sal/textenc/textenc.cxx

   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 /* TODO! This file should not be called textenc.c, because it is not the
  21    implementation of rtl/textenc.h.  Rather, it should be called
  22    gettextencodingdata.c. */
  23
  24 #include <sal/config.h>
  25
  26 #include <cstddef>
  27 #include <cstdlib>
  28
  29 #include <osl/module.hxx>
  30 #include <rtl/textenc.h>
  31 #include <rtl/ustring.h>
  32 #include <rtl/ustring.hxx>
  33 #include <sal/log.hxx>
  34 #include <sal/macros.h>
  35 #include <sal/types.h>
  36
  37 #include "convertsimple.hxx"
  38 #include "gettextencodingdata.hxx"
  39 #include "tcvtbyte.hxx"
  40 #include "tcvtutf8.hxx"
  41 #include "tenchelp.hxx"
  42
  43 #define NOTABUNI_START 0xFF
  44 #define NOTABUNI_END 0x00
  45
  46 #define NOTABCHAR_START 0xFFFF
  47 #define NOTABCHAR_END 0x0000
  48
  49 #define SAME8090UNI_START 0x80
  50 #define SAME8090UNI_END 0x9F
  51 sal_uInt16 const aImpl8090SameToUniTab[SAME8090UNI_END
  52                                                   - SAME8090UNI_START
  53                                                   + 1]
  54 = { 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087, /* 0x80 */
  55     0x0088, 0x0089, 0x008A, 0x008B, 0x008C, 0x008D, 0x008E, 0x008F,
  56     0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097, /* 0x90 */
  57     0x0098, 0x0099, 0x009A, 0x009B, 0x009C, 0x009D, 0x009E, 0x009F };
  58
  59 #define SAME8090CHAR_START 0x0080
  60 #define SAME8090CHAR_END 0x009F
  61 unsigned char const aImpl8090SameToCharTab[SAME8090CHAR_END
  62                                                   - SAME8090CHAR_START
  63                                                   + 1]
  64     = { 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, /* 0x0080 */
  65         0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x8D, 0x8E, 0x8F,
  66         0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, /* 0x0090 */
  67         0x98, 0x99, 0x9A, 0x9B, 0x9C, 0x9D, 0x9E, 0x9F };
  68
  69 #define SAMEA0FFCHAR_START 0x00A0
  70 #define SAMEA0FFCHAR_END 0x00FF
  71 unsigned char const aImplA0FFSameToCharTab[SAMEA0FFCHAR_END
  72                                                   - SAMEA0FFCHAR_START
  73                                                   + 1]
  74     = { 0xA0, 0xA1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, /* 0x00A0 */
  75         0xA8, 0xA9, 0xAA, 0xAB, 0xAC, 0xAD, 0xAE, 0xAF,
  76         0xB0, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7, /* 0x00B0 */
  77         0xB8, 0xB9, 0xBA, 0xBB, 0xBC, 0xBD, 0xBE, 0xBF,
  78         0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, /* 0x00C0 */
  79         0xC8, 0xC9, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF,
  80         0xD0, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7, /* 0x00D0 */
  81         0xD8, 0xD9, 0xDA, 0xDB, 0xDC, 0xDD, 0xDE, 0xDF,
  82         0xE0, 0xE1, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, /* 0x00E0 */
  83         0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF,
  84         0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, /* 0x00F0 */
  85         0xF8, 0xF9, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE, 0xFF };
  86
  87 /* ======================================================================= */
  88
  89 /* MS-1252 */
