xref: /titanic_44/usr/src/lib/libc/port/locale/runetype.h (revision 2d08521bd15501c8370ba2153b9cca4f094979d0)
1 /*
2  * Copyright 2013 Garrett D'Amore <garrett@damore.org>
3  * Copyright 2010 Nexenta Systems, Inc.  All rights reserved.
4  * Copyright (c) 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  * Paul Borman at Krystal Technologies.
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  * Because we borrowed the __rune_t defintions from _types.h, we need
36  * this copyright notice as well:
37  *
38  * Copyright (c) 2002 Mike Barcroft <mike@FreeBSD.org>
39  * All rights reserved.
40  *
41  * Redistribution and use in source and binary forms, with or without
42  * modification, are permitted provided that the following conditions
43  * are met:
44  * 1. Redistributions of source code must retain the above copyright
45  *    notice, this list of conditions and the following disclaimer.
46  * 2. Redistributions in binary form must reproduce the above copyright
47  *    notice, this list of conditions and the following disclaimer in the
48  *    documentation and/or other materials provided with the distribution.
49  *
50  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
51  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
52  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
53  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
54  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
55  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
56  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
57  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
58  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
59  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
60  * SUCH DAMAGE.
61  */
62 
63 #ifndef	_RUNETYPE_H_
64 #define	_RUNETYPE_H_
65 
66 #define	_CACHED_RUNES	(1 << 8)		/* Must be a power of 2 */
67 #define	_CRMASK		(~(_CACHED_RUNES - 1))
68 
69 /*
70  * rune_t is declared to be an ``int'' instead of the more natural
71  * ``unsigned long'' or ``long''.  Two things are happening here.  It is not
72  * unsigned so that EOF (-1) can be naturally assigned to it and used.  Also,
73  * it looks like 10646 will be a 31 bit standard.  This means that if your
74  * ints cannot hold 32 bits, you will be in trouble.  The reason an int was
75  * chosen over a long is that the is*() and to*() routines take ints (says
76  * ANSI C), but they use __ct_rune_t instead of int.
77  *
78  * NOTE: rune_t is not covered by ANSI nor other standards, and should not
79  * be instantiated outside of lib/libc/locale.  Use wchar_t.  wchar_t and
80  * rune_t must be the same type.  Also, wint_t must be no narrower than
81  * wchar_t, and should be able to hold all members of the largest
82  * character set plus one extra value (WEOF), and must be at least 16 bits.
83  *
84  * For compatibility with Solaris, we want to use long in ILP32, and int in
85  * LP64.  This is due to historical Solaris legacy.  (See <wchar.h> for
86  * the definition.)
87  */
88 
89 #if	defined(_LP64)
90 typedef int		__ct_rune_t;	/* arg type for ctype funcs */
91 #else
92 typedef	int		__ct_rune_t;
93 #endif
94 
95 typedef	int		__rune_t;	/* rune_t (see above) */
96 
97 /*
98  * The lower 8 bits of runetype[] contain the digit value of the rune.
99  */
100 typedef struct {
101 	__rune_t	__min;		/* First rune of the range */
102 	__rune_t	__max;		/* Last rune (inclusive) of the range */
103 	__rune_t	__map;		/* What first maps to in maps */
104 	unsigned	*__types;	/* Array of types in range */
105 } _RuneEntry;
106 
107 typedef struct {
108 	int		__nranges;	/* Number of ranges stored */
109 	_RuneEntry	*__ranges;	/* Pointer to the ranges */
110 } _RuneRange;
111 
112 typedef struct {
113 	char		__magic[8];	/* Magic saying what version we are */
114 	char		__encoding[32];	/* ASCII name of encoding */
115 
116 	unsigned int	__runetype[_CACHED_RUNES];
117 	__rune_t	__maplower[_CACHED_RUNES];
118 	__rune_t	__mapupper[_CACHED_RUNES];
119 
120 	/*
121 	 * The following are to deal with Runes larger than _CACHED_RUNES - 1.
122 	 * Their data is actually contiguous with this structure so as to make
123 	 * it easier to read/write from/to disk.
124 	 */
125 	_RuneRange	__runetype_ext;
126 	_RuneRange	__maplower_ext;
127 	_RuneRange	__mapupper_ext;
128 
129 	void		*__variable;	/* Data which depends on the encoding */
130 	int		__variable_len;	/* how long that data is */
131 } _RuneLocale;
132 
133 #define	_RUNE_MAGIC_1	"RuneMagi"	/* Indicates version 0 of RuneLocale */
134 
135 extern _RuneLocale _DefaultRuneLocale;
136 
137 unsigned int __runetype(const _RuneLocale *, int);
138 
139 #endif	/* !_RUNETYPE_H_ */
140