i18npool/source/breakiterator/gendict.cxx

   1 /*************************************************************************
   2  *
   3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   4  *
   5  * Copyright 2000, 2010 Oracle and/or its affiliates.
   6  *
   7  * OpenOffice.org - a multi-platform office productivity suite
   8  *
   9  * This file is part of OpenOffice.org.
  10  *
  11  * OpenOffice.org is free software: you can redistribute it and/or modify
  12  * it under the terms of the GNU Lesser General Public License version 3
  13  * only, as published by the Free Software Foundation.
  14  *
  15  * OpenOffice.org is distributed in the hope that it will be useful,
  16  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  17  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  18  * GNU Lesser General Public License version 3 for more details
  19  * (a copy is included in the LICENSE file that accompanied this code).
  20  *
  21  * You should have received a copy of the GNU Lesser General Public License
  22  * version 3 along with OpenOffice.org.  If not, see
  23  * <http://www.openoffice.org/license.html>
  24  * for a copy of the LGPLv3 License.
  25  *
  26  ************************************************************************/
  27
  28 // MARKER(update_precomp.py): autogen include statement, do not remove
  29 #include "precompiled_i18npool.hxx"
  30
  31 #include <stdio.h>
  32 #include <string.h>
  33 #include <stdlib.h>
  34 #include <sal/main.h>
  35 #include <sal/types.h>
  36 #include <rtl/strbuf.hxx>
  37 #include <rtl/ustring.hxx>
  38
  39 using namespace ::rtl;
  40
  41 /* Main Procedure */
  42
  43 SAL_IMPLEMENT_MAIN_WITH_ARGS(argc, argv)
  44 {
  45     FILE *sfp, *cfp;
  46
  47     if (argc < 3) exit(-1);
  48
  49     sfp = fopen(argv[1], "rb"); // open the source file for read;
  50     if (sfp == NULL)
  51     {
  52         printf("Open the dictionary source file failed.");
  53         return -1;
  54     }
  55
  56     // create the C source file to write
  57     cfp = fopen(argv[2], "wb");
  58     if (cfp == NULL) {
  59         fclose(sfp);
  60         printf("Can't create the C source file.");
  61         return -1;
  62     }
  63
  64     fprintf(cfp, "/*\n");
  65     fprintf(cfp, " * Copyright(c) 1999 - 2000, Sun Microsystems, Inc.\n");
  66     fprintf(cfp, " * All Rights Reserved.\n");
  67     fprintf(cfp, " */\n\n");
  68     fprintf(cfp, "/* !!!The file is generated automatically. DONOT edit the file manually!!! */\n\n");
  69     fprintf(cfp, "#include <sal/types.h>\n\n");
  70     fprintf(cfp, "extern \"C\" {\n");
  71
  72     sal_Int32 count, i, j;
  73     sal_Int32 lenArrayCurr = 0, lenArrayCount = 0, lenArrayLen = 0, *lenArray = NULL, charArray[0x10000];
  74     sal_Bool exist[0x10000];
  75     for (i = 0; i < 0x10000; i++) {
  76         exist[i] = sal_False;
  77         charArray[i] = 0;
  78     }
  79
  80     // generate main dict. data array
  81     fprintf(cfp, "static const sal_Unicode dataArea[] = {");
  82     sal_Char str[1024];
  83     sal_Unicode current = 0;
  84     count = 0;
  85     while (fgets(str, 1024, sfp)) {
  86         // input file is in UTF-8 encoding
  87         // don't convert last new line character to Ostr.
  88         OUString Ostr((const sal_Char *)str, strlen(str) - 1, RTL_TEXTENCODING_UTF8);
  89         const sal_Unicode *u = Ostr.getStr();
  90
  91         sal_Int32 len = Ostr.getLength();
  92
  93         i=0;
  94         Ostr.iterateCodePoints(&i, 1);
  95         if (len == i) continue; // skip one character word
  96
  97         if (*u != current) {
  98         if (*u < current)
  99         printf("u %x, current %x, count %d, lenArrayCount %d\n", *u, current,
 100                     sal::static_int_cast<int>(count), sal::static_int_cast<int>(lenArrayCount));
 101         current = *u;
 102         charArray[current] = lenArrayCount;
 103         }
 104
 105         if (lenArrayLen <= lenArrayCount+1)
 106         lenArray = (sal_Int32*) realloc(lenArray, (lenArrayLen += 1000) * sizeof(sal_Int32));
 107         lenArray[lenArrayCount++] = lenArrayCurr;
 108
 109         exist[u[0]] = sal_True;
 110         for (i = 1; i < len; i++) {     // start from second character,
 111         exist[u[i]] = sal_True;     // since the first character is captured in charArray.
