AuroraMiddleware/scuffed-code
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

ICU-647 update API docs

Browse Source 
X-SVN-Rev: 3200
This commit is contained in:
Markus Scherer 2000-12-12 20:31:47 +00:00
parent 07a0851ec4
commit 96c50174b4
4 changed files with 347 additions and 89 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

235
icu4c/source/common/unicode/utf.h
View File

@ -16,13 +16,50 @@
/** /**
* \file * \file
* \brief Description of UChar and UChar32 data types * \brief C API: UChar and UChar32 data types and UTF macros for C Unicode string handling
* *
* This file defines the UChar and UChar32 data types for Unicode code units * <p>This file defines the UChar and UChar32 data types for Unicode code units
* and code points, as well as macros for efficiently getting code points * and code points, as well as macros for efficiently getting code points
* in and out of a string. * in and out of a string.</p>
* utf.h is included by utypes.h and itself includes the utfXX.h after some *
* common definitions. Those files define the macros for each UTF-size. * <p>utf.h is included by utypes.h and itself includes the utfXX.h after some
* common definitions. Those files define the macros for each UTF-size.</p>
*
* <p>ICU allows in principle to set which UTF (UTF-8/16/32) is used internally
* by defining UTF_SIZE to either 8, 16, or 32. utf.h would then define the UChar type
* accordingly. UTF-16 is the default.<br>
* In praxis, since a lot of the ICU source code &mdash; especially low-level code like
* conversion and collation &mdash; assumes UTF-16, utf.h enforces the default of UTF-16.
* This is unlikely to change in the future. Only some files (ubidi.h) should work with any UTF.</p>
*
* <p>Accordinly, utf.h defines UChar to be an unsigned 16-bit integer. If this matches wchar_t, then
* UChar is defined to be exactly wchar_t, otherwise uint16_t.</p>
*
* <p>UChar32 is always defined to be a 32-bit integer to be large enough for a 21-bit
* Unicode code point (Unicode scalar value, 0..0x10ffff). If wchar_t is a 32-bit type, then
* UChar32 is defined to be exactly wchar_t, <em>regardless of whether wchar_t is signed or unsigned.
* This means that UChar32 may be signed or unsigned depending on the platform!</em>
* If wchar_t is not a 32-bit type, then UChar32 is defined to be uint32_t.</p>
*
* <p>utf.h also defines a number of C macros for handling single Unicode code points and
* for using UTF Unicode strings. It includes utf8.h, utf16.h, and utf32.h for the actual
* implementations of those macros and then aliases one set of them (for UTF-16) for general use.
* The UTF-specific macros have the UTF size in the macro name prefixes (UTF16_...), while
* the general alias macros always begin with UTF_...</p>
*
* <p>Many string operations can be done with or without error checking.
* Where such a distinction is useful, there are two versions of the macros, "unsafe" and "safe"
* ones with ..._UNSAFE and ..._SAFE suffixes. The unsafe macros are fast but may cause
* program failures if the strings are not well-formed. The safe macros have an additional, boolean
* parameter "strict". If strict is FALSE, then only illegal sequences are detected.
* Otherwise, irregular sequences are detected as well (like single surrogates in UTF-8/32).
* Safe macros return special error code points for illegal/irregular sequences:
* Typically, U+ffff, or for UTF-8 values that would result in a byte sequence of the same length
* as the illegal input sequence.</p>
*
* <p>It is possible to change the choice for the general alias macros to be unsafe, safe/not strict or safe/strict.
* The default is safe/not strict. It is not recommended to select the unsafe macros as the basis for
* Unicode string handling in ICU! To select this, define UTF_SAFE, UTF_STRICT, or UTF_UNSAFE.</p>
*/ */
#ifndef __UTF_H__ #ifndef __UTF_H__
@ -40,12 +77,17 @@
/* If there is no compiler option for the preferred UTF size, then default to UTF-16. */ /* If there is no compiler option for the preferred UTF size, then default to UTF-16. */
#ifndef UTF_SIZE #ifndef UTF_SIZE
/** Number of bits in a Unicode string code unit, same as x in UTF-x (8, 16, or 32). */
# define UTF_SIZE 16 # define UTF_SIZE 16
#endif #endif
/** Number of bytes in a UChar (sizeof(UChar)). */
#define U_SIZEOF_UCHAR (UTF_SIZE>>3) #define U_SIZEOF_UCHAR (UTF_SIZE>>3)
/* Do we have wchar.h on this platform? It is there on most platforms. */ /*!