  90 /* Windows Standard CharSet (ANSI) for Western Script */
  91 /* 1-Byte, 0x00-0x7F ASCII without exception */
  92 /* Convert-Tables: mappings/vendors/micsft/windows/cp1252.txt from 04/15/98 Version 2.01 */
  93 /* Last-Changes from us: */
  94
  95 /* ----------------------------------------------------------------------- */
  96
  97 #define MS1252UNI_START                 0x80
  98 #define MS1252UNI_END                   0xFF
  99 sal_uInt16 const aImplMS1252ToUniTab[MS1252UNI_END - MS1252UNI_START + 1] =
 100 {
 101 /*       0       1       2       3       4       5       6       7 */
 102 /*       8       9       A       B       C       D       E       F */
 103     0x20AC,      0, 0x201A, 0x0192, 0x201E, 0x2026, 0x2020, 0x2021, /* 0x80 */
 104     0x02C6, 0x2030, 0x0160, 0x2039, 0x0152,      0, 0x017D,      0, /* 0x80 */
 105          0, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014, /* 0x90 */
 106     0x02DC, 0x2122, 0x0161, 0x203A, 0x0153,      0, 0x017E, 0x0178, /* 0x90 */
 107     0x00A0, 0x00A1, 0x00A2, 0x00A3, 0x00A4, 0x00A5, 0x00A6, 0x00A7, /* 0xA0 */
 108     0x00A8, 0x00A9, 0x00AA, 0x00AB, 0x00AC, 0x00AD, 0x00AE, 0x00AF, /* 0xA0 */
 109     0x00B0, 0x00B1, 0x00B2, 0x00B3, 0x00B4, 0x00B5, 0x00B6, 0x00B7, /* 0xB0 */
 110     0x00B8, 0x00B9, 0x00BA, 0x00BB, 0x00BC, 0x00BD, 0x00BE, 0x00BF, /* 0xB0 */
 111     0x00C0, 0x00C1, 0x00C2, 0x00C3, 0x00C4, 0x00C5, 0x00C6, 0x00C7, /* 0xC0 */
 112     0x00C8, 0x00C9, 0x00CA, 0x00CB, 0x00CC, 0x00CD, 0x00CE, 0x00CF, /* 0xC0 */
 113     0x00D0, 0x00D1, 0x00D2, 0x00D3, 0x00D4, 0x00D5, 0x00D6, 0x00D7, /* 0xD0 */
 114     0x00D8, 0x00D9, 0x00DA, 0x00DB, 0x00DC, 0x00DD, 0x00DE, 0x00DF, /* 0xD0 */
 115     0x00E0, 0x00E1, 0x00E2, 0x00E3, 0x00E4, 0x00E5, 0x00E6, 0x00E7, /* 0xE0 */
 116     0x00E8, 0x00E9, 0x00EA, 0x00EB, 0x00EC, 0x00ED, 0x00EE, 0x00EF, /* 0xE0 */
 117     0x00F0, 0x00F1, 0x00F2, 0x00F3, 0x00F4, 0x00F5, 0x00F6, 0x00F7, /* 0xF0 */
 118     0x00F8, 0x00F9, 0x00FA, 0x00FB, 0x00FC, 0x00FD, 0x00FE, 0x00FF  /* 0xF0 */
 119 };
 120
 121 /* ----------------------------------------------------------------------- */
 122
 123 #define MS1252TOCHARTABEX_COUNT         27
 124 ImplUniCharTabData const aImplMS1252ToCharTabEx[MS1252TOCHARTABEX_COUNT] =
 125 {
 126   { 0x0152, 0x8C, 0 },
 127   { 0x0153, 0x9C, 0 },
 128   { 0x0160, 0x8A, 0 },
 129   { 0x0161, 0x9A, 0 },
 130   { 0x0178, 0x9F, 0 },
 131   { 0x017D, 0x8E, 0 },
 132   { 0x017E, 0x9E, 0 },
 133   { 0x0192, 0x83, 0 },
 134   { 0x02C6, 0x88, 0 },
 135   { 0x02DC, 0x98, 0 },
 136   { 0x2013, 0x96, 0 },
 137   { 0x2014, 0x97, 0 },
 138   { 0x2018, 0x91, 0 },
 139   { 0x2019, 0x92, 0 },
 140   { 0x201A, 0x82, 0 },
 141   { 0x201C, 0x93, 0 },
 142   { 0x201D, 0x94, 0 },
 143   { 0x201E, 0x84, 0 },
 144   { 0x2020, 0x86, 0 },
 145   { 0x2021, 0x87, 0 },
 146   { 0x2022, 0x95, 0 },
 147   { 0x2026, 0x85, 0 },
 148   { 0x2030, 0x89, 0 },
 149   { 0x2039, 0x8B, 0 },
 150   { 0x203A, 0x9B, 0 },
 151   { 0x20AC, 0x80, 0 },
 152   { 0x2122, 0x99, 0 },
 153 };
 154
 155 /* ----------------------------------------------------------------------- */
 156
 157 ImplByteConvertData const aImplMS1252ByteCvtData =
 158 {
 159     aImplMS1252ToUniTab,
 160     nullptr,
 161     aImplA0FFSameToCharTab,
 162     nullptr,
 163     aImplMS1252ToCharTabEx,
 164     MS1252UNI_START, MS1252UNI_END,
 165     NOTABUNI_START, NOTABUNI_END,
