mksyntax.c revision 99110 
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
37#ifndef lint
38static char const copyright[] =
39"@(#) Copyright (c) 1991, 1993\n\
40	The Regents of the University of California.  All rights reserved.\n";
41#endif /* not lint */
42
43#ifndef lint
44#if 0
45static char sccsid[] = "@(#)mksyntax.c	8.2 (Berkeley) 5/4/95";
46#endif
47#endif /* not lint */
48#include <sys/cdefs.h>
49__FBSDID("$FreeBSD: head/bin/sh/mksyntax.c 99110 2002-06-30 05:15:05Z obrien $");
50
51/*
52 * This program creates syntax.h and syntax.c.
53 */
54
55#include <stdio.h>
56#include <stdlib.h>
57#include <string.h>
58#include "parser.h"
59
60
61struct synclass {
62	char *name;
63	char *comment;
64};
65
66/* Syntax classes */
67struct synclass synclass[] = {
68	{ "CWORD",	"character is nothing special" },
69	{ "CNL",	"newline character" },
70	{ "CBACK",	"a backslash character" },
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	{ 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 */
90struct 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
99static char writer[] = "\
100/*\n\
101 * This file was generated by the mksyntax program.\n\
102 */\n\
103\n";
104
105
106static FILE *cfile;
107static FILE *hfile;
108static char *syntax[513];
109static int base;
110static int size;	/* number of values which a char variable can have */
111static int nbits;	/* number of bits in a character */
112static int digit_contig;/* true if digits are contiguous */
113
114static void filltable(char *);
115static void init(void);
116static void add(char *, char *);
117static void print(char *);
118static void output_type_macros(void);
119static void digit_convert(void);
120
121int
122main(int argc __unused, char **argv __unused)
123{
124	char c;
125	char d;
126	int sign;
127	int i;
128	char buf[80];
129	int pos;
130	static char digit[] = "0123456789";
131
132	/* Create output files */
133	if ((cfile = fopen("syntax.c", "w")) == NULL) {
134		perror("syntax.c");
135		exit(2);
136	}
137	if ((hfile = fopen("syntax.h", "w")) == NULL) {
138		perror("syntax.h");
139		exit(2);
140	}
141	fputs(writer, hfile);
142	fputs(writer, cfile);
143
144	/* Determine the characteristics of chars. */
145	c = -1;
146	if (c < 0)
147		sign = 1;
148	else
149		sign = 0;
150	for (nbits = 1 ; ; nbits++) {
151		d = (1 << nbits) - 1;
152		if (d == c)
153			break;
154	}
155#if 0
156	printf("%s %d bit chars\n", sign? "signed" : "unsigned", nbits);
157#endif
158	if (nbits > 9) {
159		fputs("Characters can't have more than 9 bits\n", stderr);
160		exit(2);
161	}
162	size = (1 << nbits) + 1;
163	base = 1;
164	if (sign)
165		base += 1 << (nbits - 1);
166	digit_contig = 1;
167	for (i = 0 ; i < 10 ; i++) {
168		if (digit[i] != '0' + i)
169			digit_contig = 0;
170	}
171
172	fputs("#include <sys/cdefs.h>\n", hfile);
173	fputs("#include <ctype.h>\n", hfile);
174
175	/* Generate the #define statements in the header file */
176	fputs("/* Syntax classes */\n", hfile);
177	for (i = 0 ; synclass[i].name ; i++) {
178		sprintf(buf, "#define %s %d", synclass[i].name, i);
179		fputs(buf, hfile);
180		for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
181			putc('\t', hfile);
182		fprintf(hfile, "/* %s */\n", synclass[i].comment);
183	}
184	putc('\n', hfile);
185	fputs("/* Syntax classes for is_ functions */\n", hfile);
186	for (i = 0 ; is_entry[i].name ; i++) {
187		sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i);
188		fputs(buf, hfile);
189		for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
190			putc('\t', hfile);
191		fprintf(hfile, "/* %s */\n", is_entry[i].comment);
192	}
193	putc('\n', hfile);
194	fprintf(hfile, "#define SYNBASE %d\n", base);
195	fprintf(hfile, "#define PEOF %d\n\n", -base);
196	putc('\n', hfile);
197	fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
198	fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
199	fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
200	fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
201	putc('\n', hfile);
202	output_type_macros();		/* is_digit, etc. */
203	putc('\n', hfile);
204
205	/* Generate the syntax tables. */
206	fputs("#include \"shell.h\"\n", cfile);
207	fputs("#include \"syntax.h\"\n\n", cfile);
208	init();
209	fputs("/* syntax table used when not in quotes */\n", cfile);
210	add("\n", "CNL");
211	add("\\", "CBACK");
212	add("'", "CSQUOTE");
213	add("\"", "CDQUOTE");
214	add("`", "CBQUOTE");
215	add("$", "CVAR");
216	add("}", "CENDVAR");
217	add("<>();&| \t", "CSPCL");
218	print("basesyntax");
219	init();
220	fputs("\n/* syntax table used when in double quotes */\n", cfile);
221	add("\n", "CNL");
222	add("\\", "CBACK");
223	add("\"", "CENDQUOTE");
224	add("`", "CBQUOTE");
225	add("$", "CVAR");
226	add("}", "CENDVAR");
227	/* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
228	add("!*?[=~:/-", "CCTL");
229	print("dqsyntax");
230	init();
231	fputs("\n/* syntax table used when in single quotes */\n", cfile);
232	add("\n", "CNL");
233	add("'", "CENDQUOTE");
234	/* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
235	add("!*?[=~:/-", "CCTL");
236	print("sqsyntax");
237	init();
238	fputs("\n/* syntax table used when in arithmetic */\n", cfile);
239	add("\n", "CNL");
240	add("\\", "CBACK");
241	add("`", "CBQUOTE");
242	add("'", "CSQUOTE");
243	add("\"", "CDQUOTE");
244	add("$", "CVAR");
245	add("}", "CENDVAR");
246	add("(", "CLP");
247	add(")", "CRP");
248	print("arisyntax");
249	filltable("0");
250	fputs("\n/* character classification table */\n", cfile);
251	add("0123456789", "ISDIGIT");
252	add("abcdefghijklmnopqrstucvwxyz", "ISLOWER");
253	add("ABCDEFGHIJKLMNOPQRSTUCVWXYZ", "ISUPPER");
254	add("_", "ISUNDER");
255	add("#?$!-*@", "ISSPECL");
256	print("is_type");
257	if (! digit_contig)
258		digit_convert();
259	exit(0);
260}
261
262
263
264/*
265 * Clear the syntax table.
