Line data Source code
1 : /* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
2 : /*
3 : * This file is part of the LibreOffice project.
4 : *
5 : * This Source Code Form is subject to the terms of the Mozilla Public
6 : * License, v. 2.0. If a copy of the MPL was not distributed with this
7 : * file, You can obtain one at http://mozilla.org/MPL/2.0/.
8 : *
9 : * This file incorporates work covered by the following license notice:
10 : *
11 : * Licensed to the Apache Software Foundation (ASF) under one or more
12 : * contributor license agreements. See the NOTICE file distributed
13 : * with this work for additional information regarding copyright
14 : * ownership. The ASF licenses this file to you under the Apache
15 : * License, Version 2.0 (the "License"); you may not use this file
16 : * except in compliance with the License. You may obtain a copy of
17 : * the License at http://www.apache.org/licenses/LICENSE-2.0 .
18 : */
19 :
20 :
21 : // generated list of languages
22 : #include "lrl_include.hxx"
23 :
24 : #include <rtl/ustrbuf.hxx>
25 : #include <i18nlangtag/languagetag.hxx>
26 : #include <i18nlangtag/languagetagicu.hxx>
27 : #include <collator_unicode.hxx>
28 : #include <localedata.hxx>
29 : #include <com/sun/star/i18n/CollatorOptions.hpp>
30 : #include <cppuhelper/supportsservice.hxx>
31 :
32 : using namespace ::com::sun::star;
33 : using namespace ::com::sun::star::lang;
34 : using namespace ::com::sun::star::uno;
35 : using namespace ::rtl;
36 :
37 : namespace com { namespace sun { namespace star { namespace i18n {
38 :
39 0 : Collator_Unicode::Collator_Unicode()
40 : {
41 0 : implementationName = "com.sun.star.i18n.Collator_Unicode";
42 0 : collator = NULL;
43 0 : uca_base = NULL;
44 : #ifndef DISABLE_DYNLOADING
45 0 : hModule = NULL;
46 : #endif
47 0 : }
48 :
49 0 : Collator_Unicode::~Collator_Unicode()
50 : {
51 0 : if (collator) delete collator;
52 0 : if (uca_base) delete uca_base;
53 : #ifndef DISABLE_DYNLOADING
54 0 : if (hModule) osl_unloadModule(hModule);
55 : #endif
56 0 : }
57 :
58 : #ifdef DISABLE_DYNLOADING
59 :
60 : extern "C" {
61 :
62 : // For DISABLE_DYNLOADING the generated functions have names that
63 : // start with get_collator_data_ to avoid clashing with a few
64 : // functions in the generated libindex_data that are called just
65 : // get_zh_pinyin for instance.
66 :
67 : const sal_uInt8* get_collator_data_ca_charset();
68 : const sal_uInt8* get_collator_data_dz_charset();
69 : const sal_uInt8* get_collator_data_hu_charset();
70 : const sal_uInt8* get_collator_data_ja_charset();
71 : const sal_uInt8* get_collator_data_ja_phonetic_alphanumeric_first();
72 : const sal_uInt8* get_collator_data_ja_phonetic_alphanumeric_last();
73 : const sal_uInt8* get_collator_data_ko_charset();
74 : const sal_uInt8* get_collator_data_ku_alphanumeric();
75 : const sal_uInt8* get_collator_data_ln_charset();
76 : const sal_uInt8* get_collator_data_my_dictionary();
77 : const sal_uInt8* get_collator_data_ne_charset();
78 : const sal_uInt8* get_collator_data_sid_charset();
79 : const sal_uInt8* get_collator_data_zh_TW_charset();
80 : const sal_uInt8* get_collator_data_zh_TW_radical();
81 : const sal_uInt8* get_collator_data_zh_TW_stroke();
82 : const sal_uInt8* get_collator_data_zh_charset();
83 : const sal_uInt8* get_collator_data_zh_pinyin();
84 : const sal_uInt8* get_collator_data_zh_radical();
85 : const sal_uInt8* get_collator_data_zh_stroke();
86 : const sal_uInt8* get_collator_data_zh_zhuyin();
87 :
88 : }
89 :
90 : #endif
91 :
92 : sal_Int32 SAL_CALL
93 0 : Collator_Unicode::compareSubstring( const OUString& str1, sal_Int32 off1, sal_Int32 len1,
94 : const OUString& str2, sal_Int32 off2, sal_Int32 len2) throw(RuntimeException, std::exception)
95 : {
