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 334008 2018-05-21 21:52:48Z jilles $"); 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 { "CQNL", "newline character in quotes" }, 69 { "CBACK", "a backslash character" }, 70 { "CSBACK", "a backslash character in single quotes" }, 71 { "CSQUOTE", "single quote" }, 72 { "CDQUOTE", "double quote" }, 73 { "CENDQUOTE", "a terminating quote" }, 74 { "CBQUOTE", "backwards single quote" }, 75 { "CVAR", "a dollar sign" }, 76 { "CENDVAR", "a '}' character" }, 77 { "CLP", "a left paren in arithmetic" }, 78 { "CRP", "a right paren in arithmetic" }, 79 { "CEOF", "end of file" }, 80 { "CCTL", "like CWORD, except it must be escaped" }, 81 { "CSPCL", "these terminate a word" }, 82 { "CIGN", "character should be ignored" }, 83 { NULL, NULL } 84 }; 85 86 87 /* 88 * Syntax classes for is_ functions. Warning: if you add new classes 89 * you may have to change the definition of the is_in_name macro. 90 */ 91 static const struct synclass is_entry[] = { 92 { "ISDIGIT", "a digit" }, 93 { "ISUPPER", "an upper case letter" }, 94 { "ISLOWER", "a lower case letter" }, 95 { "ISUNDER", "an underscore" }, 96 { "ISSPECL", "the name of a special parameter" }, 97 { NULL, NULL } 98 }; 99 100 static const char writer[] = "\ 101 /*\n\ 102 * This file was generated by the mksyntax program.\n\ 103 */\n\ 104 \n"; 105 106 107 static FILE *cfile; 108 static FILE *hfile; 109 110 static void add_default(void); 111 static void finish(void); 112 static void init(const char *); 113 static void add(const char *, const char *); 114 static void output_type_macros(void); 115 116 int 117 main(int argc __unused, char **argv __unused) 118 { 119 int i; 120 char buf[80]; 121 int pos; 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 fputs("#include <sys/cdefs.h>\n", hfile); 136 fputs("#include <limits.h>\n\n", hfile); 137 138 /* Generate the #define statements in the header file */ 139 fputs("/* Syntax classes */\n", hfile); 140 for (i = 0 ; synclass[i].name ; i++) { 141 sprintf(buf, "#define %s %d", synclass[i].name, i); 142 fputs(buf, hfile); 143 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07) 144 putc('\t', hfile); 145 fprintf(hfile, "/* %s */\n", synclass[i].comment); 146 } 147 putc('\n', hfile); 148 fputs("/* Syntax classes for is_ functions */\n", hfile); 149 for (i = 0 ; is_entry[i].name ; i++) { 150 sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i); 151 fputs(buf, hfile); 152 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07) 153 putc('\t', hfile); 154 fprintf(hfile, "/* %s */\n", is_entry[i].comment); 155 } 156 putc('\n', hfile); 157 fputs("#define SYNBASE (1 - CHAR_MIN)\n", hfile); 158 fputs("#define PEOF -SYNBASE\n\n", hfile); 159 putc('\n', hfile); 160 fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile); 161 fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile); 162 fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile); 163 fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile); 164 putc('\n', hfile); 165 output_type_macros(); /* is_digit, etc. */ 166 putc('\n', hfile); 167 168 /* Generate the syntax tables. */ 169 fputs("#include \"parser.h\"\n", cfile); 170 fputs("#include \"shell.h\"\n", cfile); 171 fputs("#include \"syntax.h\"\n\n", cfile); 172 173 fputs("/* syntax table used when not in quotes */\n", cfile); 174 init("basesyntax"); 175 add_default(); 176 add("\n", "CNL"); 177 add("\\", "CBACK"); 178 add("'", "CSQUOTE"); 179 add("\"", "CDQUOTE"); 180 add("`", "CBQUOTE"); 181 add("$", "CVAR"); 182 add("}", "CENDVAR"); 183 add("<>();&| \t", "CSPCL"); 184 finish(); 185 186 fputs("\n/* syntax table used when in double quotes */\n", cfile); 187 init("dqsyntax"); 188 add_default(); 189 add("\n", "CQNL"); 190 add("\\", "CBACK"); 191 add("\"", "CENDQUOTE"); 192 add("`", "CBQUOTE"); 193 add("$", "CVAR"); 194 add("}", "CENDVAR"); 195 /* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */ 196 add("!