266 */
267
268static void
269filltable(char *dftval)
270{
271	int i;
272
273	for (i = 0 ; i < size ; i++)
274		syntax[i] = dftval;
275}
276
277
278/*
279 * Initialize the syntax table with default values.
280 */
281
282static void
283init(void)
284{
285	filltable("CWORD");
286	syntax[0] = "CEOF";
287	syntax[base + CTLESC] = "CCTL";
288	syntax[base + CTLVAR] = "CCTL";
289	syntax[base + CTLENDVAR] = "CCTL";
290	syntax[base + CTLBACKQ] = "CCTL";
291	syntax[base + CTLBACKQ + CTLQUOTE] = "CCTL";
292	syntax[base + CTLARI] = "CCTL";
293	syntax[base + CTLENDARI] = "CCTL";
294	syntax[base + CTLQUOTEMARK] = "CCTL";
295}
296
297
298/*
299 * Add entries to the syntax table.
300 */
301
302static void
303add(char *p, char *type)
304{
305	while (*p)
306		syntax[*p++ + base] = type;
307}
308
309
310
311/*
312 * Output the syntax table.
313 */
314
315static void
316print(char *name)
317{
318	int i;
319	int col;
320
321	fprintf(hfile, "extern const char %s[];\n", name);
322	fprintf(cfile, "const char %s[%d] = {\n", name, size);
323	col = 0;
324	for (i = 0 ; i < size ; i++) {
325		if (i == 0) {
326			fputs("      ", cfile);
327		} else if ((i & 03) == 0) {
328			fputs(",\n      ", cfile);
329			col = 0;
330		} else {
331			putc(',', cfile);
332			while (++col < 9 * (i & 03))
333				putc(' ', cfile);
334		}
335		fputs(syntax[i], cfile);
336		col += strlen(syntax[i]);
337	}
338	fputs("\n};\n", cfile);
339}
340
341
342
343/*
344 * Output character classification macros (e.g. is_digit).  If digits are
345 * contiguous, we can test for them quickly.
346 */
347
348static char *macro[] = {
349	"#define is_digit(c)\t((is_type+SYNBASE)[c] & ISDIGIT)",
350	"#define is_alpha(c)\t((c) != PEOF && ((c) < CTLESC || (c) > CTLQUOTEMARK) && isalpha((unsigned char) (c)))",
351	"#define is_name(c)\t((c) != PEOF && ((c) < CTLESC || (c) > CTLQUOTEMARK) && ((c) == '_' || isalpha((unsigned char) (c))))",
352	"#define is_in_name(c)\t((c) != PEOF && ((c) < CTLESC || (c) > CTLQUOTEMARK) && ((c) == '_' || isalnum((unsigned char) (c))))",
353	"#define is_special(c)\t((is_type+SYNBASE)[c] & (ISSPECL|ISDIGIT))",
354	NULL
355};
356
357static void
358output_type_macros(void)
359{
360	char **pp;
361
362	if (digit_contig)
363		macro[0] = "#define is_digit(c)\t((unsigned)((c) - '0') <= 9)";
364	for (pp = macro ; *pp ; pp++)
365		fprintf(hfile, "%s\n", *pp);
366	if (digit_contig)
367		fputs("#define digit_val(c)\t((c) - '0')\n", hfile);
368	else
369		fputs("#define digit_val(c)\t(digit_value[c])\n", hfile);
370}
371
372
373
374/*
375 * Output digit conversion table (if digits are not contiguous).
376 */
377
378static void
379digit_convert(void)
380{
381	int maxdigit;
382	static char digit[] = "0123456789";
383	char *p;
384	int i;
385
386	maxdigit = 0;
387	for (p = digit ; *p ; p++)
388		if (*p > maxdigit)
389			maxdigit = *p;
390	fputs("extern const char digit_value[];\n", hfile);
391	fputs("\n\nconst char digit_value[] = {\n", cfile);
392	for (i = 0 ; i <= maxdigit ; i++) {
393		for (p = digit ; *p && *p != i ; p++);
394		if (*p == '\0')
395			p = digit;
396		fprintf(cfile, "      %d,\n", p - digit);
397	}
398	fputs("};\n", cfile);
399}
400