xref: /dragonfly/bin/sh/mksyntax.c (revision 20c2db9a)
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: head/bin/sh/mksyntax.c 246522 2013-02-07 22:42:33Z jilles $
39  */
40 
41 /*
42  * This program creates syntax.h and syntax.c.
43  */
44 
45 #include <stdio.h>
46 #include <stdlib.h>
47 #include <string.h>
48 #include "parser.h"
49 
50 
51 struct synclass {
52 	const char *name;
53 	const char *comment;
54 };
55 
56 /* Syntax classes */
57 struct synclass synclass[] = {
58 	{ "CWORD",	"character is nothing special" },
59 	{ "CNL",	"newline character" },
60 	{ "CBACK",	"a backslash character" },
61 	{ "CSBACK",	"a backslash character in single quotes" },
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 	{ "CIGN",       "character should be ignored" },
74 	{ NULL,		NULL }
75 };
76 
77 
78 /*
79  * Syntax classes for is_ functions.  Warning:  if you add new classes
80  * you may have to change the definition of the is_in_name macro.
81  */
82 struct synclass is_entry[] = {
83 	{ "ISDIGIT",	"a digit" },
84 	{ "ISUPPER",	"an upper case letter" },
85 	{ "ISLOWER",	"a lower case letter" },
86 	{ "ISUNDER",	"an underscore" },
87 	{ "ISSPECL",	"the name of a special parameter" },
88 	{ NULL, 	NULL }
89 };
90 
91 static char writer[] = "\
92 /*\n\
93  * This file was generated by the mksyntax program.\n\
94  */\n\
95 \n";
96 
97 
98 static FILE *cfile;
99 static FILE *hfile;
100 
101 static void add_default(void);
102 static void finish(void);
103 static void init(const char *);
104 static void add(const char *, const char *);
105 static void output_type_macros(void);
106 
107 int
108 main(int argc __unused, char **argv __unused)
109 {
110 	int i;
111 	char buf[80];
112 	int pos;
113 
114 	/* Create output files */
115 	if ((cfile = fopen("syntax.c", "w")) == NULL) {
116 		perror("syntax.c");
117 		exit(2);
118 	}
119 	if ((hfile = fopen("syntax.h", "w")) == NULL) {
120 		perror("syntax.h");
121 		exit(2);
122 	}
123 	fputs(writer, hfile);
124 	fputs(writer, cfile);
125 
126 	fputs("#include <sys/cdefs.h>\n", hfile);
127 	fputs("#include <limits.h>\n\n", hfile);
128 
129 	/* Generate the #define statements in the header file */
130 	fputs("/* Syntax classes */\n", hfile);
131 	for (i = 0 ; synclass[i].name ; i++) {
132 		sprintf(buf, "#define %s %d", synclass[i].name, i);
133 		fputs(buf, hfile);
134 		for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
135 			putc('\t', hfile);
136 		fprintf(hfile, "/* %s */\n", synclass[i].comment);
137 	}
138 	putc('\n', hfile);
139 	fputs("/* Syntax classes for is_ functions */\n", hfile);
140 	for (i = 0 ; is_entry[i].name ; i++) {
141 		sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << 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", is_entry[i].comment);
146 	}
147 	putc('\n', hfile);
148 	fputs("#define SYNBASE (1 - CHAR_MIN)\n", hfile);
149 	fputs("#define PEOF -SYNBASE\n\n", hfile);
150 	putc('\n', hfile);
151 	fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
152 	fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
153 	fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
154 	fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
155 	putc('\n', hfile);
156 	output_type_macros();		/* is_digit, etc. */
157 	putc('\n', hfile);
158 
159 	/* Generate the syntax tables. */
160 	fputs("#include \"parser.h\"\n", cfile);
161 	fputs("#include \"shell.h\"\n", cfile);
162 	fputs("#include \"syntax.h\"\n\n", cfile);
163 
164 	fputs("/* syntax table used when not in quotes */\n", cfile);
165 	init("basesyntax");
166 	add_default();
167 	add("\n", "CNL");
168 	add("\\", "CBACK");
169 	add("'", "CSQUOTE");
170 	add("\"", "CDQUOTE");
171 	add("`", "CBQUOTE");
172 	add("$", "CVAR");
173 	add("}", "CENDVAR");
174 	add("<>();&| \t", "CSPCL");
175 	finish();
176 