 166     SAMEA0FFCHAR_START, SAMEA0FFCHAR_END,
 167     NOTABCHAR_START, NOTABCHAR_END,
 168     MS1252TOCHARTABEX_COUNT
 169 };
 170
 171 /* ----------------------------------------------------------------------- */
 172
 173 ImplTextEncodingData const aImplMS1252TextEncodingData
 174     = { { &aImplMS1252ByteCvtData,
 175           sal::detail::textenc::convertCharToUnicode,
 176           sal::detail::textenc::convertUnicodeToChar,
 177           nullptr,
 178           nullptr,
 179           nullptr,
 180           nullptr,
 181           nullptr,
 182           nullptr },
 183         "iso8859-1",
 184         "windows-1252",
 185         1,
 186         1,
 187         1,
 188         0,
 189         RTL_TEXTENCODING_INFO_ASCII | RTL_TEXTENCODING_INFO_MIME };
 190     /* WIN, SCRIPT_LATIN, pc code page 850 */
 191
 192 /* ======================================================================= */
 193
 194 /* ISO-8859-1 */
 195 /* Unix Standard CharSet (Latin1) for Western Script */
 196 /* 1-Byte, 0x00-0x7F ASCII without exception, 0x80-0x9F control character like in Unicode */
 197 /* Convert-Tables: mappings/iso8859/8859-1.txt from 07/27/99 Version 1.0 (based on Unicode 3.0) */
 198 /* Last-Changes from us: */
 199
 200 #define ISO88591UNI_START               0xA0
 201 #define ISO88591UNI_END                 0xFF
 202 sal_uInt16 const aImplISO88591ToUniTab[ISO88591UNI_END - ISO88591UNI_START + 1] =
 203 {
 204 /*       0       1       2       3       4       5       6       7 */
 205 /*       8       9       A       B       C       D       E       F */
 206     0x00A0, 0x00A1, 0x00A2, 0x00A3, 0x00A4, 0x00A5, 0x00A6, 0x00A7, /* 0xA0 */
 207     0x00A8, 0x00A9, 0x00AA, 0x00AB, 0x00AC, 0x00AD, 0x00AE, 0x00AF, /* 0xA0 */
 208     0x00B0, 0x00B1, 0x00B2, 0x00B3, 0x00B4, 0x00B5, 0x00B6, 0x00B7, /* 0xB0 */
 209     0x00B8, 0x00B9, 0x00BA, 0x00BB, 0x00BC, 0x00BD, 0x00BE, 0x00BF, /* 0xB0 */
 210     0x00C0, 0x00C1, 0x00C2, 0x00C3, 0x00C4, 0x00C5, 0x00C6, 0x00C7, /* 0xC0 */
 211     0x00C8, 0x00C9, 0x00CA, 0x00CB, 0x00CC, 0x00CD, 0x00CE, 0x00CF, /* 0xC0 */
 212     0x00D0, 0x00D1, 0x00D2, 0x00D3, 0x00D4, 0x00D5, 0x00D6, 0x00D7, /* 0xD0 */
 213     0x00D8, 0x00D9, 0x00DA, 0x00DB, 0x00DC, 0x00DD, 0x00DE, 0x00DF, /* 0xD0 */
 214     0x00E0, 0x00E1, 0x00E2, 0x00E3, 0x00E4, 0x00E5, 0x00E6, 0x00E7, /* 0xE0 */
 215     0x00E8, 0x00E9, 0x00EA, 0x00EB, 0x00EC, 0x00ED, 0x00EE, 0x00EF, /* 0xE0 */
 216     0x00F0, 0x00F1, 0x00F2, 0x00F3, 0x00F4, 0x00F5, 0x00F6, 0x00F7, /* 0xF0 */
 217     0x00F8, 0x00F9, 0x00FA, 0x00FB, 0x00FC, 0x00FD, 0x00FE, 0x00FF  /* 0xF0 */
 218 };
 219
 220 /* ----------------------------------------------------------------------- */
 221
 222 ImplByteConvertData const aImplISO88591ByteCvtData =
 223 {
 224     aImplISO88591ToUniTab,
 225     aImpl8090SameToUniTab,
 226     aImplA0FFSameToCharTab,
 227     aImpl8090SameToCharTab,
 228     nullptr,
 229     ISO88591UNI_START, ISO88591UNI_END,
 230     SAME8090UNI_START, SAME8090UNI_END,
 231     SAMEA0FFCHAR_START, SAMEA0FFCHAR_END,
 232     SAME8090CHAR_START, SAME8090CHAR_END,
 233     0
 234 };
 235
 236 /* ----------------------------------------------------------------------- */
 237
 238 ImplTextEncodingData const aImplISO88591TextEncodingData
 239     = { { &aImplISO88591ByteCvtData,
 240           sal::detail::textenc::convertCharToUnicode,
