1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/os/textandloc/fontservices/textshaperplugin/IcuSource/common/unicode/utf.h Fri Jun 15 03:10:57 2012 +0200
1.3 @@ -0,0 +1,221 @@
1.4 +/*
1.5 +*******************************************************************************
1.6 +*
1.7 +* Copyright (C) 1999-2005, International Business Machines
1.8 +* Corporation and others. All Rights Reserved.
1.9 +*
1.10 +*******************************************************************************
1.11 +* file name: utf.h
1.12 +* encoding: US-ASCII
1.13 +* tab size: 8 (not used)
1.14 +* indentation:4
1.15 +*
1.16 +* created on: 1999sep09
1.17 +* created by: Markus W. Scherer
1.18 +*/
1.19 +
1.20 +/**
1.21 + * \file
1.22 + * \brief C API: Code point macros
1.23 + *
1.24 + * This file defines macros for checking whether a code point is
1.25 + * a surrogate or a non-character etc.
1.26 + *
1.27 + * The UChar and UChar32 data types for Unicode code units and code points
1.28 + * are defined in umachines.h because they can be machine-dependent.
1.29 + *
1.30 + * utf.h is included by utypes.h and itself includes utf8.h and utf16.h after some
1.31 + * common definitions. Those files define macros for efficiently getting code points
1.32 + * in and out of UTF-8/16 strings.
1.33 + * utf16.h macros have "U16_" prefixes.
1.34 + * utf8.h defines similar macros with "U8_" prefixes for UTF-8 string handling.
1.35 + *
1.36 + * ICU processes 16-bit Unicode strings.
1.37 + * Most of the time, such strings are well-formed UTF-16.
1.38 + * Single, unpaired surrogates must be handled as well, and are treated in ICU
1.39 + * like regular code points where possible.
1.40 + * (Pairs of surrogate code points are indistinguishable from supplementary
1.41 + * code points encoded as pairs of supplementary code units.)
1.42 + *
1.43 + * In fact, almost all Unicode code points in normal text (>99%)
1.44 + * are on the BMP (<=U+ffff) and even <=U+d7ff.
1.45 + * ICU functions handle supplementary code points (U+10000..U+10ffff)
1.46 + * but are optimized for the much more frequently occurring BMP code points.
1.47 + *
1.48 + * utf.h defines UChar to be an unsigned 16-bit integer. If this matches wchar_t, then
1.49 + * UChar is defined to be exactly wchar_t, otherwise uint16_t.
1.50 + *
1.51 + * UChar32 is defined to be a signed 32-bit integer (int32_t), large enough for a 21-bit
1.52 + * Unicode code point (Unicode scalar value, 0..0x10ffff).
1.53 + * Before ICU 2.4, the definition of UChar32 was similarly platform-dependent as
1.54 + * the definition of UChar. For details see the documentation for UChar32 itself.
1.55 + *
1.56 + * utf.h also defines a small number of C macros for single Unicode code points.
1.57 + * These are simple checks for surrogates and non-characters.
1.58 + * For actual Unicode character properties see uchar.h.
1.59 + *
1.60 + * By default, string operations must be done with error checking in case
1.61 + * a string is not well-formed UTF-16.
1.62 + * The macros will detect if a surrogate code unit is unpaired
1.63 + * (lead unit without trail unit or vice versa) and just return the unit itself
1.64 + * as the code point.
1.65 + * (It is an accidental property of Unicode and UTF-16 that all
1.66 + * malformed sequences can be expressed unambiguously with a distinct subrange
1.67 + * of Unicode code points.)
1.68 + *
1.69 + * When it is safe to assume that text is well-formed UTF-16
1.70 + * (does not contain single, unpaired surrogates), then one can use
1.71 + * U16_..._UNSAFE macros.
1.72 + * These do not check for proper code unit sequences or truncated text and may
1.73 + * yield wrong results or even cause a crash if they are used with "malformed"
1.74 + * text.
1.75 + * In practice, U16_..._UNSAFE macros will produce slightly less code but
1.76 + * should not be faster because the processing is only different when a
1.77 + * surrogate code unit is detected, which will be rare.
1.78 + *
1.79 + * Similarly for UTF-8, there are "safe" macros without a suffix,
1.80 + * and U8_..._UNSAFE versions.
1.81 + * The performance differences are much larger here because UTF-8 provides so
1.82 + * many opportunities for malformed sequences.
1.83 + * The unsafe UTF-8 macros are entirely implemented inside the macro definitions
1.84 + * and are fast, while the safe UTF-8 macros call functions for all but the
1.85 + * trivial (ASCII) cases.
1.86 + *
1.87 + * Unlike with UTF-16, malformed sequences cannot be expressed with distinct
1.88 + * code point values (0..U+10ffff). They are indicated with negative values instead.
1.89 + *
1.90 + * For more information see the ICU User Guide Strings chapter
1.91 + * (http://icu.sourceforge.net/userguide/strings.html).
1.92 + *
1.93 + * <em>Usage:</em>
1.94 + * ICU coding guidelines for if() statements should be followed when using these macros.
1.95 + * Compound statements (curly braces {}) must be used for if-else-while...
1.96 + * bodies and all macro statements should be terminated with semicolon.
1.97 + *
1.98 + * @stable ICU 2.4
1.99 + */
1.100 +
1.101 +#ifndef __UTF_H__
1.102 +#define __UTF_H__
1.103 +
1.104 +#include "unicode/utypes.h"
1.105 +/* include the utfXX.h after the following definitions */
1.106 +
1.107 +/* single-code point definitions -------------------------------------------- */
1.108 +
1.109 +/**
1.110 + * This value is intended for sentinel values for APIs that
1.111 + * (take or) return single code points (UChar32).
