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/*
36 * This program creates syntax.h and syntax.c.
37 */
38
39#include <stdio.h>
40#include <stdlib.h>
41#include <string.h>
42#include "parser.h"
43
44#ifndef __unused
45#define __unused __attribute__((__unused__))
46#endif
47
48
49struct synclass {
50 const char *name;
51 const char *comment;
52};
53
54/* Syntax classes */
55static const struct synclass synclass[] = {
56 { "CWORD", "character is nothing special" },
57 { "CNL", "newline character" },
58 { "CQNL", "newline character in quotes" },
59 { "CBACK", "a backslash character" },
60 { "CSBACK", "a backslash character in single quotes" },
61 { "CSQUOTE", "single quote" },
62 { "CDQUOTE", "double quote" },
63 { "CENDQUOTE", "a terminating quote" },
64 { "CBQUOTE", "backwards single quote" },
65 { "CVAR", "a dollar sign" },
66 { "CENDVAR", "a '}' character" },
67 { "CLP", "a left paren in arithmetic" },
68 { "CRP", "a right paren in arithmetic" },
69 { "CEOF", "end of file" },
70 { "CCTL", "like CWORD, except it must be escaped" },
71 { "CSPCL", "these terminate a word" },
72 { "CIGN", "character should be ignored" },
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 */
81static const 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
90static const char writer[] = "\
91/*\n\
92 * This file was generated by the mksyntax program.\n\
93 */\n\
94\n";
95
96
97static FILE *cfile;
98static FILE *hfile;
99
100static void add_default(void);
101static void finish(void);
102static void init(const char *);
103static void add(const char *, const char *);
104static void output_type_macros(void);
105
106int
107main(int argc __unused, char **argv __unused)
108{
109 int i;
110 char buf[80];
111 int pos;
112
113 /* Create output files */
114 if ((cfile = fopen("syntax.c", "w")) == NULL) {
115 perror("syntax.c");
116 exit(2);
117 }
118 if ((hfile = fopen("syntax.h", "w")) == NULL) {
119 perror("syntax.h");
120 exit(2);
121 }
122 fputs(writer, hfile);
123 fputs(writer, cfile);
124
125 fputs("#include <limits.h>\n\n", hfile);
126
127 /* Generate the #define statements in the header file */
128 fputs("/* Syntax classes */\n", hfile);
129 for (i = 0 ; synclass[i].name ; i++) {
130 if (strcmp(synclass[i].name, "CEOF") == 0) {
131 fputs("#ifdef CEOF\n#undef CEOF\n#endif\n", hfile);
132 }
133 sprintf(buf, "#define %s %d", synclass[i].name, i);
134 fputs(buf, hfile);
135 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
136 putc('\t', hfile);
137 fprintf(hfile, "/* %s */\n", synclass[i].comment);
138 }
139 putc('\n', hfile);
140 fputs("/* Syntax classes for is_ functions */\n", hfile);
141 for (i = 0 ; is_entry[i].name ; i++) {
142 sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i);
143 fputs(buf, hfile);
144 for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
145 putc('\t', hfile);
146 fprintf(hfile, "/* %s */\n", is_entry[i].comment);
147 }
148 putc('\n', hfile);
149 fputs("#define SYNBASE (1 - CHAR_MIN)\n", hfile);
150 fputs("#define PEOF -SYNBASE\n\n", hfile);
151 putc('\n', hfile);
152 fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
153 fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
154 fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
155 fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
156 putc('\n', hfile);
157 output_type_macros(); /* is_digit, etc. */
158 putc('\n', hfile);
159
160 /* Generate the syntax tables. */
161 fputs("#include \"parser.h\"\n", cfile);
162 fputs("#include \"shell.h\"\n", cfile);
163 fputs("#include \"syntax.h\"\n\n", cfile);
164
165 fputs("/* syntax table used when not in quotes */\n", cfile);
166 init("basesyntax");
167 add_default();
168 add("\n", "CNL");
169 add("\\", "CBACK");
170 add("'", "CSQUOTE");