 241           sal::detail::textenc::convertUnicodeToChar,
 242           nullptr,
 243           nullptr,
 244           nullptr,
 245           nullptr,
 246           nullptr,
 247           nullptr },
 248         "iso8859-1",
 249         "iso-8859-1",
 250         1,
 251         1,
 252         1,
 253         0,
 254         RTL_TEXTENCODING_INFO_ASCII | RTL_TEXTENCODING_INFO_MIME };
 255     /* SCRIPT_LATIN, pc code page 850 */
 256
 257 /* ======================================================================= */
 258
 259 /* US-ASCII */
 260 /* 7-Bit ASCII */
 261 /* 1-Byte, 0x00-0x7F ASCII without exception */
 262 /* For the import we use ISO-8859-1 with MS extension (MS-1252), because */
 263 /* when the 8-Bit is set, the chance, that this is an ISO-8859-1 character */
 264 /* is the greatest. For the export all chars greater than 127 are not */
 265 /* converted and are replaced by the replacement character. */
 266 /* Last-Changes from us: */
 267
 268 /* ----------------------------------------------------------------------- */
 269
 270 ImplByteConvertData const aImplUSASCIIByteCvtData =
 271 {
 272     aImplMS1252ToUniTab,
 273     nullptr,
 274     nullptr,
 275     nullptr,
 276     nullptr,
 277     MS1252UNI_START, MS1252UNI_END,
 278     NOTABUNI_START, NOTABUNI_END,
 279     NOTABCHAR_START, NOTABCHAR_END,
 280     NOTABCHAR_START, NOTABCHAR_END,
 281     0
 282 };
 283
 284 /* ----------------------------------------------------------------------- */
 285
 286 ImplTextEncodingData const aImplUSASCIITextEncodingData
 287     = { { &aImplUSASCIIByteCvtData,
 288           sal::detail::textenc::convertCharToUnicode,
 289           sal::detail::textenc::convertUnicodeToChar,
 290           nullptr,
 291           nullptr,
 292           nullptr,
 293           nullptr,
 294           nullptr,
 295           nullptr },
 296         "iso8859-1",
 297         "us-ascii",
 298         1,
 299         1,
 300         1,
 301         0,
 302         RTL_TEXTENCODING_INFO_ASCII
 303             | RTL_TEXTENCODING_INFO_7BIT
 304             | RTL_TEXTENCODING_INFO_MIME };
 305     /* SCRIPT_LATIN, pc code page 437 */
 306
 307 ImplTextEncodingData const aImplUTF8TextEncodingData
 308     = { { nullptr,
 309           &ImplConvertUtf8ToUnicode,
 310           &ImplConvertUnicodeToUtf8,
 311           &ImplCreateUtf8ToUnicodeContext,
 312           &ImplDestroyUtf8ToUnicodeContext,
 313           &ImplResetUtf8ToUnicodeContext,
 314           &ImplCreateUnicodeToUtf8Context,
 315           &ImplDestroyUnicodeToUtf8Context,
 316           &ImplResetUnicodeToUtf8Context },
 317         "iso8859-1",
 318         "utf-8",
 319         1,
 320         6,
 321         1,
 322         0,
 323         RTL_TEXTENCODING_INFO_ASCII
 324             | RTL_TEXTENCODING_INFO_UNICODE
 325             | RTL_TEXTENCODING_INFO_MULTIBYTE
 326             | RTL_TEXTENCODING_INFO_MIME };
 327     /* SCRIPT_UNICODE, pc code page 850 */
 328
 329 static char aImplJavaUtf8TextConverterTag;
 330     /* The value of this tag is irrelevant.  Only its address != NULL is used to
 331        distinguish between RTL_TEXTENCODING_UTF8 and
 332        RTL_TEXTENCODING_JAVA_UTF8. */
 333
 334 ImplTextEncodingData const aImplJavaUtf8TextEncodingData
 335     = { { &aImplJavaUtf8TextConverterTag,
 336           &ImplConvertUtf8ToUnicode,
 337           &ImplConvertUnicodeToUtf8,