1.112 + * It is outside of the Unicode code point range 0..0x10ffff.
1.113 + *
1.114 + * For example, a "done" or "error" value in a new API
1.115 + * could be indicated with U_SENTINEL.
1.116 + *
1.117 + * ICU APIs designed before ICU 2.4 usually define service-specific "done"
1.118 + * values, mostly 0xffff.
1.119 + * Those may need to be distinguished from
1.120 + * actual U+ffff text contents by calling functions like
1.121 + * CharacterIterator::hasNext() or UnicodeString::length().
1.122 + *
1.123 + * @return -1
1.124 + * @see UChar32
1.125 + * @stable ICU 2.4
1.126 + */
1.127 +#define U_SENTINEL (-1)
1.128 +
1.129 +/**
1.130 + * Is this code point a Unicode noncharacter?
1.131 + * @param c 32-bit code point
1.132 + * @return TRUE or FALSE
1.133 + * @stable ICU 2.4
1.134 + */
1.135 +#define U_IS_UNICODE_NONCHAR(c) \
1.136 + ((c)>=0xfdd0 && \
1.137 + ((uint32_t)(c)<=0xfdef || ((c)&0xfffe)==0xfffe) && \
1.138 + (uint32_t)(c)<=0x10ffff)
1.139 +
1.140 +/**
1.141 + * Is c a Unicode code point value (0..U+10ffff)
1.142 + * that can be assigned a character?
1.143 + *
1.144 + * Code points that are not characters include:
1.145 + * - single surrogate code points (U+d800..U+dfff, 2048 code points)
1.146 + * - the last two code points on each plane (U+__fffe and U+__ffff, 34 code points)
1.147 + * - U+fdd0..U+fdef (new with Unicode 3.1, 32 code points)
1.148 + * - the highest Unicode code point value is U+10ffff
1.149 + *
1.150 + * This means that all code points below U+d800 are character code points,
1.151 + * and that boundary is tested first for performance.
1.152 + *
1.153 + * @param c 32-bit code point
1.154 + * @return TRUE or FALSE
1.155 + * @stable ICU 2.4
1.156 + */
1.157 +#define U_IS_UNICODE_CHAR(c) \
1.158 + ((uint32_t)(c)<0xd800 || \
1.159 + ((uint32_t)(c)>0xdfff && \
1.160 + (uint32_t)(c)<=0x10ffff && \
1.161 + !U_IS_UNICODE_NONCHAR(c)))
1.162 +
1.163 +#ifndef U_HIDE_DRAFT_API
1.164 +
1.165 +/**
1.166 + * Is this code point a BMP code point (U+0000..U+ffff)?
1.167 + * @param c 32-bit code point
1.168 + * @return TRUE or FALSE
1.169 + * @stable ICU 2.8
1.170 + */
1.171 +#define U_IS_BMP(c) ((uint32_t)(c)<=0xffff)
1.172 +
1.173 +/**
1.174 + * Is this code point a supplementary code point (U+10000..U+10ffff)?
1.175 + * @param c 32-bit code point
1.176 + * @return TRUE or FALSE
1.177 + * @stable ICU 2.8
1.178 + */
1.179 +#define U_IS_SUPPLEMENTARY(c) ((uint32_t)((c)-0x10000)<=0xfffff)
1.180 +
1.181 +#endif /*U_HIDE_DRAFT_API*/
1.182 +
1.183 +/**
1.184 + * Is this code point a lead surrogate (U+d800..U+dbff)?
1.185 + * @param c 32-bit code point
1.186 + * @return TRUE or FALSE
1.187 + * @stable ICU 2.4
1.188 + */
1.189 +#define U_IS_LEAD(c) (((c)&0xfffffc00)==0xd800)
1.190 +
1.191 +/**
1.192 + * Is this code point a trail surrogate (U+dc00..U+dfff)?
1.193 + * @param c 32-bit code point
1.194 + * @return TRUE or FALSE
1.195 + * @stable ICU 2.4
1.196 + */
1.197 +#define U_IS_TRAIL(c) (((c)&0xfffffc00)==0xdc00)
1.198 +
1.199 +/**
1.200 + * Is this code point a surrogate (U+d800..U+dfff)?
1.201 + * @param c 32-bit code point
1.202 + * @return TRUE or FALSE
1.203 + * @stable ICU 2.4
1.204 + */
1.205 +#define U_IS_SURROGATE(c) (((c)&0xfffff800)==0xd800)
1.206 +
1.207 +/**
1.208 + * Assuming c is a surrogate code point (U_IS_SURROGATE(c)),
1.209 + * is it a lead surrogate?
1.210 + * @param c 32-bit code point
1.211 + * @return TRUE or FALSE
1.212 + * @stable ICU 2.4
1.213 + */
1.214 +#define U_IS_SURROGATE_LEAD(c) (((c)&0x400)==0)
1.215 +
1.216 +/* include the utfXX.h ------------------------------------------------------ */
1.217 +
1.218 +#include "unicode/utf8.h"
1.219 +#include "unicode/utf16.h"
1.220 +
1.221 +/* utf_old.h contains deprecated, pre-ICU 2.4 definitions */
1.222 +#include "unicode/utf_old.h"
1.223 +
1.224 +#endif