bin/sh/mksyntax.c

   1 /*-
   2  * Copyright (c) 1991, 1993
   3  *      The Regents of the University of California.  All rights reserved.
   4  *
   5  * This code is derived from software contributed to Berkeley by
   6  * Kenneth Almquist.
   7  *
   8  * Redistribution and use in source and binary forms, with or without
   9  * modification, are permitted provided that the following conditions
  10  * are met:
  11  * 1. Redistributions of source code must retain the above copyright
  12  *    notice, this list of conditions and the following disclaimer.
  13  * 2. Redistributions in binary form must reproduce the above copyright
  14  *    notice, this list of conditions and the following disclaimer in the
  15  *    documentation and/or other materials provided with the distribution.
  16  * 3. All advertising materials mentioning features or use of this software
  17  *    must display the following acknowledgement:
  18  *      This product includes software developed by the University of
  19  *      California, Berkeley and its contributors.
  20  * 4. Neither the name of the University nor the names of its contributors
  21  *    may be used to endorse or promote products derived from this software
  22  *    without specific prior written permission.
  23  *
  24  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  25  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  26  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  27  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  28  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  29  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  30  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  31  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  32  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  33  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  34  * SUCH DAMAGE.
  35  *
  36  * @(#) Copyright (c) 1991, 1993 The Regents of the University of California.  All rights reserved.
  37  * @(#)mksyntax.c       8.2 (Berkeley) 5/4/95
  38  * $FreeBSD: src/bin/sh/mksyntax.c,v 1.14.2.3 2002/07/19 04:38:51 tjr Exp $
  39  * $DragonFly: src/bin/sh/mksyntax.c,v 1.4 2006/09/28 04:19:40 pavalos Exp $
  40  */
  41
  42 /*
  43  * This program creates syntax.h and syntax.c.
  44  */
  45
  46 #include <stdio.h>
  47 #include <stdlib.h>
  48 #include <string.h>
  49 #include "parser.h"
  50
  51
  52 struct synclass {
  53         const char *name;
  54         const char *comment;
  55 };
  56
  57 /* Syntax classes */
  58 struct synclass synclass[] = {
  59         { "CWORD",      "character is nothing special" },
  60         { "CNL",        "newline character" },
  61         { "CBACK",      "a backslash character" },
  62         { "CSQUOTE",    "single quote" },
  63         { "CDQUOTE",    "double quote" },
  64         { "CENDQUOTE",  "a terminating quote" },
  65         { "CBQUOTE",    "backwards single quote" },
  66         { "CVAR",       "a dollar sign" },
  67         { "CENDVAR",    "a '}' character" },
  68         { "CLP",        "a left paren in arithmetic" },
  69         { "CRP",        "a right paren in arithmetic" },
  70         { "CEOF",       "end of file" },
  71         { "CCTL",       "like CWORD, except it must be escaped" },
  72         { "CSPCL",      "these terminate a word" },
  73         { NULL,         NULL }
  74 };
  75
  76
  77 /*
  78  * Syntax classes for is_ functions.  Warning:  if you add new classes
  79  * you may have to change the definition of the is_in_name macro.
  80  */
  81 struct synclass is_entry[] = {
  82         { "ISDIGIT",    "a digit" },
  83         { "ISUPPER",    "an upper case letter" },
  84         { "ISLOWER",    "a lower case letter" },
  85         { "ISUNDER",    "an underscore" },
  86         { "ISSPECL",    "the name of a special parameter" },
  87         { NULL,         NULL }
  88 };
  89
  90 static char writer[] = "\
  91 /*\n\
  92  * This file was generated by the mksyntax program.\n\
  93  */\n\
  94 \n";
  95
  96
  97 static FILE *cfile;
  98 static FILE *hfile;
  99 static const char *syntax[513];
 100 static int base;
 101 static int size;        /* number of values which a char variable can have */
 102 static int nbits;       /* number of bits in a character */
 103 static int digit_contig;/* true if digits are contiguous */
 104
 105 static void filltable(const char *);
 106 static void init(void);
 107 static void add(const char *, const char *);
 108 static void print(const char *);
 109 static void output_type_macros(void);
 110 static void digit_convert(void);
 111
 112 int
 113 main(int argc __unused, char **argv __unused)
 114 {
 115         char c;
 116         char d;
 117         int sign;
 118         int i;
 119         char buf[80];
 120         int pos;
 121         static char digit[] = "0123456789";
 122
 123         /* Create output files */
 124         if ((cfile = fopen("syntax.c", "w")) == NULL) {
 125                 perror("syntax.c");
 126                 exit(2);
 127         }