 338           &ImplCreateUtf8ToUnicodeContext,
 339           &ImplDestroyUtf8ToUnicodeContext,
 340           &ImplResetUtf8ToUnicodeContext,
 341           &ImplCreateUnicodeToUtf8Context,
 342           &ImplDestroyUnicodeToUtf8Context,
 343           &ImplResetUnicodeToUtf8Context },
 344         nullptr,
 345         nullptr,
 346         1,
 347         3,
 348         1,
 349         0,
 350         RTL_TEXTENCODING_INFO_UNICODE | RTL_TEXTENCODING_INFO_MULTIBYTE };
 351
 352 namespace {
 353
 354 #ifndef COND_LIB_SAL_TEXTENC
 355
 356 extern "C" ImplTextEncodingData const * sal_getFullTextEncodingData(
 357     rtl_TextEncoding); // from tables.cxx in sal_textenc library
 358
 359 class FullTextEncodingData {
 360 public:
 361     ImplTextEncodingData const * get(rtl_TextEncoding encoding) {
 362         (void) this; // loplugin:staticmethods
 363         return sal_getFullTextEncodingData(encoding);
 364     }
 365     FullTextEncodingData() = default;
 366     FullTextEncodingData(const FullTextEncodingData&) = delete;
 367     FullTextEncodingData& operator=(const FullTextEncodingData&) = delete;
 368 };
 369
 370 #else
 371
 372 extern "C" {
 373
 374 typedef ImplTextEncodingData const * TextEncodingFunction(rtl_TextEncoding);
 375
 376 void thisModule() {}
 377
 378 }
 379
 380 class FullTextEncodingData {
 381 public:
 382     FullTextEncodingData() {
 383         if (!module_.loadRelative(&thisModule, SAL_MODULENAME("sal_textenclo")))
 384         {
 385             SAL_WARN( "sal.textenc", "Loading sal_textenc library failed" );
 386             std::abort();
 387         }
 388         function_ = reinterpret_cast< TextEncodingFunction * >(
 389             module_.getFunctionSymbol("sal_getFullTextEncodingData"));
 390         if (function_ == nullptr) {
 391             SAL_WARN( "sal.textenc", "Obtaining sal_getFullTextEncodingData function from sal_textenc"
 392                 " library failed");
 393             std::abort();
 394         }
 395     }
 396
 397     ImplTextEncodingData const * get(rtl_TextEncoding encoding) const {
 398         return (*function_)(encoding);
 399     }
 400
 401     FullTextEncodingData(const FullTextEncodingData&) = delete;
 402     FullTextEncodingData& operator=(const FullTextEncodingData&) = delete;
 403
 404 private:
 405     osl::Module module_;
 406     TextEncodingFunction * function_;
 407 };
 408
 409 #endif
 410
 411 }
 412
 413 ImplTextEncodingData const *
 414 Impl_getTextEncodingData(rtl_TextEncoding nEncoding)
 415 {
 416     switch(nEncoding)
 417     {
 418         case RTL_TEXTENCODING_ASCII_US:
 419             return &aImplUSASCIITextEncodingData; break;
 420         case RTL_TEXTENCODING_MS_1252:
 421             return &aImplMS1252TextEncodingData; break;
 422         case RTL_TEXTENCODING_UTF8:
 423             return &aImplUTF8TextEncodingData; break;
 424         case RTL_TEXTENCODING_JAVA_UTF8:
 425             return &aImplJavaUtf8TextEncodingData; break;
 426         case RTL_TEXTENCODING_ISO_8859_1:
 427             return &aImplISO88591TextEncodingData; break;
 428         default:
 429         {
 430             static FullTextEncodingData gFullTextEncodingData;
 431             return gFullTextEncodingData.get(nEncoding);
 432         }
 433     }
 434 }
 435
 436 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */