sl@0: /*
sl@0: *******************************************************************************
sl@0: *
sl@0: *   Copyright (C) 1999-2005, International Business Machines
sl@0: *   Corporation and others.  All Rights Reserved.
sl@0: *
sl@0: *******************************************************************************
sl@0: *   file name:  utf.h
sl@0: *   encoding:   US-ASCII
sl@0: *   tab size:   8 (not used)
sl@0: *   indentation:4
sl@0: *
sl@0: *   created on: 1999sep09
sl@0: *   created by: Markus W. Scherer
sl@0: */
sl@0: 
sl@0: /**
sl@0:  * \file
sl@0:  * \brief C API: Code point macros
sl@0:  *
sl@0:  * This file defines macros for checking whether a code point is
sl@0:  * a surrogate or a non-character etc.
sl@0:  *
sl@0:  * The UChar and UChar32 data types for Unicode code units and code points
sl@0:  * are defined in umachines.h because they can be machine-dependent.
sl@0:  *
sl@0:  * utf.h is included by utypes.h and itself includes utf8.h and utf16.h after some
sl@0:  * common definitions. Those files define macros for efficiently getting code points
sl@0:  * in and out of UTF-8/16 strings.
sl@0:  * utf16.h macros have "U16_" prefixes.
sl@0:  * utf8.h defines similar macros with "U8_" prefixes for UTF-8 string handling.
sl@0:  *
sl@0:  * ICU processes 16-bit Unicode strings.
sl@0:  * Most of the time, such strings are well-formed UTF-16.
sl@0:  * Single, unpaired surrogates must be handled as well, and are treated in ICU
sl@0:  * like regular code points where possible.
sl@0:  * (Pairs of surrogate code points are indistinguishable from supplementary
sl@0:  * code points encoded as pairs of supplementary code units.)
sl@0:  *
sl@0:  * In fact, almost all Unicode code points in normal text (>99%)
sl@0:  * are on the BMP (<=U+ffff) and even <=U+d7ff.
sl@0:  * ICU functions handle supplementary code points (U+10000..U+10ffff)
sl@0:  * but are optimized for the much more frequently occurring BMP code points.
sl@0:  *
sl@0:  * utf.h defines UChar to be an unsigned 16-bit integer. If this matches wchar_t, then
sl@0:  * UChar is defined to be exactly wchar_t, otherwise uint16_t.
sl@0:  *
sl@0:  * UChar32 is defined to be a signed 32-bit integer (int32_t), large enough for a 21-bit
sl@0:  * Unicode code point (Unicode scalar value, 0..0x10ffff).
sl@0:  * Before ICU 2.4, the definition of UChar32 was similarly platform-dependent as
sl@0:  * the definition of UChar. For details see the documentation for UChar32 itself.
sl@0:  *
sl@0:  * utf.h also defines a small number of C macros for single Unicode code points.
sl@0:  * These are simple checks for surrogates and non-characters.
sl@0:  * For actual Unicode character properties see uchar.h.
sl@0:  *
sl@0:  * By default, string operations must be done with error checking in case
sl@0:  * a string is not well-formed UTF-16.
sl@0:  * The macros will detect if a surrogate code unit is unpaired
sl@0:  * (lead unit without trail unit or vice versa) and just return the unit itself
sl@0:  * as the code point.
sl@0:  * (It is an accidental property of Unicode and UTF-16 that all
sl@0:  * malformed sequences can be expressed unambiguously with a distinct subrange
sl@0:  * of Unicode code points.)
sl@0:  *
sl@0:  * When it is safe to assume that text is well-formed UTF-16
sl@0:  * (does not contain single, unpaired surrogates), then one can use
sl@0:  * U16_..._UNSAFE macros.
sl@0:  * These do not check for proper code unit sequences or truncated text and may
sl@0:  * yield wrong results or even cause a crash if they are used with "malformed"
sl@0:  * text.
sl@0:  * In practice, U16_..._UNSAFE macros will produce slightly less code but
sl@0:  * should not be faster because the processing is only different when a
sl@0:  * surrogate code unit is detected, which will be rare.
sl@0:  *
sl@0:  * Similarly for UTF-8, there are "safe" macros without a suffix,
sl@0:  * and U8_..._UNSAFE versions.
sl@0:  * The performance differences are much larger here because UTF-8 provides so
sl@0:  * many opportunities for malformed sequences.
sl@0:  * The unsafe UTF-8 macros are entirely implemented inside the macro definitions
sl@0:  * and are fast, while the safe UTF-8 macros call functions for all but the
sl@0:  * trivial (ASCII) cases.
sl@0:  *
sl@0:  * Unlike with UTF-16, malformed sequences cannot be expressed with distinct
sl@0:  * code point values (0..U+10ffff). They are indicated with negative values instead.
sl@0:  *
sl@0:  * For more information see the ICU User Guide Strings chapter
sl@0:  * (http://icu.sourceforge.net/userguide/strings.html).
sl@0:  *
sl@0:  * <em>Usage:</em>
sl@0:  * ICU coding guidelines for if() statements should be followed when using these macros.
sl@0:  * Compound statements (curly braces {}) must be used  for if-else-while... 
sl@0:  * bodies and all macro statements should be terminated with semicolon.
sl@0:  *
sl@0:  * @stable ICU 2.4
sl@0:  */
sl@0: 
sl@0: #ifndef __UTF_H__
sl@0: #define __UTF_H__
sl@0: 
sl@0: #include "unicode/utypes.h"
sl@0: /* include the utfXX.h after the following definitions */
sl@0: 
sl@0: /* single-code point definitions -------------------------------------------- */
sl@0: 
sl@0: /**
sl@0:  * This value is intended for sentinel values for APIs that
sl@0:  * (take or) return single code points (UChar32).
