sh/mksyntax.c

   1 /*-
   2  * SPDX-License-Identifier: BSD-3-Clause
   3  *
   4  * Copyright (c) 1991, 1993
   5  *      The Regents of the University of California.  All rights reserved.
   6  *
   7  * This code is derived from software contributed to Berkeley by
   8  * Kenneth Almquist.
   9  *
  10  * Redistribution and use in source and binary forms, with or without
  11  * modification, are permitted provided that the following conditions
  12  * are met:
  13  * 1. Redistributions of source code must retain the above copyright
  14  *    notice, this list of conditions and the following disclaimer.
  15  * 2. Redistributions in binary form must reproduce the above copyright
  16  *    notice, this list of conditions and the following disclaimer in the
  17  *    documentation and/or other materials provided with the distribution.
  18  * 3. Neither the name of the University nor the names of its contributors
  19  *    may be used to endorse or promote products derived from this software
  20  *    without specific prior written permission.
  21  *
  22  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  23  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  24  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  25  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  26  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  27  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  28  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  29  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  30  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  31  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  32  * SUCH DAMAGE.
  33  */
  34
  35 #if 0
  36 #ifndef lint
  37 static char const copyright[] =
  38 "@(#) Copyright (c) 1991, 1993\n\
  39         The Regents of the University of California.  All rights reserved.\n";
  40 #endif /* not lint */
  41
  42 #ifndef lint
  43 static char sccsid[] = "@(#)mksyntax.c  8.2 (Berkeley) 5/4/95";
  44 #endif /* not lint */
  45 #endif
  46 #include <sys/cdefs.h>
  47 __FBSDID("$FreeBSD: head/bin/sh/mksyntax.c 326025 2017-11-20 19:49:47Z pfg $");
  48
  49 /*
  50  * This program creates syntax.h and syntax.c.
  51  */
  52
  53 #include <stdio.h>
  54 #include <stdlib.h>
  55 #include <string.h>
  56 #include "parser.h"
  57
  58
  59 struct synclass {
  60         const char *name;
  61         const char *comment;
  62 };
  63
  64 /* Syntax classes */
  65 static const struct synclass synclass[] = {
  66         { "CWORD",      "character is nothing special" },
  67         { "CNL",        "newline character" },
  68         { "CBACK",      "a backslash character" },
  69         { "CSBACK",     "a backslash character in single quotes" },
  70         { "CSQUOTE",    "single quote" },
  71         { "CDQUOTE",    "double quote" },
  72         { "CENDQUOTE",  "a terminating quote" },
  73         { "CBQUOTE",    "backwards single quote" },
  74         { "CVAR",       "a dollar sign" },
  75         { "CENDVAR",    "a '}' character" },
  76         { "CLP",        "a left paren in arithmetic" },
  77         { "CRP",        "a right paren in arithmetic" },
  78         { "CEOF",       "end of file" },
  79         { "CCTL",       "like CWORD, except it must be escaped" },
  80         { "CSPCL",      "these terminate a word" },
  81         { "CIGN",       "character should be ignored" },
  82         { NULL,         NULL }
  83 };
  84
  85
  86 /*
  87  * Syntax classes for is_ functions.  Warning:  if you add new classes
  88  * you may have to change the definition of the is_in_name macro.
