag, thTHTHTag).stream()
244 .filter(stag -> lang.equals(stag.substring(0,2)))
245 .flatMap(t -> includeLocaleFiles(t.replaceAll("-", "_")).stream())
246 .collect(Collectors.toList()));
247
248 // Add possible UN.M49 files (unconditional for now) for each language
249 files.addAll(includeLocaleFiles(lang + "_[0-9]{3}"));
250 if (!isoSpecial.isEmpty()) {
251 files.addAll(includeLocaleFiles(isoSpecial + "_[0-9]{3}"));
252 }
253
254 // Add Thai BreakIterator related data files
255 if (lang.equals("th")) {
256 files.add(".+sun/text/resources/thai_dict");
257 files.add(".+sun/text/resources/[^_]+BreakIteratorData_th");
258 }
259
260 // Add Taiwan resource bundles for Hong Kong
261 if (tag.startsWith("zh-HK")) {
262 files.addAll(includeLocaleFiles("zh_TW"));
263 }
264
265 return files;
266 }
267
268 private List<String> includeLocaleFiles(String localeStr) {
269 return INCLUDE_LOCALE_FILES.stream()
270 .map(s -> s + localeStr + ".class")
271 .collect(Collectors.toList());
272 }
273
274 private boolean stripUnsupportedLocales(byte[] bytes, ClassReader cr) {
275 char[] buf = new char[cr.getMaxStringLength()];
276 boolean[] modified = new boolean[1];
277
278 IntStream.range(1, cr.getItemCount())
279 .map(item -> cr.getItem(item))
280 .forEach(itemIndex -> {
281 if (bytes[itemIndex - 1] == 1 && // UTF-8
282 bytes[itemIndex + 2] == (byte)' ') { // fast check for leading space
283 int length = cr.readUnsignedShort(itemIndex);
284 byte[] b = new byte[length];
285 System.arraycopy(bytes, itemIndex + 2, b, 0, length);
286 if (filterOutUnsupportedTags(b)) {
287 // copy back
288 System.arraycopy(b, 0, bytes, itemIndex + 2, length);
289 modified[0] = true;
290 }
291 }
292 });
293
294 return modified[0];
295 }
296
297 private boolean filterOutUnsupportedTags(byte[] b) {
298 List<Locale> locales;
299
300 try {
301 locales = Arrays.asList(new String(b).split(" ")).stream()
302 .filter(tag -> !tag.isEmpty())
303 .map(IncludeLocalesPlugin::tagToLocale)
304 .collect(Collectors.toList());
305 } catch (IllformedLocaleException ile) {
306 // Seems not an available locales string literal.
307 return false;
308 }
309
310 byte[] filteredBytes = filterLocales(locales).stream()
311 .collect(Collectors.joining(" "))
312 .getBytes();
313 System.arraycopy(filteredBytes, 0, b, 0, filteredBytes.length);
314 Arrays.fill(b, filteredBytes.length, b.length, (byte)' ');
315 return true;
316 }
317
318 private List<String> filterLocales(List<Locale> locales) {
319 List<String> ret =
320 Locale.filter(priorityList, locales, Locale.FilteringMode.EXTENDED_FILTERING).stream()
321 .map(loc ->
322 // Locale.filter() does not preserve the case, which is
323 // significant for "variant" equality. Retrieve the original
324 // locales from the pre-filtered list.
325 locales.stream()
326 .filter(l -> l.toString().equalsIgnoreCase(loc.toString()))
327 .findAny()
328 .orElse(Locale.ROOT)
329 .toLanguageTag())
330 .collect(Collectors.toList());
331
332 // no-NO-NY.toLanguageTag() returns "nn-NO", so specially handle it here
333 if (ret.contains("no-NO")) {
334 ret.add(noNONYTag);
335 }
336
337 return ret;
338 }
339
340 private static final Locale.Builder LOCALE_BUILDER = new Locale.Builder();
341 private static Locale tagToLocale(String tag) {
342 // ISO3166 compatibility
343 tag = tag.replaceFirst("^iw", "he").replaceFirst("^ji", "yi").replaceFirst("^in", "id");
344
345 switch (tag) {
346 case jaJPJPTag:
347 return jaJPJP;
348 case noNONYTag:
349 return noNONY;
350 case thTHTHTag:
351 return thTHTH;
352 default:
353 LOCALE_BUILDER.clear();
354 LOCALE_BUILDER.setLanguageTag(tag);
355 return LOCALE_BUILDER.build();
356 }
357 }
358 }