1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ******************************************************************************* |
5 | * |
6 | * Copyright (C) 1999-2011, International Business Machines |
7 | * Corporation and others. All Rights Reserved. |
8 | * |
9 | ******************************************************************************* |
10 | * file name: utf.h |
11 | * encoding: UTF-8 |
12 | * tab size: 8 (not used) |
13 | * indentation:4 |
14 | * |
15 | * created on: 1999sep09 |
16 | * created by: Markus W. Scherer |
17 | */ |
18 | |
19 | /** |
20 | * \file |
21 | * \brief C API: Code point macros |
22 | * |
23 | * This file defines macros for checking whether a code point is |
24 | * a surrogate or a non-character etc. |
25 | * |
26 | * If U_NO_DEFAULT_INCLUDE_UTF_HEADERS is 0 then utf.h is included by utypes.h |
27 | * and itself includes utf8.h and utf16.h after some |
28 | * common definitions. |
29 | * If U_NO_DEFAULT_INCLUDE_UTF_HEADERS is 1 then each of these headers must be |
30 | * included explicitly if their definitions are used. |
31 | * |
32 | * utf8.h and utf16.h define macros for efficiently getting code points |
33 | * in and out of UTF-8/16 strings. |
34 | * utf16.h macros have "U16_" prefixes. |
35 | * utf8.h defines similar macros with "U8_" prefixes for UTF-8 string handling. |
36 | * |
37 | * ICU mostly processes 16-bit Unicode strings. |
38 | * Most of the time, such strings are well-formed UTF-16. |
39 | * Single, unpaired surrogates must be handled as well, and are treated in ICU |
40 | * like regular code points where possible. |
41 | * (Pairs of surrogate code points are indistinguishable from supplementary |
42 | * code points encoded as pairs of supplementary code units.) |
43 | * |
44 | * In fact, almost all Unicode code points in normal text (>99%) |
45 | * are on the BMP (<=U+ffff) and even <=U+d7ff. |
46 | * ICU functions handle supplementary code points (U+10000..U+10ffff) |
47 | * but are optimized for the much more frequently occurring BMP code points. |
48 | * |
49 | * umachine.h defines UChar to be an unsigned 16-bit integer. |
50 | * Since ICU 59, ICU uses char16_t in C++, UChar only in C, |
51 | * and defines UChar=char16_t by default. See the UChar API docs for details. |
52 | * |
53 | * UChar32 is defined to be a signed 32-bit integer (int32_t), large enough for a 21-bit |
54 | * Unicode code point (Unicode scalar value, 0..0x10ffff) and U_SENTINEL (-1). |
55 | * Before ICU 2.4, the definition of UChar32 was similarly platform-dependent as |
56 | * the definition of UChar. For details see the documentation for UChar32 itself. |
57 | * |
58 | * utf.h defines a small number of C macros for single Unicode code points. |
59 | * These are simple checks for surrogates and non-characters. |
60 | * For actual Unicode character properties see uchar.h. |
61 | * |
62 | * By default, string operations must be done with error checking in case |
63 | * a string is not well-formed UTF-16 or UTF-8. |
64 | * |
65 | * The U16_ macros detect if a surrogate code unit is unpaired |
66 | * (lead unit without trail unit or vice versa) and just return the unit itself |
67 | * as the code point. |
68 | * |
69 | * The U8_ macros detect illegal byte sequences and return a negative value. |
70 | * Starting with ICU 60, the observable length of a single illegal byte sequence |
71 | * skipped by one of these macros follows the Unicode 6+ recommendation |
72 | * which is consistent with the W3C Encoding Standard. |
73 | * |
74 | * There are ..._OR_FFFD versions of both U16_ and U8_ macros |
75 | * that return U+FFFD for illegal code unit sequences. |
76 | * |
77 | * The regular "safe" macros require that the initial, passed-in string index |
78 | * is within bounds. They only check the index when they read more than one |
79 | * code unit. This is usually done with code similar to the following loop: |
80 | * <pre>while(i<length) { |
81 | * U16_NEXT(s, i, length, c); |
82 | * // use c |
83 | * }</pre> |
84 | * |
85 | * When it is safe to assume that text is well-formed UTF-16 |
86 | * (does not contain single, unpaired surrogates), then one can use |
87 | * U16_..._UNSAFE macros. |
88 | * These do not check for proper code unit sequences or truncated text and may |
89 | * yield wrong results or even cause a crash if they are used with "malformed" |
90 | * text. |
91 | * In practice, U16_..._UNSAFE macros will produce slightly less code but |
92 | * should not be faster because the processing is only different when a |
93 | * surrogate code unit is detected, which will be rare. |
94 | * |
95 | * Similarly for UTF-8, there are "safe" macros without a suffix, |
96 | * and U8_..._UNSAFE versions. |