96 0 : return collator->compare(reinterpret_cast<const UChar *>(str1.getStr()) + off1, len1, reinterpret_cast<const UChar *>(str2.getStr()) + off2, len2); // UChar != sal_Unicode in MinGW
97 : }
98 :
99 : sal_Int32 SAL_CALL
100 0 : Collator_Unicode::compareString( const OUString& str1, const OUString& str2) throw(RuntimeException, std::exception)
101 : {
102 0 : return collator->compare(reinterpret_cast<const UChar *>(str1.getStr()), reinterpret_cast<const UChar *>(str2.getStr())); // UChar != sal_Unicode in MinGW
103 : }
104 :
105 : #ifndef DISABLE_DYNLOADING
106 :
107 0 : extern "C" { static void SAL_CALL thisModule() {} }
108 :
109 : #endif
110 :
111 : sal_Int32 SAL_CALL
112 0 : Collator_Unicode::loadCollatorAlgorithm(const OUString& rAlgorithm, const lang::Locale& rLocale, sal_Int32 options)
113 : throw(RuntimeException, std::exception)
114 : {
115 0 : if (!collator) {
116 0 : UErrorCode status = U_ZERO_ERROR;
117 0 : OUString rule = LocaleDataImpl().getCollatorRuleByAlgorithm(rLocale, rAlgorithm);
118 0 : if (!rule.isEmpty()) {
119 0 : collator = new RuleBasedCollator(reinterpret_cast<const UChar *>(rule.getStr()), status); // UChar != sal_Unicode in MinGW
120 0 : if (! U_SUCCESS(status)) throw RuntimeException();
121 : }
122 0 : if (!collator && OUString::createFromAscii(LOCAL_RULE_LANGS).indexOf(rLocale.Language) >= 0) {
123 0 : const sal_uInt8* (*func)() = NULL;
124 :
125 : #ifndef DISABLE_DYNLOADING
126 0 : OUStringBuffer aBuf;
127 : #ifdef SAL_DLLPREFIX
128 0 : aBuf.appendAscii(SAL_DLLPREFIX);
129 : #endif
130 0 : aBuf.appendAscii( "collator_data" ).appendAscii( SAL_DLLEXTENSION );
131 0 : hModule = osl_loadModuleRelative( &thisModule, aBuf.makeStringAndClear().pData, SAL_LOADMODULE_DEFAULT );
132 0 : if (hModule) {
133 0 : aBuf.appendAscii("get_").append(rLocale.Language).appendAscii("_");
134 0 : if ( rLocale.Language == "zh" ) {
135 0 : OUString func_base = aBuf.makeStringAndClear();
136 0 : if (OUString("TW HK MO").indexOf(rLocale.Country) >= 0)
137 : func=(const sal_uInt8* (*)()) osl_getFunctionSymbol(hModule,
138 0 : OUString(func_base + "TW_" + rAlgorithm).pData);
139 0 : if (!func)
140 0 : func=(const sal_uInt8* (*)()) osl_getFunctionSymbol(hModule, OUString(func_base + rAlgorithm).pData);
141 : } else {
142 0 : if ( rLocale.Language == "ja" ) {
143 : // replace algorithm name to implementation name.
144 0 : if (rAlgorithm == "phonetic (alphanumeric first)")
145 0 : aBuf.appendAscii("phonetic_alphanumeric_first");
146 0 : else if (rAlgorithm == "phonetic (alphanumeric last)")
147 0 : aBuf.appendAscii("phonetic_alphanumeric_last");
148 : else
149 0 : aBuf.append(rAlgorithm);
150 : } else {
151 0 : aBuf.append(rAlgorithm);
152 : }
153 0 : func=(const sal_uInt8* (*)()) osl_getFunctionSymbol(hModule, aBuf.makeStringAndClear().pData);
154 : }
155 : }
156 : #else
157 : if ( rLocale.Language == "ca" ) {
158 : if ( rAlgorithm == "charset" )
159 : func = get_collator_data_ca_charset;
160 : } else if ( rLocale.Language == "dz" || rLocale.Language == "bo" ) {
161 : // 'bo' Tibetan uses the same collation rules as 'dz' Dzongkha
162 : if ( rAlgorithm == "charset" )
163 : func = get_collator_data_dz_charset;
164 : } else if ( rLocale.Language == "hu" ) {
165 : if ( rAlgorithm == "charset" )
166 : func = get_collator_data_hu_charset;
167 : } else if ( rLocale.Language == "ja" ) {
168 : if ( rAlgorithm == "charset" )
169 : func = get_collator_data_ja_charset;
170 : else if ( rAlgorithm == "phonetic (alphanumeric first)" )
171 : func = get_collator_data_ja_phonetic_alphanumeric_first;
172 : else if ( rAlgorithm == "phonetic (alphanumeric last)" )
173 : func = get_collator_data_ja_phonetic_alphanumeric_last;
174 : } else if ( rLocale.Language == "ko" ) {
175 : if ( rAlgorithm == "charset" )