 128         if ((hfile = fopen("syntax.h", "w")) == NULL) {
 129                 perror("syntax.h");
 130                 exit(2);
 131         }
 132         fputs(writer, hfile);
 133         fputs(writer, cfile);
 134
 135         /* Determine the characteristics of chars. */
 136         c = -1;
 137         if (c < 0)
 138                 sign = 1;
 139         else
 140                 sign = 0;
 141         for (nbits = 1 ; ; nbits++) {
 142                 d = (1 << nbits) - 1;
 143                 if (d == c)
 144                         break;
 145         }
 146 #if 0
 147         printf("%s %d bit chars\n", sign? "signed" : "unsigned", nbits);
 148 #endif
 149         if (nbits > 9) {
 150                 fputs("Characters can't have more than 9 bits\n", stderr);
 151                 exit(2);
 152         }
 153         size = (1 << nbits) + 1;
 154         base = 1;
 155         if (sign)
 156                 base += 1 << (nbits - 1);
 157         digit_contig = 1;
 158         for (i = 0 ; i < 10 ; i++) {
 159                 if (digit[i] != '0' + i)
 160                         digit_contig = 0;
 161         }
 162
 163         fputs("#include <sys/cdefs.h>\n", hfile);
 164         fputs("#include <ctype.h>\n", hfile);
 165
 166         /* Generate the #define statements in the header file */
 167         fputs("/* Syntax classes */\n", hfile);
 168         for (i = 0 ; synclass[i].name ; i++) {
 169                 sprintf(buf, "#define %s %d", synclass[i].name, i);
 170                 fputs(buf, hfile);
 171                 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
 172                         putc('\t', hfile);
 173                 fprintf(hfile, "/* %s */\n", synclass[i].comment);
 174         }
 175         putc('\n', hfile);
 176         fputs("/* Syntax classes for is_ functions */\n", hfile);
 177         for (i = 0 ; is_entry[i].name ; i++) {
 178                 sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i);
 179                 fputs(buf, hfile);
 180                 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
 181                         putc('\t', hfile);
 182                 fprintf(hfile, "/* %s */\n", is_entry[i].comment);
 183         }
 184         putc('\n', hfile);
 185         fprintf(hfile, "#define SYNBASE %d\n", base);
 186         fprintf(hfile, "#define PEOF %d\n\n", -base);
 187         putc('\n', hfile);
 188         fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
 189         fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
 190         fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
 191         fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
 192         putc('\n', hfile);
 193         output_type_macros();           /* is_digit, etc. */
 194         putc('\n', hfile);
 195
 196         /* Generate the syntax tables. */
 197         fputs("#include \"shell.h\"\n", cfile);
 198         fputs("#include \"syntax.h\"\n\n", cfile);
 199         init();
 200         fputs("/* syntax table used when not in quotes */\n", cfile);
 201         add("\n", "CNL");
 202         add("\\", "CBACK");
 203         add("'", "CSQUOTE");
 204         add("\"", "CDQUOTE");
 205         add("`", "CBQUOTE");
 206         add("$", "CVAR");
 207         add("}", "CENDVAR");
 208         add("<>();&| \t", "CSPCL");
 209         print("basesyntax");
 210         init();
 211         fputs("\n/* syntax table used when in double quotes */\n", cfile);
 212         add("\n", "CNL");
 213         add("\\", "CBACK");
 214         add("\"", "CENDQUOTE");
 215         add("`", "CBQUOTE");
 216         add("$", "CVAR");
 217         add("}", "CENDVAR");
 218         /* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
 219         add("!*?[=~:/-", "CCTL");
 220         print("dqsyntax");
 221         init();
 222         fputs("\n/* syntax table used when in single quotes */\n", cfile);
 223         add("\n", "CNL");
 224         add("'", "CENDQUOTE");
 225         /* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
 226         add("!*?[=~:/-", "CCTL");
 227         print("sqsyntax");
 228         init();
 229         fputs("\n/* syntax table used when in arithmetic */\n", cfile);
 230         add("\n", "CNL");
 231         add("\\", "CBACK");
 232         add("`", "CBQUOTE");
 233         add("'", "CSQUOTE");
 234         add("\"", "CDQUOTE");
 235         add("$", "CVAR");
 236         add("}", "CENDVAR");
 237         add("(", "CLP");
 238         add(")", "CRP");
 239         print("arisyntax");
 240         filltable("0");
 241         fputs("\n/* character classification table */\n", cfile);
 242         add("0123456789", "ISDIGIT");
 243         add("abcdefghijklmnopqrstucvwxyz", "ISLOWER");
 244         add("ABCDEFGHIJKLMNOPQRSTUCVWXYZ", "ISUPPER");
 245         add("_", "ISUNDER");
 246         add("#?$!-*@", "ISSPECL");
 247         print("is_type");
 248         if (! digit_contig)
 249                 digit_convert();
 250         exit(0);
 251 }
 252
 253
 254
 255 /*
 256  * Clear the syntax table.
