icuSources/common/unicode/utf.h

   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  * The UChar and UChar32 data types for Unicode code units and code points
  27  * are defined in umachine.h because they can be machine-dependent.
  28  *
  29  * If U_NO_DEFAULT_INCLUDE_UTF_HEADERS is 0 then utf.h is included by utypes.h
  30  * and itself includes utf8.h and utf16.h after some
  31  * common definitions.
  32  * If U_NO_DEFAULT_INCLUDE_UTF_HEADERS is 1 then each of these headers must be
  33  * included explicitly if their definitions are used.
  34  *
  35  * utf8.h and utf16.h define macros for efficiently getting code points
  36  * in and out of UTF-8/16 strings.
  37  * utf16.h macros have "U16_" prefixes.
  38  * utf8.h defines similar macros with "U8_" prefixes for UTF-8 string handling.
  39  *
  40  * ICU mostly processes 16-bit Unicode strings.
  41  * Most of the time, such strings are well-formed UTF-16.
  42  * Single, unpaired surrogates must be handled as well, and are treated in ICU
  43  * like regular code points where possible.
  44  * (Pairs of surrogate code points are indistinguishable from supplementary
  45  * code points encoded as pairs of supplementary code units.)
  46  *
  47  * In fact, almost all Unicode code points in normal text (>99%)
  48  * are on the BMP (<=U+ffff) and even <=U+d7ff.
  49  * ICU functions handle supplementary code points (U+10000..U+10ffff)
  50  * but are optimized for the much more frequently occurring BMP code points.
  51  *
  52  * umachine.h defines UChar to be an unsigned 16-bit integer.
  53  * Where available, UChar is defined to be a char16_t
  54  * or a wchar_t (if that is an unsigned 16-bit type), otherwise uint16_t.
  55  *
  56  * UChar32 is defined to be a signed 32-bit integer (int32_t), large enough for a 21-bit
  57  * Unicode code point (Unicode scalar value, 0..0x10ffff).
  58  * Before ICU 2.4, the definition of UChar32 was similarly platform-dependent as
  59  * the definition of UChar. For details see the documentation for UChar32 itself.
  60  *
  61  * utf.h defines a small number of C macros for single Unicode code points.
  62  * These are simple checks for surrogates and non-characters.
  63  * For actual Unicode character properties see uchar.h.
  64  *
  65  * By default, string operations must be done with error checking in case
  66  * a string is not well-formed UTF-16.
  67  * The macros will detect if a surrogate code unit is unpaired
  68  * (lead unit without trail unit or vice versa) and just return the unit itself
  69  * as the code point.
  70  *
  71  * The regular "safe" macros require that the initial, passed-in string index
  72  * is within bounds. They only check the index when they read more than one
  73  * code unit. This is usually done with code similar to the following loop:
  74  * <pre>while(i<length) {
  75  *   U16_NEXT(s, i, length, c);
  76  *   // use c
  77  * }</pre>
  78  *
  79  * When it is safe to assume that text is well-formed UTF-16
  80  * (does not contain single, unpaired surrogates), then one can use
  81  * U16_..._UNSAFE macros.
  82  * These do not check for proper code unit sequences or truncated text and may
  83  * yield wrong results or even cause a crash if they are used with "malformed"
  84  * text.
  85  * In practice, U16_..._UNSAFE macros will produce slightly less code but
  86  * should not be faster because the processing is only different when a
  87  * surrogate code unit is detected, which will be rare.
  88  *
  89  * Similarly for UTF-8, there are "safe" macros without a suffix,
  90  * and U8_..._UNSAFE versions.
  91  * The performance differences are much larger here because UTF-8 provides so
  92  * many opportunities for malformed sequences.
  93  * The unsafe UTF-8 macros are entirely implemented inside the macro definitions
  94  * and are fast, while the safe UTF-8 macros call functions for all but the
  95  * trivial (ASCII) cases.
  96  * (ICU 3.6 optimizes U8_NEXT() and U8_APPEND() to handle most other common
  97  * characters inline as well.)
  98  *
  99  * Unlike with UTF-16, malformed sequences cannot be expressed with distinct
 100  * code point values (0..U+10ffff). They are indicated with negative values instead.
 101  *
 102  * For more information see the ICU User Guide Strings chapter
 103  * (http://userguide.icu-project.org/strings).
 104  *
 105  * <em>Usage:</em>
 106  * ICU coding guidelines for if() statements should be followed when using these macros.
 107  * Compound statements (curly braces {}) must be used  for if-else-while...
 108  * bodies and all macro statements should be terminated with semicolon.
 109  *
 110  * @stable ICU 2.4
 111  */
 112
 113 #ifndef __UTF_H__
 114 #define __UTF_H__
 115
 116 #include "unicode/umachine.h"
 117 /* include the utfXX.h after the following definitions */
 118
 119 /* single-code point definitions -------------------------------------------- */
 120
 121 /**
 122  * Is this code point a Unicode noncharacter?
 123  * @param c 32-bit code point
 124  * @return TRUE or FALSE
 125  * @stable ICU 2.4
 126  */
 127 #define U_IS_UNICODE_NONCHAR(c) \
 128     ((c)>=0xfdd0 && \
 129      ((uint32_t)(c)<=0xfdef || ((c)&0xfffe)==0xfffe) && \
 130      (uint32_t)(c)<=0x10ffff)
 131
 132 /**
 133  * Is c a Unicode code point value (0..U+10ffff)
 134  * that can be assigned a character?
 135  *
 136  * Code points that are not characters include:
 137  * - single surrogate code points (U+d800..U+dfff, 2048 code points)
 138  * - the last two code points on each plane (U+__fffe and U+__ffff, 34 code points)
 139  * - U+fdd0..U+fdef (new with Unicode 3.1, 32 code points)
 140  * - the highest Unicode code point value is U+10ffff
 141  *
 142  * This means that all code points below U+d800 are character code points,
 143  * and that boundary is tested first for performance.
 144  *
 145  * @param c 32-bit code point
 146  * @return TRUE or FALSE
 147  * @stable ICU 2.4
 148  */
 149 #define U_IS_UNICODE_CHAR(c) \
 150     ((uint32_t)(c)<0xd800 || \
 151         ((uint32_t)(c)>0xdfff && \
 152          (uint32_t)(c)<=0x10ffff && \
 153          !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__ */