*?[]=~:/-^", "CCTL"); 197 finish(); 198 199 fputs("\n/* syntax table used when in single quotes */\n", cfile); 200 init("sqsyntax"); 201 add_default(); 202 add("\n", "CQNL"); 203 add("\\", "CSBACK"); 204 add("'", "CENDQUOTE"); 205 /* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */ 206 add("!*?[]=~:/-^", "CCTL"); 207 finish(); 208 209 fputs("\n/* syntax table used when in arithmetic */\n", cfile); 210 init("arisyntax"); 211 add_default(); 212 add("\n", "CQNL"); 213 add("\\", "CBACK"); 214 add("`", "CBQUOTE"); 215 add("\"", "CIGN"); 216 add("$", "CVAR"); 217 add("}", "CENDVAR"); 218 add("(", "CLP"); 219 add(")", "CRP"); 220 finish(); 221 222 fputs("\n/* character classification table */\n", cfile); 223 init("is_type"); 224 add("0123456789", "ISDIGIT"); 225 add("abcdefghijklmnopqrstuvwxyz", "ISLOWER"); 226 add("ABCDEFGHIJKLMNOPQRSTUVWXYZ", "ISUPPER"); 227 add("_", "ISUNDER"); 228 add("#?$!-*@", "ISSPECL"); 229 finish(); 230 231 exit(0); 232 } 233 234 235 /* 236 * Output the header and declaration of a syntax table. 237 */ 238 239 static void 240 init(const char *name) 241 { 242 fprintf(hfile, "extern const char %s[];\n", name); 243 fprintf(cfile, "const char %s[SYNBASE + CHAR_MAX + 1] = {\n", name); 244 } 245 246 247 static void 248 add_one(const char *key, const char *type) 249 { 250 fprintf(cfile, "\t[SYNBASE + %s] = %s,\n", key, type); 251 } 252 253 254 /* 255 * Add default values to the syntax table. 256 */ 257 258 static void 259 add_default(void) 260 { 261 add_one("PEOF", "CEOF"); 262 add_one("CTLESC", "CCTL"); 263 add_one("CTLVAR", "CCTL"); 264 add_one("CTLENDVAR", "CCTL"); 265 add_one("CTLBACKQ", "CCTL"); 266 add_one("CTLBACKQ + CTLQUOTE", "CCTL"); 267 add_one("CTLARI", "CCTL"); 268 add_one("CTLENDARI", "CCTL"); 269 add_one("CTLQUOTEMARK", "CCTL"); 270 add_one("CTLQUOTEEND", "CCTL"); 271 } 272 273 274 /* 275 * Output the footer of a syntax table. 276 */ 277 278 static void 279 finish(void) 280 { 281 fputs("};\n", cfile); 282 } 283 284 285 /* 286 * Add entries to the syntax table. 287 */ 288 289 static void 290 add(const char *p, const char *type) 291 { 292 for (; *p; ++p) { 293 char c = *p; 294 switch (c) { 295 case '\t': c = 't'; break; 296 case '\n': c = 'n'; break; 297 case '\'': c = '\''; break; 298 case '\\': c = '\\'; break; 299 300 default: 301 fprintf(cfile, "\t[SYNBASE + '%c'] = %s,\n", c, type); 302 continue; 303 } 304 fprintf(cfile, "\t[SYNBASE + '\\%c'] = %s,\n", c, type); 305 } 306 } 307 308 309 /* 310 * Output character classification macros (e.g. is_digit). If digits are 311 * contiguous, we can test for them quickly. 312 */ 313 314 static const char *macro[] = { 315 "#define is_digit(c)\t((unsigned int)((c) - '0') <= 9)", 316 "#define is_eof(c)\t((c) == PEOF)", 317 "#define is_alpha(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER))", 318 "#define is_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER))", 319 "#define is_in_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER|ISDIGIT))", 320 "#define is_special(c)\t((is_type+SYNBASE)[(int)c] & (ISSPECL|ISDIGIT))", 321 "#define digit_val(c)\t((c) - '0')", 322 NULL 323 }; 324 325 static void 326 output_type_macros(void) 327 { 328 const char **pp; 329 330 for (pp = macro ; *pp ; pp++) 331 fprintf(hfile, "%s\n", *pp); 332 } 333