  89  */
  90 static const struct synclass is_entry[] = {
  91         { "ISDIGIT",    "a digit" },
  92         { "ISUPPER",    "an upper case letter" },
  93         { "ISLOWER",    "a lower case letter" },
  94         { "ISUNDER",    "an underscore" },
  95         { "ISSPECL",    "the name of a special parameter" },
  96         { NULL,         NULL }
  97 };
  98
  99 static const char writer[] = "\
 100 /*\n\
 101  * This file was generated by the mksyntax program.\n\
 102  */\n\
 103 \n";
 104
 105
 106 static FILE *cfile;
 107 static FILE *hfile;
 108
 109 static void add_default(void);
 110 static void finish(void);
 111 static void init(const char *);
 112 static void add(const char *, const char *);
 113 static void output_type_macros(void);
 114
 115 int
 116 main(int argc __unused, char **argv __unused)
 117 {
 118         int i;
 119         char buf[80];
 120         int pos;
 121
 122         /* Create output files */
 123         if ((cfile = fopen("syntax.c", "w")) == NULL) {
 124                 perror("syntax.c");
 125                 exit(2);
 126         }
 127         if ((hfile = fopen("syntax.h", "w")) == NULL) {
 128                 perror("syntax.h");
 129                 exit(2);
 130         }
 131         fputs(writer, hfile);
 132         fputs(writer, cfile);
 133
 134         fputs("#include <sys/cdefs.h>\n", hfile);
 135         fputs("#include <limits.h>\n\n", hfile);
 136
 137         /* Generate the #define statements in the header file */
 138         fputs("/* Syntax classes */\n", hfile);
 139         for (i = 0 ; synclass[i].name ; i++) {
 140                 sprintf(buf, "#define %s %d", synclass[i].name, i);
 141                 fputs(buf, hfile);
 142                 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
 143                         putc('\t', hfile);
 144                 fprintf(hfile, "/* %s */\n", synclass[i].comment);
 145         }
 146         putc('\n', hfile);
 147         fputs("/* Syntax classes for is_ functions */\n", hfile);
 148         for (i = 0 ; is_entry[i].name ; i++) {
 149                 sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i);
 150                 fputs(buf, hfile);
 151                 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
 152                         putc('\t', hfile);
 153                 fprintf(hfile, "/* %s */\n", is_entry[i].comment);
 154         }
 155         putc('\n', hfile);
 156         fputs("#define SYNBASE (1 - CHAR_MIN)\n", hfile);
 157         fputs("#define PEOF -SYNBASE\n\n", hfile);
 158         putc('\n', hfile);
 159         fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
 160         fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
 161         fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
 162         fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
 163         putc('\n', hfile);
 164         output_type_macros();           /* is_digit, etc. */
 165         putc('\n', hfile);
 166
 167         /* Generate the syntax tables. */
 168         fputs("#include \"parser.h\"\n", cfile);
 169         fputs("#include \"shell.h\"\n", cfile);
 170         fputs("#include \"syntax.h\"\n\n", cfile);
 171
 172         fputs("/* syntax table used when not in quotes */\n", cfile);
 173         init("basesyntax");
 174         add_default();
 175         add("\n", "CNL");
 176         add("\\", "CBACK");
 177         add("'", "CSQUOTE");
 178         add("\"", "CDQUOTE");
 179         add("`", "CBQUOTE");
 180         add("$", "CVAR");
 181         add("}", "CENDVAR");
 182         add("<>();&| \t", "CSPCL");
 183         finish();
 184
 185         fputs("\n/* syntax table used when in double quotes */\n", cfile);
 186         init("dqsyntax");
 187         add_default();
 188         add("\n", "CNL");
 189         add("\\", "CBACK");
 190         add("\"", "CENDQUOTE");
 191         add("`", "CBQUOTE");
 192         add("$", "CVAR");
 193         add("}", "CENDVAR");
 194         /* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
 195         add("!*?[]=~:/-^", "CCTL");
 196         finish();
 197
 198         fputs("\n/* syntax table used when in single quotes */\n", cfile);
 199         init("sqsyntax");
 200         add_default();
 201         add("\n", "CNL");
 202         add("\\", "CSBACK");
 203         add("'", "CENDQUOTE");
 204         /* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
 205         add("!*?[]=~:/-^", "CCTL");
 206         finish();
 207
 208         fputs("\n/* syntax table used when in arithmetic */\n", cfile);
 209         init("arisyntax");
 210         add_default();
 211         add("\n", "CNL");
 212         add("\\", "CBACK");
 213         add("`", "CBQUOTE");
 214         add("\"", "CIGN");
 215         add("$", "CVAR");
 216         add("}", "CENDVAR");
 217         add("(", "CLP");
 218         add(")", "CRP");
 219         finish();
 220
 221         fputs("\n/* character classification table */\n", cfile);
 222         init("is_type");
 223         add("0123456789", "ISDIGIT");
 224         add("abcdefghijklmnopqrstuvwxyz", "ISLOWER");
 225         add("ABCDEFGHIJKLMNOPQRSTUVWXYZ", "ISUPPER");
 226         add("_", "ISUNDER");
 227         add("#?$!-*@", "ISSPECL");
 228         finish();
 229
 230         exit(0);
 231 }
 232
 233
 234 /*
 235  * Output the header and declaration of a syntax table.
