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. Neither the name of the University nor the names of its contributors
  17  *    may be used to endorse or promote products derived from this software
  18  *    without specific prior written permission.
  19  *
  20  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  21  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  22  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  23  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  24  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  25  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  26  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  27  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  28  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  29  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  30  * SUCH DAMAGE.
  31  */
  32
  33 #if 0
  34 #ifndef lint
  35 static char const copyright[] =
  36 "@(#) Copyright (c) 1991, 1993\n\
  37         The Regents of the University of California.  All rights reserved.\n";
  38 #endif /* not lint */
  39
  40 #ifndef lint
  41 static char sccsid[] = "@(#)mksyntax.c  8.2 (Berkeley) 5/4/95";
  42 #endif /* not lint */
  43 #endif
  44 #include <sys/cdefs.h>
  45 __FBSDID("$FreeBSD$");
  46
  47 /*
  48  * This program creates syntax.h and syntax.c.
  49  */
  50
  51 #include <stdio.h>
  52 #include <stdlib.h>
  53 #include <string.h>
  54 #include "parser.h"
  55
  56
  57 struct synclass {
  58         const char *name;
  59         const char *comment;
  60 };
  61
  62 /* Syntax classes */
  63 static const struct synclass synclass[] = {
  64         { "CWORD",      "character is nothing special" },
  65         { "CNL",        "newline character" },
  66         { "CBACK",      "a backslash character" },
  67         { "CSBACK",     "a backslash character in single quotes" },
  68         { "CSQUOTE",    "single quote" },
  69         { "CDQUOTE",    "double quote" },
  70         { "CENDQUOTE",  "a terminating quote" },
  71         { "CBQUOTE",    "backwards single quote" },
  72         { "CVAR",       "a dollar sign" },
  73         { "CENDVAR",    "a '}' character" },
  74         { "CLP",        "a left paren in arithmetic" },
  75         { "CRP",        "a right paren in arithmetic" },
  76         { "CEOF",       "end of file" },
  77         { "CCTL",       "like CWORD, except it must be escaped" },
  78         { "CSPCL",      "these terminate a word" },
  79         { "CIGN",       "character should be ignored" },
  80         { NULL,         NULL }
  81 };
  82
  83
  84 /*
  85  * Syntax classes for is_ functions.  Warning:  if you add new classes
  86  * you may have to change the definition of the is_in_name macro.
  87  */
  88 static const struct synclass is_entry[] = {
  89         { "ISDIGIT",    "a digit" },
  90         { "ISUPPER",    "an upper case letter" },
  91         { "ISLOWER",    "a lower case letter" },
  92         { "ISUNDER",    "an underscore" },
  93         { "ISSPECL",    "the name of a special parameter" },
  94         { NULL,         NULL }
  95 };
  96
  97 static const char writer[] = "\
  98 /*\n\
  99  * This file was generated by the mksyntax program.\n\
 100  */\n\
 101 \n";
 102
 103
 104 static FILE *cfile;
 105 static FILE *hfile;
 106
 107 static void add_default(void);
 108 static void finish(void);
 109 static void init(const char *);
 110 static void add(const char *, const char *);
 111 static void output_type_macros(void);
 112
 113 int
 114 main(int argc __unused, char **argv __unused)
 115 {
 116         int i;
 117         char buf[80];
 118         int pos;
 119
 120         /* Create output files */
 121         if ((cfile = fopen("syntax.c", "w")) == NULL) {
 122                 perror("syntax.c");
 123                 exit(2);
 124         }
 125         if ((hfile = fopen("syntax.h", "w")) == NULL) {
 126                 perror("syntax.h");
 127                 exit(2);
 128         }
 129         fputs(writer, hfile);
 130         fputs(writer, cfile);
 131
 132         fputs("#include <sys/cdefs.h>\n", hfile);
 133         fputs("#include <limits.h>\n\n", hfile);
 134
 135         /* Generate the #define statements in the header file */
 136         fputs("/* Syntax classes */\n", hfile);
 137         for (i = 0 ; synclass[i].name ; i++) {
 138                 sprintf(buf, "#define %s %d", synclass[i].name, i);
 139                 fputs(buf, hfile);
 140                 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
 141                         putc('\t', hfile);
 142                 fprintf(hfile, "/* %s */\n", synclass[i].comment);
 143         }
 144         putc('\n', hfile);
 145         fputs("/* Syntax classes for is_ functions */\n", hfile);
 146         for (i = 0 ; is_entry[i].name ; i++) {
 147                 sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i);
 148                 fputs(buf, hfile);
 149                 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
 150                         putc('\t', hfile);
 151                 fprintf(hfile, "/* %s */\n", is_entry[i].comment);
 152         }
 153         putc('\n', hfile);
 154         fputs("#define SYNBASE (1 - CHAR_MIN)\n", hfile);
 155         fputs("#define PEOF -SYNBASE\n\n", hfile);
 156         putc('\n', hfile);
 157         fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
 158         fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
