1 /*        $NetBSD: syntax.h,v 1.13 2024/06/15 05:02:24 kre Exp $      */
2 
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 #include <sys/cdefs.h>
36 #include <ctype.h>
37 #include <limits.h>
38 
39 /* Syntax classes */
40 #define CWORD 0                         /* character is nothing special */
41 #define CNL 1                           /* newline character */
42 #define CBACK 2                         /* a backslash character */
43 #define CSQUOTE 3             /* single quote */
44 #define CDQUOTE 4             /* double quote */
45 #define CBQUOTE 5             /* backwards single quote */
46 #define CVAR 6                          /* a dollar sign */
47 #define CENDVAR 7             /* a '}' character */
48 #define CLP 8                           /* a left paren in arithmetic */
49 #define CRP 9                           /* a right paren in arithmetic */
50 #define CEOF 10                         /* end of file */
51 #define CSPCL 11              /* these terminate a word */
52 #define CCTL 12                         /* like CWORD, except it must be escaped */
53 #define CSBACK 13             /* a backslash in a single quote syntax */
54 #define CFAKE 14              /* a delimiter that does not exist */
55           /*
56            * note CSBACK == (CCTL|1)
57            * the code does not rely upon that, but keeping it allows a
58            * smart enough compiler to optimise some tests
59            */
60 
61 /* Syntax classes for is_ functions */
62 #define ISDIGIT 01            /* a digit */
63 #define ISUPPER 02            /* an upper case letter */
64 #define ISLOWER 04            /* a lower case letter */
65 #define ISUNDER 010           /* an underscore */
66 #define ISSPECL 020           /* the name of a special parameter */
67 #define ISSPACE 040           /* a white space character */
68 
69 #define PEOF        (CHAR_MIN - 1)
70 #define PFAKE       (CHAR_MIN - 2)
71 #define SYNBASE     (-PFAKE)
72 
73 
74 #define BASESYNTAX (basesyntax + SYNBASE)
75 #define DQSYNTAX (dqsyntax + SYNBASE)
76 #define SQSYNTAX (sqsyntax + SYNBASE)
77 #define ARISYNTAX (arisyntax + SYNBASE)
78 
79 /* These defines assume that the digits are contiguous (which is guaranteed) */
80 #define   is_digit(c)         ((unsigned)((c) - '0') <= 9)
81 #define   sh_ctype(c)         (is_type+SYNBASE)[(int)(c)]
82 #define   is_upper(c)         (sh_ctype(c) & ISUPPER)
83 #define   is_lower(c)         (sh_ctype(c) & ISLOWER)
84 #define   is_alpha(c)         (sh_ctype(c) & (ISUPPER|ISLOWER))
85 #define   is_name(c)          (sh_ctype(c) & (ISUPPER|ISLOWER|ISUNDER))
86 #define   is_in_name(c)       (sh_ctype(c) & (ISUPPER|ISLOWER|ISUNDER|ISDIGIT))
87 #define   is_special(c)       (sh_ctype(c) & (ISSPECL|ISDIGIT))
88 #define   is_space(c)         (sh_ctype(c) & ISSPACE)
89 #define   digit_val(c)        ((c) - '0')
90 
91 /* true if the arg char needs CTLESC to protect it */
92 #define   NEEDESC(c)          (SQSYNTAX[(int)(c)] == CCTL || \
93                                SQSYNTAX[(int)(c)] == CSBACK)
94 
95 #define   ISCTL(c)  ((c) >= CTL_FIRST && (c) <= CTL_LAST)
96 #if 0                                   /* alternative form (generally slower) */
97 #define   ISCTL(c)  (BASESYNTAX[(int)(c)] == CCTL)
98 #endif
99 
100 extern const char basesyntax[];
101 extern const char dqsyntax[];
102 extern const char sqsyntax[];
103 extern const char arisyntax[];
104 extern const char is_type[];
105