 236  */
 237
 238 static void
 239 init(const char *name)
 240 {
 241         fprintf(hfile, "extern const char %s[];\n", name);
 242         fprintf(cfile, "const char %s[SYNBASE + CHAR_MAX + 1] = {\n", name);
 243 }
 244
 245
 246 static void
 247 add_one(const char *key, const char *type)
 248 {
 249         fprintf(cfile, "\t[SYNBASE + %s] = %s,\n", key, type);
 250 }
 251
 252
 253 /*
 254  * Add default values to the syntax table.
 255  */
 256
 257 static void
 258 add_default(void)
 259 {
 260         add_one("PEOF",                "CEOF");
 261         add_one("CTLESC",              "CCTL");
 262         add_one("CTLVAR",              "CCTL");
 263         add_one("CTLENDVAR",           "CCTL");
 264         add_one("CTLBACKQ",            "CCTL");
 265         add_one("CTLBACKQ + CTLQUOTE", "CCTL");
 266         add_one("CTLARI",              "CCTL");
 267         add_one("CTLENDARI",           "CCTL");
 268         add_one("CTLQUOTEMARK",        "CCTL");
 269         add_one("CTLQUOTEEND",         "CCTL");
 270 }
 271
 272
 273 /*
 274  * Output the footer of a syntax table.
 275  */
 276
 277 static void
 278 finish(void)
 279 {
 280         fputs("};\n", cfile);
 281 }
 282
 283
 284 /*
 285  * Add entries to the syntax table.
 286  */
 287
 288 static void
 289 add(const char *p, const char *type)
 290 {
 291         for (; *p; ++p) {
 292                 char c = *p;
 293                 switch (c) {
 294                 case '\t': c = 't';  break;
 295                 case '\n': c = 'n';  break;
 296                 case '\'': c = '\''; break;
 297                 case '\\': c = '\\'; break;
 298
 299                 default:
 300                         fprintf(cfile, "\t[SYNBASE + '%c'] = %s,\n", c, type);
 301                         continue;
 302                 }
 303                 fprintf(cfile, "\t[SYNBASE + '\\%c'] = %s,\n", c, type);
 304         }
 305 }
 306
 307
 308 /*
 309  * Output character classification macros (e.g. is_digit).  If digits are
 310  * contiguous, we can test for them quickly.
 311  */
 312
 313 static const char *macro[] = {
 314         "#define is_digit(c)\t((unsigned int)((c) - '0') <= 9)",
 315         "#define is_eof(c)\t((c) == PEOF)",
 316         "#define is_alpha(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER))",
 317         "#define is_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER))",
 318         "#define is_in_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER|ISDIGIT))",
 319         "#define is_special(c)\t((is_type+SYNBASE)[(int)c] & (ISSPECL|ISDIGIT))",
 320         "#define digit_val(c)\t((c) - '0')",
 321         NULL
 322 };
 323
 324 static void
 325 output_type_macros(void)
 326 {
 327         const char **pp;
 328
 329         for (pp = macro ; *pp ; pp++)
 330                 fprintf(hfile, "%s\n", *pp);
 331 }