 112         lenArrayCurr++;
 113         if ((count++) % 0x10 == 0)
 114             fprintf(cfp, "\n\t");
 115         fprintf(cfp, "0x%04x, ", u[i]);
 116         }
 117     }
 118     lenArray[lenArrayCount++] = lenArrayCurr; // store last ending pointer
 119     charArray[current+1] = lenArrayCount;
 120     fprintf(cfp, "\n};\n");
 121
 122     // generate lenArray
 123     fprintf(cfp, "static const sal_Int32 lenArray[] = {\n\t");
 124     count = 1;
 125     fprintf(cfp, "0x%x, ", 0); // insert one slat for skipping 0 in index2 array.
 126     for (i = 0; i < lenArrayCount; i++) {
 127         fprintf(cfp, "0x%lx, ", static_cast<long unsigned int>(lenArray[i]));
 128         if (count == 0xf) {
 129         count = 0;
 130         fprintf(cfp, "\n\t");
 131         } else count++;
 132     }
 133     fprintf(cfp, "\n};\n");
 134
 135     free(lenArray);
 136
 137     // generate index1 array
 138     fprintf (cfp, "static const sal_Int16 index1[] = {\n\t");
 139     sal_Int16 set[0x100];
 140     count = 0;
 141     for (i = 0; i < 0x100; i++) {
 142         for (j = 0; j < 0x100; j++)
 143         if (charArray[(i*0x100) + j] != 0)
 144             break;
 145
 146         fprintf(cfp, "0x%02x, ", set[i] = (j < 0x100 ? sal::static_int_cast<sal_Int16>(count++) : 0xff));
 147         if ((i+1) % 0x10 == 0)
 148         fprintf (cfp, "\n\t");
 149     }
 150     fprintf (cfp, "};\n");
 151
 152     // generate index2 array
 153     fprintf (cfp, "static const sal_Int32 index2[] = {\n\t");
 154     sal_Int32 prev = 0;
 155     for (i = 0; i < 0x100; i++) {
 156         if (set[i] != 0xff) {
 157         for (j = 0; j < 0x100; j++) {
 158             sal_Int32 k = (i*0x100) + j;
 159             if (prev != 0 && charArray[k] == 0) {
 160             for (k++; k < 0x10000; k++)
 161                 if (charArray[k] != 0)
 162                 break;
 163             }
 164             prev = charArray[(i*0x100) + j];
 165             fprintf(
 166                 cfp, "0x%lx, ",
 167                 sal::static_int_cast< unsigned long >(
 168                     k < 0x10000 ? charArray[k] + 1 : 0));
 169             if ((j+1) % 0x10 == 0)
 170             fprintf (cfp, "\n\t");
 171         }
 172         fprintf (cfp, "\n\t");
 173         }
 174     }
 175     fprintf (cfp, "\n};\n");
 176
 177     // generate existMark array
 178     count = 0;
 179     fprintf (cfp, "static const sal_uInt8 existMark[] = {\n\t");
 180     for (i = 0; i < 0x1FFF; i++) {
 181         sal_uInt8 bit = 0;
 182         for (j = 0; j < 8; j++)
 183         if (exist[i * 8 + j])
 184             bit |= 1 << j;
 185         fprintf(cfp, "0x%02x, ", bit);
 186         if (count == 0xf) {
 187         count = 0;
 188         fprintf(cfp, "\n\t");
 189         } else count++;
 190     }
 191     fprintf (cfp, "\n};\n");
 192
 193     // create function to return arrays
 194     fprintf (cfp, "\tconst sal_uInt8* getExistMark() { return existMark; }\n");
 195     fprintf (cfp, "\tconst sal_Int16* getIndex1() { return index1; }\n");
 196     fprintf (cfp, "\tconst sal_Int32* getIndex2() { return index2; }\n");
 197     fprintf (cfp, "\tconst sal_Int32* getLenArray() { return lenArray; }\n");
 198     fprintf (cfp, "\tconst sal_Unicode* getDataArea() { return dataArea; }\n");
 199     fprintf (cfp, "}\n");
 200
 201     fclose(sfp);
 202     fclose(cfp);
 203
 204     return 0;
 205 }   // End of main