171 add("\"", "CDQUOTE");
172 add("`", "CBQUOTE");
173 add("$", "CVAR");
174 add("}", "CENDVAR");
175 add("<>();&| \t", "CSPCL");
176 finish();
177
178 fputs("\n/* syntax table used when in double quotes */\n", cfile);
179 init("dqsyntax");
180 add_default();
181 add("\n", "CQNL");
182 add("\\", "CBACK");
183 add("\"", "CENDQUOTE");
184 add("`", "CBQUOTE");
185 add("$", "CVAR");
186 add("}", "CENDVAR");
187 /* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
188 add("!*?[]=~:/-^", "CCTL");
189 finish();
190
191 fputs("\n/* syntax table used when in single quotes */\n", cfile);
192 init("sqsyntax");
193 add_default();
194 add("\n", "CQNL");
195 add("\\", "CSBACK");
196 add("'", "CENDQUOTE");
197 /* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
198 add("!*?[]=~:/-^", "CCTL");
199 finish();
200
201 fputs("\n/* syntax table used when in arithmetic */\n", cfile);
202 init("arisyntax");
203 add_default();
204 add("\n", "CQNL");
205 add("\\", "CBACK");
206 add("`", "CBQUOTE");
207 add("\"", "CIGN");
208 add("$", "CVAR");
209 add("}", "CENDVAR");
210 add("(", "CLP");
211 add(")", "CRP");
212 finish();
213
214 fputs("\n/* character classification table */\n", cfile);
215 init("is_type");
216 add("0123456789", "ISDIGIT");
217 add("abcdefghijklmnopqrstuvwxyz", "ISLOWER");
218 add("ABCDEFGHIJKLMNOPQRSTUVWXYZ", "ISUPPER");
219 add("_", "ISUNDER");
220 add("#?$!-*@", "ISSPECL");
221 finish();
222
223 exit(0);
224}
225
226
227/*
228 * Output the header and declaration of a syntax table.
229 */
230
231static void
232init(const char *name)
233{
234 fprintf(hfile, "extern const char %s[];\n", name);
235 fprintf(cfile, "const char %s[SYNBASE + CHAR_MAX + 1] = {\n", name);
236}
237
238
239static void
240add_one(const char *key, const char *type)
241{
242 fprintf(cfile, "\t[SYNBASE + %s] = %s,\n", key, type);
243}
244
245
246/*
247 * Add default values to the syntax table.
248 */
249
250static void
251add_default(void)
252{
253 add_one("PEOF", "CEOF");
254 add_one("CTLESC", "CCTL");
255 add_one("CTLVAR", "CCTL");
256 add_one("CTLENDVAR", "CCTL");
257 add_one("CTLBACKQ", "CCTL");
258 add_one("CTLBACKQ + CTLQUOTE", "CCTL");
259 add_one("CTLARI", "CCTL");
260 add_one("CTLENDARI", "CCTL");
261 add_one("CTLQUOTEMARK", "CCTL");
262 add_one("CTLQUOTEEND", "CCTL");
263}
264
265
266/*
267 * Output the footer of a syntax table.
268 */
269
270static void
271finish(void)
272{
273 fputs("};\n", cfile);
274}
275
276
277/*
278 * Add entries to the syntax table.
279 */
280
281static void
282add(const char *p, const char *type)
283{
284 for (; *p; ++p) {
285 char c = *p;
286 switch (c) {
287 case '\t': c = 't'; break;
288 case '\n': c = 'n'; break;
289 case '\'': c = '\''; break;
290 case '\\': c = '\\'; break;
291
292 default:
293 fprintf(cfile, "\t[SYNBASE + '%c'] = %s,\n", c, type);
294 continue;
295 }
296 fprintf(cfile, "\t[SYNBASE + '\\%c'] = %s,\n", c, type);
297 }
298}
299
300
301/*
302 * Output character classification macros (e.g. is_digit). If digits are
303 * contiguous, we can test for them quickly.
304 */
305
306static const char *macro[] = {
307 "#define is_digit(c)\t((unsigned int)((c) - '0') <= 9)",
308 "#define is_eof(c)\t((c) == PEOF)",
309 "#define is_alpha(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER))",
310 "#define is_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER))",
311 "#define is_in_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER|ISDIGIT))",
312 "#define is_special(c)\t((is_type+SYNBASE)[(int)c] & (ISSPECL|ISDIGIT))",
313 "#define digit_val(c)\t((c) - '0')",
314 NULL
315};
316
317static void
318output_type_macros(void)
319{
320 const char **pp;
321
322 for (pp = macro ; *pp ; pp++)
323 fprintf(hfile, "%s\n", *pp);
324}