97 | * The performance differences are much larger here because UTF-8 provides so |
98 | * many opportunities for malformed sequences. |
99 | * The unsafe UTF-8 macros are entirely implemented inside the macro definitions |
100 | * and are fast, while the safe UTF-8 macros call functions for some complicated cases. |
101 | * |
102 | * Unlike with UTF-16, malformed sequences cannot be expressed with distinct |
103 | * code point values (0..U+10ffff). They are indicated with negative values instead. |
104 | * |
105 | * For more information see the ICU User Guide Strings chapter |
106 | * (https://unicode-org.github.io/icu/userguide/strings). |
107 | * |
108 | * <em>Usage:</em> |
109 | * ICU coding guidelines for if() statements should be followed when using these macros. |
110 | * Compound statements (curly braces {}) must be used for if-else-while... |
111 | * bodies and all macro statements should be terminated with semicolon. |
112 | * |
113 | * @stable ICU 2.4 |
114 | */ |
115 | |
116 | #ifndef __UTF_H__ |
117 | #define __UTF_H__ |
118 | |
119 | #include "unicode/umachine.h" |
120 | /* include the utfXX.h after the following definitions */ |
121 | |
122 | /* single-code point definitions -------------------------------------------- */ |
123 | |
124 | /** |
125 | * Is this code point a Unicode noncharacter? |
126 | * @param c 32-bit code point |
127 | * @return true or false |
128 | * @stable ICU 2.4 |
129 | */ |
130 | #define U_IS_UNICODE_NONCHAR(c) \ |
131 | ((c)>=0xfdd0 && \ |
132 | ((c)<=0xfdef || ((c)&0xfffe)==0xfffe) && (c)<=0x10ffff) |
133 | |
134 | /** |
135 | * Is c a Unicode code point value (0..U+10ffff) |
136 | * that can be assigned a character? |
137 | * |
138 | * Code points that are not characters include: |
139 | * - single surrogate code points (U+d800..U+dfff, 2048 code points) |
140 | * - the last two code points on each plane (U+__fffe and U+__ffff, 34 code points) |
141 | * - U+fdd0..U+fdef (new with Unicode 3.1, 32 code points) |
142 | * - the highest Unicode code point value is U+10ffff |
143 | * |
144 | * This means that all code points below U+d800 are character code points, |
145 | * and that boundary is tested first for performance. |
146 | * |
147 | * @param c 32-bit code point |
148 | * @return true or false |
149 | * @stable ICU 2.4 |
150 | */ |
151 | #define U_IS_UNICODE_CHAR(c) \ |
152 | ((uint32_t)(c)<0xd800 || \ |
153 | (0xdfff<(c) && (c)<=0x10ffff && !U_IS_UNICODE_NONCHAR(c))) |
154 | |
155 | /** |
156 | * Is this code point a BMP code point (U+0000..U+ffff)? |
157 | * @param c 32-bit code point |
158 | * @return true or false |
159 | * @stable ICU 2.8 |
160 | */ |
161 | #define U_IS_BMP(c) ((uint32_t)(c)<=0xffff) |
162 | |
163 | /** |
164 | * Is this code point a supplementary code point (U+10000..U+10ffff)? |
165 | * @param c 32-bit code point |
166 | * @return true or false |
167 | * @stable ICU 2.8 |
168 | */ |
169 | #define U_IS_SUPPLEMENTARY(c) ((uint32_t)((c)-0x10000)<=0xfffff) |
170 | |
171 | /** |
172 | * Is this code point a lead surrogate (U+d800..U+dbff)? |
173 | * @param c 32-bit code point |
174 | * @return true or false |
175 | * @stable ICU 2.4 |
176 | */ |
177 | #define U_IS_LEAD(c) (((c)&0xfffffc00)==0xd800) |
178 | |
179 | /** |
180 | * Is this code point a trail surrogate (U+dc00..U+dfff)? |
181 | * @param c 32-bit code point |
182 | * @return true or false |
183 | * @stable ICU 2.4 |
184 | */ |
185 | #define U_IS_TRAIL(c) (((c)&0xfffffc00)==0xdc00) |
186 | |
187 | /** |
188 | * Is this code point a surrogate (U+d800..U+dfff)? |
189 | * @param c 32-bit code point |
190 | * @return true or false |
191 | * @stable ICU 2.4 |
192 | */ |
193 | #define U_IS_SURROGATE(c) (((c)&0xfffff800)==0xd800) |
194 | |
195 | /** |
196 | * Assuming c is a surrogate code point (U_IS_SURROGATE(c)), |
197 | * is it a lead surrogate? |
198 | * @param c 32-bit code point |
199 | * @return true or false |
200 | * @stable ICU 2.4 |
201 | */ |
202 | #define U_IS_SURROGATE_LEAD(c) (((c)&0x400)==0) |
203 | |
204 | /** |
205 | * Assuming c is a surrogate code point (U_IS_SURROGATE(c)), |
206 | * is it a trail surrogate? |
207 | * @param c 32-bit code point |
208 | * @return true or false |
209 | * @stable ICU 4.2 |
210 | */ |
211 | #define U_IS_SURROGATE_TRAIL(c) (((c)&0x400)!=0) |
212 | |
213 | /* include the utfXX.h ------------------------------------------------------ */ |
214 | |
215 | #if !U_NO_DEFAULT_INCLUDE_UTF_HEADERS |
216 | |
217 | #include "unicode/utf8.h" |
218 | #include "unicode/utf16.h" |
219 | |
220 | /* utf_old.h contains deprecated, pre-ICU 2.4 definitions */ |
221 | #include "unicode/utf_old.h" |
222 | |
223 | #endif /* !U_NO_DEFAULT_INCLUDE_UTF_HEADERS */ |
224 | |
225 | #endif /* __UTF_H__ */ |
226 | |