xref: /dragonfly/bin/sh/mksyntax.c (revision d4ef6694)
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  * @(#) Copyright (c) 1991, 1993 The Regents of the University of California.  All rights reserved.
33  * @(#)mksyntax.c	8.2 (Berkeley) 5/4/95
34  * $FreeBSD: head/bin/sh/mksyntax.c 246522 2013-02-07 22:42:33Z jilles $
35  */
36 
37 /*
38  * This program creates syntax.h and syntax.c.
39  */
40 
41 #include <stdio.h>
42 #include <stdlib.h>
43 #include <string.h>
44 #include "parser.h"
45 
46 
47 struct synclass {
48 	const char *name;
49 	const char *comment;
50 };
51 
52 /* Syntax classes */
53 struct synclass synclass[] = {
54 	{ "CWORD",	"character is nothing special" },
55 	{ "CNL",	"newline character" },
56 	{ "CBACK",	"a backslash character" },
57 	{ "CSBACK",	"a backslash character in single quotes" },
58 	{ "CSQUOTE",	"single quote" },
59 	{ "CDQUOTE",	"double quote" },
60 	{ "CENDQUOTE",	"a terminating quote" },
61 	{ "CBQUOTE",	"backwards single quote" },
62 	{ "CVAR",	"a dollar sign" },
63 	{ "CENDVAR",	"a '}' character" },
64 	{ "CLP",	"a left paren in arithmetic" },
65 	{ "CRP",	"a right paren in arithmetic" },
66 	{ "CEOF",	"end of file" },
67 	{ "CCTL",	"like CWORD, except it must be escaped" },
68 	{ "CSPCL",	"these terminate a word" },
69 	{ "CIGN",       "character should be ignored" },
70 	{ NULL,		NULL }
71 };
72 
73 
74 /*
75  * Syntax classes for is_ functions.  Warning:  if you add new classes
76  * you may have to change the definition of the is_in_name macro.
77  */
78 struct synclass is_entry[] = {
79 	{ "ISDIGIT",	"a digit" },
80 	{ "ISUPPER",	"an upper case letter" },
81 	{ "ISLOWER",	"a lower case letter" },
82 	{ "ISUNDER",	"an underscore" },
83 	{ "ISSPECL",	"the name of a special parameter" },
84 	{ NULL, 	NULL }
85 };
86 
87 static char writer[] = "\
88 /*\n\
89  * This file was generated by the mksyntax program.\n\
90  */\n\
91 \n";
92 
93 
94 static FILE *cfile;
95 static FILE *hfile;
96 
97 static void add_default(void);
98 static void finish(void);
99 static void init(const char *);
100 static void add(const char *, const char *);
101 static void output_type_macros(void);
102 
103 int
104 main(int argc __unused, char **argv __unused)
105 {
106 	int i;
107 	char buf[80];
108 	int pos;
109 
110 	/* Create output files */
111 	if ((cfile = fopen("syntax.c", "w")) == NULL) {
112 		perror("syntax.c");
113 		exit(2);
114 	}
115 	if ((hfile = fopen("syntax.h", "w")) == NULL) {
116 		perror("syntax.h");
117 		exit(2);
118 	}
119 	fputs(writer, hfile);
120 	fputs(writer, cfile);
121 
122 	fputs("#include <sys/cdefs.h>\n", hfile);
123 	fputs("#include <limits.h>\n\n", hfile);
124 
125 	/* Generate the #define statements in the header file */
126 	fputs("/* Syntax classes */\n", hfile);
127 	for (i = 0 ; synclass[i].name ; i++) {
128 		sprintf(buf, "#define %s %d", synclass[i].name, i);
129 		fputs(buf, hfile);
130 		for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
131 			putc('\t', hfile);
132 		fprintf(hfile, "/* %s */\n", synclass[i].comment);
133 	}
134 	putc('\n', hfile);
135 	fputs("/* Syntax classes for is_ functions */\n", hfile);
136 	for (i = 0 ; is_entry[i].name ; i++) {
137 		sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i);
138 		fputs(buf, hfile);
139 		for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
140 			putc('\t', hfile);
141 		fprintf(hfile, "/* %s */\n", is_entry[i].comment);
142 	}
143 	putc('\n', hfile);
144 	fputs("#define SYNBASE (1 - CHAR_MIN)\n", hfile);
145 	fputs("#define PEOF -SYNBASE\n\n", hfile);
146 	putc('\n', hfile);
147 	fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
148 	fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
149 	fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
150 	fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
151 	putc('\n', hfile);
152 	output_type_macros();		/* is_digit, etc. */
153 	putc('\n', hfile);
154 
155 	/* Generate the syntax tables. */
156 	fputs("#include \"parser.h\"\n", cfile);
157 	fputs("#include \"shell.h\"\n", cfile);
158 	fputs("#include \"syntax.h\"\n\n", cfile);
159 
160 	fputs("/* syntax table used when not in quotes */\n", cfile);