* \def U_SIZEOF_WCHAR_T
* Do we have wchar.h on this platform? It is there on most platforms.
*/
#ifndef U_HAVE_WCHAR_H #ifndef U_HAVE_WCHAR_H
# define U_HAVE_WCHAR_H 1 # define U_HAVE_WCHAR_H 1
#endif #endif
@ -53,48 +95,67 @@
/* U_SIZEOF_WCHAR_T==sizeof(wchar_t) (0 means it is not defined or autoconf could not set it) */ /* U_SIZEOF_WCHAR_T==sizeof(wchar_t) (0 means it is not defined or autoconf could not set it) */
#if U_SIZEOF_WCHAR_T==0 #if U_SIZEOF_WCHAR_T==0
# undef U_SIZEOF_WCHAR_T # undef U_SIZEOF_WCHAR_T
/** U_SIZEOF_WCHAR_T==sizeof(wchar_t). */
# define U_SIZEOF_WCHAR_T 4 # define U_SIZEOF_WCHAR_T 4
#endif #endif
/* Define UChar32 to be compatible with wchar_t if possible. */ /*!
* \var UChar32
* Define UChar32 to be wchar_t if that is 32 bits wide; may be signed or unsigned!
* If wchar_t is not 32 bits wide, then define UChar32 to be uint32_t.
*/
#if U_SIZEOF_WCHAR_T==4 #if U_SIZEOF_WCHAR_T==4
typedef wchar_t UChar32; typedef wchar_t UChar32;
#else #else
typedef uint32_t UChar32; typedef uint32_t UChar32;
#endif #endif
/* Unicode string and array offset and index type */ /**
* Unicode string and array offset and index type.
* ICU always counts Unicode code units (UChars) for string offsets, indexes, and lengths, not Unicode code points.
*/
typedef int32_t UTextOffset; typedef int32_t UTextOffset;
/* Specify which macro versions are the default ones - safe or fast. */ /* Specify which macro versions are the default ones - safe or fast. */
#if !defined(UTF_SAFE) && !defined(UTF_STRICT) && !defined(UTF_UNSAFE) #if !defined(UTF_SAFE) && !defined(UTF_STRICT) && !defined(UTF_UNSAFE)
/**
* The default choice for general Unicode string macros is to use the ..._SAFE macro implementations
* with strict=FALSE. See the utf.h file description.
*/
# define UTF_SAFE # define UTF_SAFE
#endif #endif
/* internal definitions ----------------------------------------------------- */ /* internal definitions ----------------------------------------------------- */
/* /**
* Special error values for UTF-8, * <p>Special error values for UTF-8,
* which need 1 or 2 bytes in UTF-8: * which need 1 or 2 bytes in UTF-8:<br>
* U+0015 = NAK = Negative Acknowledge, C0 control character * U+0015 = NAK = Negative Acknowledge, C0 control character<br>
* U+009f = highest C1 control character * U+009f = highest C1 control character</p>
* *
* These are used by ("safe") UTF-8 macros so that they can return an error value * <p>These are used by ("safe") UTF-8 macros so that they can return an error value
* that needs the same number of code units (bytes) as were seen by * that needs the same number of code units (bytes) as were seen by
* a macro. * a macro. They should be tested with UTF_IS_ERROR() or UTF_IS_VALID().</p>
*
* @internal
*/ */
/*@{*/
#define UTF8_ERROR_VALUE_1 0x15 #define UTF8_ERROR_VALUE_1 0x15
#define UTF8_ERROR_VALUE_2 0x9f #define UTF8_ERROR_VALUE_2 0x9f
/*@}*/
/* error value for all UTFs */ /**
* Error value for all UTFs. This code point value will be set by macros with error
* checking if an error is detected.