 257  */
 258
 259 static void
 260 filltable(const char *dftval)
 261 {
 262         int i;
 263
 264         for (i = 0 ; i < size ; i++)
 265                 syntax[i] = dftval;
 266 }
 267
 268
 269 /*
 270  * Initialize the syntax table with default values.
 271  */
 272
 273 static void
 274 init(void)
 275 {
 276         filltable("CWORD");
 277         syntax[0] = "CEOF";
 278         syntax[base + CTLESC] = "CCTL";
 279         syntax[base + CTLVAR] = "CCTL";
 280         syntax[base + CTLENDVAR] = "CCTL";
 281         syntax[base + CTLBACKQ] = "CCTL";
 282         syntax[base + CTLBACKQ + CTLQUOTE] = "CCTL";
 283         syntax[base + CTLARI] = "CCTL";
 284         syntax[base + CTLENDARI] = "CCTL";
 285         syntax[base + CTLQUOTEMARK] = "CCTL";
 286 }
 287
 288
 289 /*
 290  * Add entries to the syntax table.
 291  */
 292
 293 static void
 294 add(const char *p, const char *type)
 295 {
 296         while (*p)
 297                 syntax[*p++ + base] = type;
 298 }
 299
 300
 301
 302 /*
 303  * Output the syntax table.
 304  */
 305
 306 static void
 307 print(const char *name)
 308 {
 309         int i;
 310         int col;
 311
 312         fprintf(hfile, "extern const char %s[];\n", name);
 313         fprintf(cfile, "const char %s[%d] = {\n", name, size);
 314         col = 0;
 315         for (i = 0 ; i < size ; i++) {
 316                 if (i == 0) {
 317                         fputs("      ", cfile);
 318                 } else if ((i & 03) == 0) {
 319                         fputs(",\n      ", cfile);
 320                         col = 0;
 321                 } else {
 322                         putc(',', cfile);
 323                         while (++col < 9 * (i & 03))
 324                                 putc(' ', cfile);
 325                 }
 326                 fputs(syntax[i], cfile);
 327                 col += strlen(syntax[i]);
 328         }
 329         fputs("\n};\n", cfile);
 330 }
 331
 332
 333
 334 /*
 335  * Output character classification macros (e.g. is_digit).  If digits are
 336  * contiguous, we can test for them quickly.
 337  */
 338
 339 static const char *macro[] = {
 340         "#define is_digit(c)\t((is_type+SYNBASE)[c] & ISDIGIT)",
 341         "#define is_eof(c)\t((c) == PEOF)",
 342         "#define is_alpha(c)\t(((c) < CTLESC || (c) > CTLQUOTEMARK) && isalpha((unsigned char) (c)))",
 343         "#define is_name(c)\t(((c) < CTLESC || (c) > CTLQUOTEMARK) && ((c) == '_' || isalpha((unsigned char) (c))))",
 344         "#define is_in_name(c)\t(((c) < CTLESC || (c) > CTLQUOTEMARK) && ((c) == '_' || isalnum((unsigned char) (c))))",
 345         "#define is_special(c)\t((is_type+SYNBASE)[c] & (ISSPECL|ISDIGIT))",
 346         NULL
 347 };
 348
 349 static void
 350 output_type_macros(void)
 351 {
 352         const char **pp;
 353
 354         if (digit_contig)
 355                 macro[0] = "#define is_digit(c)\t((unsigned)((c) - '0') <= 9)";
 356         for (pp = macro ; *pp ; pp++)
 357                 fprintf(hfile, "%s\n", *pp);
 358         if (digit_contig)
 359                 fputs("#define digit_val(c)\t((c) - '0')\n", hfile);
 360         else
 361                 fputs("#define digit_val(c)\t(digit_value[c])\n", hfile);
 362 }
 363
 364
 365
 366 /*
 367  * Output digit conversion table (if digits are not contiguous).
 368  */
 369
 370 static void
 371 digit_convert(void)
 372 {
 373         int maxdigit;
 374         static char digit[] = "0123456789";
 375         char *p;
 376         int i;
 377
 378         maxdigit = 0;
 379         for (p = digit ; *p ; p++)
 380                 if (*p > maxdigit)
 381                         maxdigit = *p;
 382         fputs("extern const char digit_value[];\n", hfile);
 383         fputs("\n\nconst char digit_value[] = {\n", cfile);
 384         for (i = 0 ; i <= maxdigit ; i++) {
 385                 for (p = digit ; *p && *p != i ; p++);
 386                 if (*p == '\0')
 387                         p = digit;
 388                 fprintf(cfile, "      %td,\n", p - digit);
 389         }
 390         fputs("};\n", cfile);
 391 }