161 	init("basesyntax");
162 	add_default();
163 	add("\n", "CNL");
164 	add("\\", "CBACK");
165 	add("'", "CSQUOTE");
166 	add("\"", "CDQUOTE");
167 	add("`", "CBQUOTE");
168 	add("$", "CVAR");
169 	add("}", "CENDVAR");
170 	add("<>();&| \t", "CSPCL");
171 	finish();
172 
173 	fputs("\n/* syntax table used when in double quotes */\n", cfile);
174 	init("dqsyntax");
175 	add_default();
176 	add("\n", "CNL");
177 	add("\\", "CBACK");
178 	add("\"", "CENDQUOTE");
179 	add("`", "CBQUOTE");
180 	add("$", "CVAR");
181 	add("}", "CENDVAR");
182 	/* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
183 	add("!*?[]=~:/-^", "CCTL");
184 	finish();
185 
186 	fputs("\n/* syntax table used when in single quotes */\n", cfile);
187 	init("sqsyntax");
188 	add_default();
189 	add("\n", "CNL");
190 	add("\\", "CSBACK");
191 	add("'", "CENDQUOTE");
192 	/* ':/' for tilde expansion, '-^]' for [a\-x] pattern ranges */
193 	add("!*?[]=~:/-^", "CCTL");
194 	finish();
195 
196 	fputs("\n/* syntax table used when in arithmetic */\n", cfile);
197 	init("arisyntax");
198 	add_default();
199 	add("\n", "CNL");
200 	add("\\", "CBACK");
201 	add("`", "CBQUOTE");
202 	add("\"", "CIGN");
203 	add("$", "CVAR");
204 	add("}", "CENDVAR");
205 	add("(", "CLP");
206 	add(")", "CRP");
207 	finish();
208 
209 	fputs("\n/* character classification table */\n", cfile);
210 	init("is_type");
211 	add("0123456789", "ISDIGIT");
212 	add("abcdefghijklmnopqrstuvwxyz", "ISLOWER");
213 	add("ABCDEFGHIJKLMNOPQRSTUVWXYZ", "ISUPPER");
214 	add("_", "ISUNDER");
215 	add("#?$!-*@", "ISSPECL");
216 	finish();
217 
218 	exit(0);
219 }
220 
221 
222 /*
223  * Output the header and declaration of a syntax table.
224  */
225 
226 static void
227 init(const char *name)
228 {
229 	fprintf(hfile, "extern const char %s[];\n", name);
230 	fprintf(cfile, "const char %s[SYNBASE + CHAR_MAX + 1] = {\n", name);
231 }
232 
233 
234 static void
235 add_one(const char *key, const char *type)
236 {
237 	fprintf(cfile, "\t[SYNBASE + %s] = %s,\n", key, type);
238 }
239 
240 
241 /*
242  * Add default values to the syntax table.
243  */
244 
245 static void
246 add_default(void)
247 {
248 	add_one("PEOF",                "CEOF");
249 	add_one("CTLESC",              "CCTL");
250 	add_one("CTLVAR",              "CCTL");
251 	add_one("CTLENDVAR",           "CCTL");
252 	add_one("CTLBACKQ",            "CCTL");
253 	add_one("CTLBACKQ + CTLQUOTE", "CCTL");
254 	add_one("CTLARI",              "CCTL");
255 	add_one("CTLENDARI",           "CCTL");
256 	add_one("CTLQUOTEMARK",        "CCTL");
257 	add_one("CTLQUOTEEND",         "CCTL");
258 }
259 
260 
261 /*
262  * Output the footer of a syntax table.
263  */
264 
265 static void
266 finish(void)
267 {
268 	fputs("};\n", cfile);
269 }
270 
271 
272 /*
273  * Add entries to the syntax table.
274  */
275 
276 static void
277 add(const char *p, const char *type)
278 {
279 	for (; *p; ++p) {
280 		char c = *p;
281 		switch (c) {
282 		case '\t': c = 't';  break;
283 		case '\n': c = 'n';  break;
284 		case '\'': c = '\''; break;
285 		case '\\': c = '\\'; break;
286 
287 		default:
288 			fprintf(cfile, "\t[SYNBASE + '%c'] = %s,\n", c, type);
289 			continue;
290 		}
291 		fprintf(cfile, "\t[SYNBASE + '\\%c'] = %s,\n", c, type);
292 	}
293 }
294 
295 
296 /*
297  * Output character classification macros (e.g. is_digit).  If digits are
298  * contiguous, we can test for them quickly.
299  */
300 
301 static const char *macro[] = {
302 	"#define is_digit(c)\t((unsigned int)((c) - '0') <= 9)",
303 	"#define is_eof(c)\t((c) == PEOF)",
304 	"#define is_alpha(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER))",
305 	"#define is_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER))",
306 	"#define is_in_name(c)\t((is_type+SYNBASE)[(int)c] & (ISUPPER|ISLOWER|ISUNDER|ISDIGIT))",
307 	"#define is_special(c)\t((is_type+SYNBASE)[(int)c] & (ISSPECL|ISDIGIT))",
308 	"#define digit_val(c)\t((c) - '0')",
309 	NULL
310 };
311 
312 static void
313 output_type_macros(void)
314 {
315 	const char **pp;
316 
317 	for (pp = macro ; *pp ; pp++)
318 		fprintf(hfile, "%s\n", *pp);
319 }
320