libjava/scripts/MakeCharTables.java

   1 /* MakeCharTables.java - converts gnu.java.lang.CharData into
   2                          include/java-chartables.h
   3    Copyright (C) 2002 Free Software Foundation, Inc.
   4
   5 This file is part of GNU Classpath.
   6
   7 GNU Classpath is free software; you can redistribute it and/or modify
   8 it under the terms of the GNU General Public License as published by
   9 the Free Software Foundation; either version 2, or (at your option)
  10 any later version.
  11
  12 GNU Classpath is distributed in the hope that it will be useful, but
  13 WITHOUT ANY WARRANTY; without even the implied warranty of
  14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15 General Public License for more details.
  16
  17 You should have received a copy of the GNU General Public License
  18 along with GNU Classpath; see the file COPYING.  If not, write to the
  19 Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  20 02111-1307 USA.
  21
  22 Linking this library statically or dynamically with other modules is
  23 making a combined work based on this library.  Thus, the terms and
  24 conditions of the GNU General Public License cover the whole
  25 combination.
  26
  27 As a special exception, the copyright holders of this library give you
  28 permission to link this library with independent modules to produce an
  29 executable, regardless of the license terms of these independent
  30 modules, and to copy and distribute the resulting executable under
  31 terms of your choice, provided that you also meet, for each linked
  32 independent module, the terms and conditions of the license of that
  33 module.  An independent module is a module which is not derived from
  34 or based on this library.  If you modify this library, you may extend
  35 this exception to your version of the library, but you are not
  36 obligated to do so.  If you do not wish to do so, delete this
  37 exception statement from your version. */
  38
  39 import gnu.java.lang.CharData;
  40
  41 public class MakeCharTables implements CharData
  42 {
  43   public static void main(String[] args)
  44   {
  45     System.out.println("/* java-chartables.h -- Character tables for java.lang.Character -*- c++ -*-\n"
  46                        + "   Copyright (C) 2002 Free Software Foundation, Inc.\n"
  47                        + "   *** This file is generated by scripts/MakeCharTables.java ***\n"
  48                        + "\n"
  49                        + "This file is part of GNU Classpath.\n"
  50                        + "\n"
  51                        + "GNU Classpath is free software; you can redistribute it and/or modify\n"
  52                        + "it under the terms of the GNU General Public License as published by\n"
  53                        + "the Free Software Foundation; either version 2, or (at your option)\n"
  54                        + "any later version.\n"
  55                        + "\n"
  56                        + "GNU Classpath is distributed in the hope that it will be useful, but\n"
  57                        + "WITHOUT ANY WARRANTY; without even the implied warranty of\n"
  58                        + "MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU\n"
  59                        + "General Public License for more details.\n"
  60                        + "\n"
  61                        + "You should have received a copy of the GNU General Public License\n"
  62                        + "along with GNU Classpath; see the file COPYING.  If not, write to the\n"
  63                        + "Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA\n"
  64                        + "02111-1307 USA.\n"
  65                        + "\n"
  66                        + "Linking this library statically or dynamically with other modules is\n"
  67                        + "making a combined work based on this library.  Thus, the terms and\n"
  68                        + "conditions of the GNU General Public License cover the whole\n"
  69                        + "combination.\n"
  70                        + "\n"
  71                        + "As a special exception, the copyright holders of this library give you\n"
  72                        + "permission to link this library with independent modules to produce an\n"
  73                        + "executable, regardless of the license terms of these independent\n"
  74                        + "modules, and to copy and distribute the resulting executable under\n"
  75                        + "terms of your choice, provided that you also meet, for each linked\n"
  76                        + "independent module, the terms and conditions of the license of that\n"
  77                        + "module.  An independent module is a module which is not derived from\n"
  78                        + "or based on this library.  If you modify this library, you may extend\n"
  79                        + "this exception to your version of the library, but you are not\n"
  80                        + "obligated to do so.  If you do not wish to do so, delete this\n"
  81                        + "exception statement from your version. */\n"
  82                        + "\n"
  83                        + "#ifndef __JAVA_CHARTABLES_H__\n"
  84                        + "#define __JAVA_CHARTABLES_H__\n"
  85                        + "\n"
  86                        + "// These tables are automatically generated by scripts/MakeCharTables.java.\n"
  87                        + "// This is in turn parsing gnu.java.lang.CharData, which is generated by\n"
  88                        + "// scripts/unicode-muncher.pl.  The Unicode data comes from\n"
  89                        + "// www.unicode.org; this header is based on\n"
  90                        + "// " + SOURCE + ". JDK 1.4 uses Unicode version 3.0.0.\n"
  91                        + "// DO NOT EDIT the tables.  Instead, fix the upstream scripts and run\n"
  92                        + "// them again.\n"
  93                        + "\n"
  94                        + "// The data is stored in C style arrays of the appropriate CNI types, to\n"
  95                        + "// guarantee that the data is constant and non-relocatable.  The field\n"