*/
#define UTF_ERROR_VALUE 0xffff #define UTF_ERROR_VALUE 0xffff
/* single-code point definitions -------------------------------------------- */ /* single-code point definitions -------------------------------------------- */
/* is this code unit or code point a surrogate? */ /** Is this code unit or code point a surrogate (U+d800..U+dfff)? */
#define UTF_IS_SURROGATE(uchar) (((uchar)&0xfffff800)==0xd800) #define UTF_IS_SURROGATE(uchar) (((uchar)&0xfffff800)==0xd800)
/* /**
* Is a given 32-bit code point/Unicode scalar value * Is a given 32-bit code point/Unicode scalar value
* actually a valid Unicode (abstract) character? * actually a valid Unicode (abstract) character?
*/ */
@ -102,14 +163,14 @@ typedef int32_t UTextOffset;
((uint32_t)(c)<=0x10ffff && \ ((uint32_t)(c)<=0x10ffff && \
!UTF_IS_SURROGATE(c) && ((c)&0xfffe)!=0xfffe) !UTF_IS_SURROGATE(c) && ((c)&0xfffe)!=0xfffe)
/* /**
* Is a given 32-bit code an error value * Is a given 32-bit code an error value
* as returned by one of the macros for any UTF? * as returned by one of the macros for any UTF?
*/ */
#define UTF_IS_ERROR(c) \ #define UTF_IS_ERROR(c) \
(((c)&0xfffe)==0xfffe || (c)==UTF8_ERROR_VALUE_1 || (c)==UTF8_ERROR_VALUE_2) (((c)&0xfffe)==0xfffe || (c)==UTF8_ERROR_VALUE_1 || (c)==UTF8_ERROR_VALUE_2)
/* This is a combined macro: is c a valid Unicode value _and_ not an error code? */ /** This is a combined macro: Is c a valid Unicode value _and_ not an error code? */
#define UTF_IS_VALID(c) \ #define UTF_IS_VALID(c) \
((uint32_t)(c)<=0x10ffff && \ ((uint32_t)(c)<=0x10ffff && \
!UTF_IS_SURROGATE(c) && \ !UTF_IS_SURROGATE(c) && \
@ -124,6 +185,19 @@ typedef int32_t UTextOffset;
/* Define types and macros according to the selected UTF size. -------------- */ /* Define types and macros according to the selected UTF size. -------------- */
/*!
* \var UChar
* Define UChar to be wchar_t if that is 16 bits wide; always assumed to be unsigned.
* If wchar_t is not 16 bits wide, then define UChar to be uint16_t.
*/
/**
* <p>All these macros are aliases to the selected UTF implementation macros.
* In an ICU build, they are always macros to the UTF-16 macros (with UTF16_ prefixes).
* In essence, they remove the UTF size from the macro names so that all macros will
* have a UTF_ prefix.</p>
*/
/*@{*/
#if UTF_SIZE==8 #if UTF_SIZE==8
# error UTF-8 is not implemented, undefine UTF_SIZE or define it to 16 # error UTF-8 is not implemented, undefine UTF_SIZE or define it to 16
@ -144,24 +218,27 @@ typedef int32_t UTextOffset;
typedef uint16_t UChar; typedef uint16_t UChar;
# endif # endif
/** Does this code unit alone encode a code point? */
# define UTF_IS_SINGLE(uchar) UTF16_IS_SINGLE(uchar) # define UTF_IS_SINGLE(uchar) UTF16_IS_SINGLE(uchar)
/** Is this code unit the first one of several? */