sl@0:  * It is outside of the Unicode code point range 0..0x10ffff.
sl@0:  * 
sl@0:  * For example, a "done" or "error" value in a new API
sl@0:  * could be indicated with U_SENTINEL.
sl@0:  *
sl@0:  * ICU APIs designed before ICU 2.4 usually define service-specific "done"
sl@0:  * values, mostly 0xffff.
sl@0:  * Those may need to be distinguished from
sl@0:  * actual U+ffff text contents by calling functions like
sl@0:  * CharacterIterator::hasNext() or UnicodeString::length().
sl@0:  *
sl@0:  * @return -1
sl@0:  * @see UChar32
sl@0:  * @stable ICU 2.4
sl@0:  */
sl@0: #define U_SENTINEL (-1)
sl@0: 
sl@0: /**
sl@0:  * Is this code point a Unicode noncharacter?
sl@0:  * @param c 32-bit code point
sl@0:  * @return TRUE or FALSE
sl@0:  * @stable ICU 2.4
sl@0:  */
sl@0: #define U_IS_UNICODE_NONCHAR(c) \
sl@0:     ((c)>=0xfdd0 && \
sl@0:      ((uint32_t)(c)<=0xfdef || ((c)&0xfffe)==0xfffe) && \
sl@0:      (uint32_t)(c)<=0x10ffff)
sl@0: 
sl@0: /**
sl@0:  * Is c a Unicode code point value (0..U+10ffff)
sl@0:  * that can be assigned a character?
sl@0:  *
sl@0:  * Code points that are not characters include:
sl@0:  * - single surrogate code points (U+d800..U+dfff, 2048 code points)
sl@0:  * - the last two code points on each plane (U+__fffe and U+__ffff, 34 code points)
sl@0:  * - U+fdd0..U+fdef (new with Unicode 3.1, 32 code points)
sl@0:  * - the highest Unicode code point value is U+10ffff
sl@0:  *
sl@0:  * This means that all code points below U+d800 are character code points,
sl@0:  * and that boundary is tested first for performance.
sl@0:  *
sl@0:  * @param c 32-bit code point
sl@0:  * @return TRUE or FALSE
sl@0:  * @stable ICU 2.4
sl@0:  */
sl@0: #define U_IS_UNICODE_CHAR(c) \
sl@0:     ((uint32_t)(c)<0xd800 || \
sl@0:         ((uint32_t)(c)>0xdfff && \
sl@0:          (uint32_t)(c)<=0x10ffff && \
sl@0:          !U_IS_UNICODE_NONCHAR(c)))
sl@0: 
sl@0: #ifndef U_HIDE_DRAFT_API
sl@0: 
sl@0: /**
sl@0:  * Is this code point a BMP code point (U+0000..U+ffff)?
sl@0:  * @param c 32-bit code point
sl@0:  * @return TRUE or FALSE
sl@0:  * @stable ICU 2.8
sl@0:  */
sl@0: #define U_IS_BMP(c) ((uint32_t)(c)<=0xffff)
sl@0: 
sl@0: /**
sl@0:  * Is this code point a supplementary code point (U+10000..U+10ffff)?
sl@0:  * @param c 32-bit code point
sl@0:  * @return TRUE or FALSE
sl@0:  * @stable ICU 2.8
sl@0:  */
sl@0: #define U_IS_SUPPLEMENTARY(c) ((uint32_t)((c)-0x10000)<=0xfffff)
sl@0: 
sl@0: #endif /*U_HIDE_DRAFT_API*/
sl@0:  
sl@0: /**
sl@0:  * Is this code point a lead surrogate (U+d800..U+dbff)?
sl@0:  * @param c 32-bit code point
sl@0:  * @return TRUE or FALSE
sl@0:  * @stable ICU 2.4
sl@0:  */
sl@0: #define U_IS_LEAD(c) (((c)&0xfffffc00)==0xd800)
sl@0: 
sl@0: /**
sl@0:  * Is this code point a trail surrogate (U+dc00..U+dfff)?
sl@0:  * @param c 32-bit code point
sl@0:  * @return TRUE or FALSE
sl@0:  * @stable ICU 2.4
sl@0:  */
sl@0: #define U_IS_TRAIL(c) (((c)&0xfffffc00)==0xdc00)
sl@0: 
sl@0: /**
sl@0:  * Is this code point a surrogate (U+d800..U+dfff)?
sl@0:  * @param c 32-bit code point
sl@0:  * @return TRUE or FALSE
sl@0:  * @stable ICU 2.4
sl@0:  */
sl@0: #define U_IS_SURROGATE(c) (((c)&0xfffff800)==0xd800)
sl@0: 
sl@0: /**
sl@0:  * Assuming c is a surrogate code point (U_IS_SURROGATE(c)),
sl@0:  * is it a lead surrogate?
sl@0:  * @param c 32-bit code point
sl@0:  * @return TRUE or FALSE
sl@0:  * @stable ICU 2.4
sl@0:  */
sl@0: #define U_IS_SURROGATE_LEAD(c) (((c)&0x400)==0)
sl@0: 
sl@0: /* include the utfXX.h ------------------------------------------------------ */
sl@0: 
sl@0: #include "unicode/utf8.h"
sl@0: #include "unicode/utf16.h"
sl@0: 
sl@0: /* utf_old.h contains deprecated, pre-ICU 2.4 definitions */
sl@0: #include "unicode/utf_old.h"
sl@0: 
sl@0: #endif