177 	fputs("\n/* syntax table used when in double quotes */\n", cfile);
178 	init("dqsyntax");
179 	add_default();
180 	add("\n", "CNL");
181 	add("\\", "CBACK");
182 	add("\"", "CENDQUOTE");
183 	add("`", "CBQUOTE");
184 	add("$", "CVAR");
185 	add("}", "CENDVAR");
186 	/* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
187 	add("!*?[]=~:/-^", "CCTL");
188 	finish();
189 
190 	fputs("\n/* syntax table used when in single quotes */\n", cfile);
191 	init("sqsyntax");
192 	add_default();
193 	add("\n", "CNL");
194 	add("\\", "CSBACK");
195 	add("'", "CENDQUOTE");
196 	/* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
197 	add("!*?[]=~:/-^", "CCTL");
198 	finish();
199 
200 	fputs("\n/* syntax table used when in arithmetic */\n", cfile);
201 	init("arisyntax");
202 	add_default();
203 	add("\n", "CNL");
204 	add("\\", "CBACK");
205 	add("`", "CBQUOTE");
206 	add("\"", "CIGN");
207 	add("$", "CVAR");
208 	add("}", "CENDVAR");
209 	add("(", "CLP");
210 	add(")", "CRP");
211 	finish();
212 
213 	fputs("\n/* character classification table */\n", cfile);
214 	init("is_type");
215 	add("0123456789", "ISDIGIT");
216 	add("abcdefghijklmnopqrstuvwxyz", "ISLOWER");
217 	add("ABCDEFGHIJKLMNOPQRSTUVWXYZ", "ISUPPER");
218 	add("_", "ISUNDER");
219 	add("#?$!-*@", "ISSPECL");
220 	finish();
221 
222 	exit(0);
223 }
224 
225 
226 /*
227  * Output the header and declaration of a syntax table.
228  */
229 
230 static void
231 init(const char *name)
232 {
233 	fprintf(hfile, "extern const char %s[];\n", name);
234 	fprintf(cfile, "const char %s[SYNBASE + CHAR_MAX + 1] = {\n", name);
235 }
236 
237 
238 static void
239 add_one(const char *key, const char *type)
240 {
241 	fprintf(cfile, "\t[SYNBASE + %s] = %s,\n", key, type);
242 }
243 
244 
245 /*
246  * Add default values to the syntax table.
247  */
248 
249 static void
250 add_default(void)
251 {
252 	add_one("PEOF",                "CEOF");
253 	add_one("CTLESC",              "CCTL");
254 	add_one("CTLVAR",              "CCTL");
255 	add_one("CTLENDVAR",           "CCTL");
256 	add_one("CTLBACKQ",            "CCTL");
257 	add_one("CTLBACKQ + CTLQUOTE", "CCTL");
258 	add_one("CTLARI",              "CCTL");
259 	add_one("CTLENDARI",           "CCTL");
260 	add_one("CTLQUOTEMARK",        "CCTL");
261 	add_one("CTLQUOTEEND",         "CCTL");
262 }
263 
264 
265 /*
266  * Output the footer of a syntax table.
267  */
268 
269 static void
270 finish(void)
271 {
272 	fputs("};\n", cfile);
273 }
274 
275 
276 /*
277  * Add entries to the syntax table.
278  */
279 
280 static void
281 add(const char *p, const char *type)
282 {
283 	for (; *p; ++p) {
284 		char c = *p;
285 		switch (c) {
286 		case '\t': c = 't';  break;
287 		case '\n': c = 'n';  break;
288 		case '\'': c = '\''; break;
289 		case '\\': c = '\\'; break;
290 
291 		default:
292 			fprintf(cfile, "\t[SYNBASE + '%c'] = %s,\n", c, type);
293 			continue;
294 		}
295 		fprintf(cfile, "\t[SYNBASE + '\\%c'] = %s,\n", c, type);
296 	}
297 }
298 
299 
300 /*
301  * Output character classification macros (e.g. is_digit).  If digits are
302  * contiguous, we can test for them quickly.
303  */
304 
305 static const char *macro[] = {
306 	"#define is_digit(c)\t((unsigned int)((c) - '0') <= 9)",
307 	"#define is_eof(c)\t((c) == PEOF)",
308 	"#define is_alpha(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER))",
309 	"#define is_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER))",
310 	"#define is_in_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER|ISDIGIT))",
311 	"#define is_special(c)\t((is_type+SYNBASE)[(int)c] & (ISSPECL|ISDIGIT))",
312 	"#define digit_val(c)\t((c) - '0')",
313 	NULL
314 };
315 
316 static void
317 output_type_macros(void)
318 {
319 	const char **pp;
320 
321 	for (pp = macro ; *pp ; pp++)
322 		fprintf(hfile, "%s\n", *pp);
323 }
324