176 : func = get_collator_data_ko_charset;
177 : } else if ( rLocale.Language == "ku" ) {
178 : if ( rAlgorithm == "alphanumeric" )
179 : func = get_collator_data_ku_alphanumeric;
180 : } else if ( rLocale.Language == "ln" ) {
181 : if ( rAlgorithm == "charset" )
182 : func = get_collator_data_ln_charset;
183 : } else if ( rLocale.Language == "my" ) {
184 : if ( rAlgorithm == "dictionary" )
185 : func = get_collator_data_my_dictionary;
186 : } else if ( rLocale.Language == "ne" ) {
187 : if ( rAlgorithm == "charset" )
188 : func = get_collator_data_ne_charset;
189 : } else if ( rLocale.Language == "sid" ) {
190 : if ( rAlgorithm == "charset" )
191 : func = get_collator_data_sid_charset;
192 : } else if ( rLocale.Language == "zh" && (rLocale.Country == "TW" || rLocale.Country == "HK" || rLocale.Country == "MO") ) {
193 : if ( rAlgorithm == "charset" )
194 : func = get_collator_data_zh_TW_charset;
195 : else if ( rAlgorithm == "radical" )
196 : func = get_collator_data_zh_TW_radical;
197 : else if ( rAlgorithm == "stroke" )
198 : func = get_collator_data_zh_TW_stroke;
199 : } else if ( rLocale.Language == "zh" ) {
200 : if ( rAlgorithm == "charset" )
201 : func = get_collator_data_zh_charset;
202 : else if ( rAlgorithm == "pinyin" )
203 : func = get_collator_data_zh_pinyin;
204 : else if ( rAlgorithm == "radical" )
205 : func = get_collator_data_zh_radical;
206 : else if ( rAlgorithm == "stroke" )
207 : func = get_collator_data_zh_stroke;
208 : else if ( rAlgorithm == "zhuyin" )
209 : func = get_collator_data_zh_zhuyin;
210 : }
211 : #endif
212 0 : if (func) {
213 0 : const sal_uInt8* ruleImage=func();
214 0 : uca_base = new RuleBasedCollator(static_cast<UChar*>(NULL), status);
215 0 : if (! U_SUCCESS(status)) throw RuntimeException();
216 0 : collator = new RuleBasedCollator(reinterpret_cast<const uint8_t*>(ruleImage), -1, uca_base, status);
217 0 : if (! U_SUCCESS(status)) throw RuntimeException();
218 0 : }
219 : }
220 0 : if (!collator) {
221 : /** ICU collators are loaded using a locale only.
222 : ICU uses Variant as collation algorithm name (like de__PHONEBOOK
223 : locale), note the empty territory (Country) designator in this special
224 : case here. The icu::Locale constructor changes the algorithm name to
225 : uppercase itself, so we don't have to bother with that.
226 : */
227 0 : icu::Locale icuLocale( LanguageTagIcu::getIcuLocale( LanguageTag( rLocale), rAlgorithm));
228 : // load ICU collator
229 0 : collator = (RuleBasedCollator*) icu::Collator::createInstance(icuLocale, status);
230 0 : if (! U_SUCCESS(status)) throw RuntimeException();
231 0 : }
232 : }
233 :
234 0 : if (options & CollatorOptions::CollatorOptions_IGNORE_CASE_ACCENT)
235 0 : collator->setStrength(Collator::PRIMARY);
236 0 : else if (options & CollatorOptions::CollatorOptions_IGNORE_CASE)
237 0 : collator->setStrength(Collator::SECONDARY);
238 : else
239 0 : collator->setStrength(Collator::TERTIARY);
240 :
241 0 : return(0);
242 : }
243 :
244 :
245 : OUString SAL_CALL
246 0 : Collator_Unicode::getImplementationName() throw( RuntimeException, std::exception )
247 : {
248 0 : return OUString::createFromAscii(implementationName);
249 : }
250 :
251 : sal_Bool SAL_CALL
252 0 : Collator_Unicode::supportsService(const OUString& rServiceName) throw( RuntimeException, std::exception )
253 : {
254 0 : return cppu::supportsService(this, rServiceName);
255 : }
256 :
257 : Sequence< OUString > SAL_CALL
258 0 : Collator_Unicode::getSupportedServiceNames() throw( RuntimeException, std::exception )
259 : {
260 0 : Sequence< OUString > aRet(1);
261 0 : aRet[0] = OUString::createFromAscii(implementationName);
262 0 : return aRet;
263 : }
264 :
265 : } } } }
266 :
267 : /* vim:set shiftwidth=4 softtabstop=4 expandtab: */
|