  96                        + "// <code>blocks</code> stores the offset of a block of 2<supSHIFT</sup>\n"
  97                        + "// characters within <code>data</code>. The data field, in turn, stores\n"
  98                        + "// information about each character in the low order bits, and an offset\n"
  99                        + "// into the attribute tables <code>upper</code>, <code>lower</code>,\n"
 100                        + "// <code>numValue</code>, and <code>direction</code>.  Notice that the\n"
 101                        + "// attribute tables are much smaller than 0xffff entries; as many characters\n"
 102                        + "// in Unicode share common attributes.  Finally, there is a listing for\n"
 103                        + "// <code>title</code> exceptions (most characters just have the same title\n"
 104                        + "// case as upper case).\n"
 105                        + "\n"
 106                        + "// This file should only be included by natCharacter.cc\n"
 107                        + "\n"
 108                        + "/**\n"
 109                        + " * The character shift amount to look up the block offset. In other words,\n"
 110                        + " * <code>(char) (blocks[ch >> SHIFT] + ch)</code> is the index where\n"
 111                        + " * <code>ch</code> is described in <code>data</code>.\n"
 112                        + " */\n"
 113                        + "#define SHIFT " + SHIFT);
 114
 115     convertString("/**\n"
 116                   + " * The mapping of character blocks to their location in <code>data</code>.\n"
 117                   + " * Each entry has been adjusted so that a modulo 16 sum with the desired\n"
 118                   + " * character gives the actual index into <code>data</code>.\n"
 119                   + " */",
 120                   char.class, "blocks", BLOCKS);
 121
 122     convertString("/**\n"
 123                   + " * Information about each character.  The low order 5 bits form the\n"
 124                   + " * character type, the next bit is a flag for non-breaking spaces, and the\n"
 125                   + " * next bit is a flag for mirrored directionality.  The high order 9 bits\n"
 126                   + " * form the offset into the attribute tables.  Note that this limits the\n"
 127                   + " * number of unique character attributes to 512, which is not a problem\n"
 128                   + " * as of Unicode version 3.2.0, but may soon become one.\n"
 129                   + " */",
 130                   char.class, "data", DATA);
 131
 132     convertString("/**\n"
 133                   + " * This is the attribute table for computing the numeric value of a\n"
 134                   + " * character.  The value is -1 if Unicode does not define a value, -2\n"
 135                   + " * if the value is not a positive integer, otherwise it is the value.\n"
 136                   + " */",
 137                   short.class, "numValue", NUM_VALUE);
 138
 139     convertString("/**\n"
 140                   + " * This is the attribute table for computing the uppercase representation\n"
 141                   + " * of a character.  The value is the difference between the character and\n"
 142                   + " * its uppercase version.\n"
 143                   + " */",
 144                   short.class, "upper", UPPER);
 145
 146     convertString("/**\n"
 147                   + " * This is the attribute table for computing the lowercase representation\n"
 148                   + " * of a character.  The value is the difference between the character and\n"
 149                   + " * its lowercase version.\n"
 150                   + " */",
 151                   short.class, "lower", LOWER);
 152
 153     convertString("/**\n"
 154                   + " * This is the attribute table for computing the directionality class\n"
 155                   + " * of a character.  At present, the value is in the range 0 - 18 if the\n"
 156                   + " * character has a direction, otherwise it is -1.\n"
 157                   + " */",
 158                   byte.class, "direction", DIRECTION);
 159
 160     convertString("/**\n"
 161                   + " * This is the listing of titlecase special cases (all other character\n"
 162                   + " * can use <code>upper</code> to determine their titlecase).  The listing\n"
 163                   + " * is a sequence of character pairs; converting the first character of the\n"
 164                   + " * pair to titlecase produces the second character.\n"
 165                   + " */",
 166                   char.class, "title", TITLE);
 167
 168     System.out.println();
 169     System.out.println("#endif /* __JAVA_CHARTABLES_H__ */");
 170   }
 171
 172   private static void convertString(String header, Class type,
 173                                     String name, String field)
 174   {
 175     System.out.println();
 176     System.out.println(header);
 177     System.out.println("static const j" + type.getName() + " " + name
 178                        + "[] = {");
 179     char[] data = field.toCharArray();
 180     int wrap;
 181     if (type == char.class)
 182       wrap = 10;
 183     else if (type == byte.class)
 184       wrap = 21;
 185     else if (type == short.class)
 186       wrap = 13;
 187     else
 188       throw new Error("Unexpeced type");
 189     for (int i = 0; i < data.length; i += wrap)
 190       {
 191         System.out.print("   ");
 192         for (int j = 0; j < wrap; j++)
 193           {
 194             if (i + j >= data.length)
 195               break;
 196             System.out.print(" ");
 197             if (type == char.class)
 198               System.out.print((int) data[i + j]);
 199             else if (type == byte.class)
 200               System.out.print((byte) data[i + j]);
 201             else if (type == short.class)
 202               System.out.print((short) data[i + j]);
 203             System.out.print(",");
 204           }
 205         System.out.println();
 206       }
 207     System.out.println("  };\n"
 208                        + "/** Length of " + name + ". */\n"
 209                        + "static const int " + name + "_length = "
 210                        + data.length + ";");
 211   }
 212 }