 159         fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
 160         fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
 161         putc('\n', hfile);
 162         output_type_macros();           /* is_digit, etc. */
 163         putc('\n', hfile);
 164
 165         /* Generate the syntax tables. */
 166         fputs("#include \"parser.h\"\n", cfile);
 167         fputs("#include \"shell.h\"\n", cfile);
 168         fputs("#include \"syntax.h\"\n\n", cfile);
 169
 170         fputs("/* syntax table used when not in quotes */\n", cfile);
 171         init("basesyntax");
 172         add_default();
 173         add("\n", "CNL");
 174         add("\\", "CBACK");
 175         add("'", "CSQUOTE");
 176         add("\"", "CDQUOTE");
 177         add("`", "CBQUOTE");
 178         add("$", "CVAR");
 179         add("}", "CENDVAR");
 180         add("<>();&| \t", "CSPCL");
 181         finish();
 182
 183         fputs("\n/* syntax table used when in double quotes */\n", cfile);
 184         init("dqsyntax");
 185         add_default();
 186         add("\n", "CNL");
 187         add("\\", "CBACK");
 188         add("\"", "CENDQUOTE");
 189         add("`", "CBQUOTE");
 190         add("$", "CVAR");
 191         add("}", "CENDVAR");
 192         /* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
 193         add("!*?[]=~:/-^", "CCTL");
 194         finish();
 195
 196         fputs("\n/* syntax table used when in single quotes */\n", cfile);
 197         init("sqsyntax");
 198         add_default();
 199         add("\n", "CNL");
 200         add("\\", "CSBACK");
 201         add("'", "CENDQUOTE");
 202         /* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
 203         add("!*?[]=~:/-^", "CCTL");
 204         finish();
 205
 206         fputs("\n/* syntax table used when in arithmetic */\n", cfile);
 207         init("arisyntax");
 208         add_default();
 209         add("\n", "CNL");
 210         add("\\", "CBACK");
 211         add("`", "CBQUOTE");
 212         add("\"", "CIGN");
 213         add("$", "CVAR");
 214         add("}", "CENDVAR");
 215         add("(", "CLP");
 216         add(")", "CRP");
 217         finish();
 218
 219         fputs("\n/* character classification table */\n", cfile);
 220         init("is_type");
 221         add("0123456789", "ISDIGIT");
 222         add("abcdefghijklmnopqrstuvwxyz", "ISLOWER");
 223         add("ABCDEFGHIJKLMNOPQRSTUVWXYZ", "ISUPPER");
 224         add("_", "ISUNDER");
 225         add("#?$!-*@", "ISSPECL");
 226         finish();
 227
 228         exit(0);
 229 }
 230
 231
 232 /*
 233  * Output the header and declaration of a syntax table.
 234  */
 235
 236 static void
 237 init(const char *name)
 238 {
 239         fprintf(hfile, "extern const char %s[];\n", name);
 240         fprintf(cfile, "const char %s[SYNBASE + CHAR_MAX + 1] = {\n", name);
 241 }
 242
 243
 244 static void
 245 add_one(const char *key, const char *type)
 246 {
 247         fprintf(cfile, "\t[SYNBASE + %s] = %s,\n", key, type);
 248 }
 249
 250
 251 /*
 252  * Add default values to the syntax table.
 253  */
 254
 255 static void
 256 add_default(void)
 257 {
 258         add_one("PEOF",                "CEOF");
 259         add_one("CTLESC",              "CCTL");
 260         add_one("CTLVAR",              "CCTL");
 261         add_one("CTLENDVAR",           "CCTL");
 262         add_one("CTLBACKQ",            "CCTL");
 263         add_one("CTLBACKQ + CTLQUOTE", "CCTL");
 264         add_one("CTLARI",              "CCTL");
 265         add_one("CTLENDARI",           "CCTL");
 266         add_one("CTLQUOTEMARK",        "CCTL");
 267         add_one("CTLQUOTEEND",         "CCTL");
 268 }
 269
 270
 271 /*
 272  * Output the footer of a syntax table.
 273  */
 274
 275 static void
 276 finish(void)
 277 {
 278         fputs("};\n", cfile);
 279 }
 280
 281
 282 /*
 283  * Add entries to the syntax table.
 284  */
 285
 286 static void
 287 add(const char *p, const char *type)
 288 {
 289         for (; *p; ++p) {
 290                 char c = *p;
 291                 switch (c) {
 292                 case '\t': c = 't';  break;
 293                 case '\n': c = 'n';  break;
 294                 case '\'': c = '\''; break;
 295                 case '\\': c = '\\'; break;
 296
 297                 default:
 298                         fprintf(cfile, "\t[SYNBASE + '%c'] = %s,\n", c, type);
 299                         continue;
 300                 }
 301                 fprintf(cfile, "\t[SYNBASE + '\\%c'] = %s,\n", c, type);
 302         }
 303 }
 304
 305
 306 /*
 307  * Output character classification macros (e.g. is_digit).  If digits are
 308  * contiguous, we can test for them quickly.
 309  */
 310
 311 static const char *macro[] = {
 312         "#define is_digit(c)\t((unsigned int)((c) - '0') <= 9)",
 313         "#define is_eof(c)\t((c) == PEOF)",
 314         "#define is_alpha(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER))",
 315         "#define is_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER))",
 316         "#define is_in_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER|ISDIGIT))",
 317         "#define is_special(c)\t((is_type+SYNBASE)[(int)c] & (ISSPECL|ISDIGIT))",
 318         "#define digit_val(c)\t((c) - '0')",
 319         NULL
 320 };
 321
 322 static void
 323 output_type_macros(void)
 324 {
 325         const char **pp;
 326
 327         for (pp = macro ; *pp ; pp++)
 328                 fprintf(hfile, "%s\n", *pp);
 329 }