# define UTF_IS_LEAD(uchar) UTF16_IS_LEAD(uchar) # define UTF_IS_LEAD(uchar) UTF16_IS_LEAD(uchar)
/** Is this code unit one of several but not the first one? */
# define UTF_IS_TRAIL(uchar) UTF16_IS_TRAIL(uchar) # define UTF_IS_TRAIL(uchar) UTF16_IS_TRAIL(uchar)
/** Does this code point require multiple code units? */
# define UTF_NEED_MULTIPLE_UCHAR(c) UTF16_NEED_MULTIPLE_UCHAR(c) # define UTF_NEED_MULTIPLE_UCHAR(c) UTF16_NEED_MULTIPLE_UCHAR(c)
/** How many code units are used to encode this code point? */
# define UTF_CHAR_LENGTH(c) UTF16_CHAR_LENGTH(c) # define UTF_CHAR_LENGTH(c) UTF16_CHAR_LENGTH(c)
/** How many code units are used at most for any Unicode code point? */
# define UTF_MAX_CHAR_LENGTH UTF16_MAX_CHAR_LENGTH # define UTF_MAX_CHAR_LENGTH UTF16_MAX_CHAR_LENGTH
/** Estimate the number of code units for a string based on the number of UTF-16 code units. */
# define UTF_ARRAY_SIZE(size) UTF16_ARRAY_SIZE(size) # define UTF_ARRAY_SIZE(size) UTF16_ARRAY_SIZE(size)
# define UTF_GET_CHAR_UNSAFE(s, i, c) UTF16_GET_CHAR_UNSAFE(s, i, c) # define UTF_GET_CHAR_UNSAFE(s, i, c) UTF16_GET_CHAR_UNSAFE(s, i, c)
# define UTF_GET_CHAR_SAFE(s, start, i, length, c, strict) UTF16_GET_CHAR_SAFE(s, start, i, length, c, strict) # define UTF_GET_CHAR_SAFE(s, start, i, length, c, strict) UTF16_GET_CHAR_SAFE(s, start, i, length, c, strict)
# define UTF_GET_CHAR_SAFE_STRICT(s, start, i, length, c) UTF16_GET_CHAR_SAFE_STRICT(s, start, i, length, c)
# define UTF_GET_CHAR_SAFE_LENIENT(s, start, i, length, c) UTF16_GET_CHAR_SAFE_LENIENT(s, start, i, length, c)
# define UTF_NEXT_CHAR_UNSAFE(s, i, c) UTF16_NEXT_CHAR_UNSAFE(s, i, c) # define UTF_NEXT_CHAR_UNSAFE(s, i, c) UTF16_NEXT_CHAR_UNSAFE(s, i, c)
# define UTF_NEXT_CHAR_SAFE(s, i, length, c, strict) UTF16_NEXT_CHAR_SAFE(s, i, length, c, strict) # define UTF_NEXT_CHAR_SAFE(s, i, length, c, strict) UTF16_NEXT_CHAR_SAFE(s, i, length, c, strict)
# define UTF_NEXT_CHAR_SAFE_STRICT(s, i, length, c) UTF16_NEXT_CHAR_SAFE_STRICT(s, i, length, c)
# define UTF_NEXT_CHAR_SAFE_LENIENT(s, i, length, c) UTF16_NEXT_CHAR_SAFE_LENIENT(s, i, length, c)
# define UTF_APPEND_CHAR_UNSAFE(s, i, c) UTF16_APPEND_CHAR_UNSAFE(s, i, c) # define UTF_APPEND_CHAR_UNSAFE(s, i, c) UTF16_APPEND_CHAR_UNSAFE(s, i, c)
# define UTF_APPEND_CHAR_SAFE(s, i, length, c) UTF16_APPEND_CHAR_SAFE(s, i, length, c) # define UTF_APPEND_CHAR_SAFE(s, i, length, c) UTF16_APPEND_CHAR_SAFE(s, i, length, c)
@ -177,8 +254,6 @@ typedef int32_t UTextOffset;
# define UTF_PREV_CHAR_UNSAFE(s, i, c) UTF16_PREV_CHAR_UNSAFE(s, i, c) # define UTF_PREV_CHAR_UNSAFE(s, i, c) UTF16_PREV_CHAR_UNSAFE(s, i, c)
# define UTF_PREV_CHAR_SAFE(s, start, i, c, strict) UTF16_PREV_CHAR_SAFE(s, start, i, c, strict) # define UTF_PREV_CHAR_SAFE(s, start, i, c, strict) UTF16_PREV_CHAR_SAFE(s, start, i, c, strict)
# define UTF_PREV_CHAR_SAFE_STRICT(s, start, i, c) UTF16_PREV_CHAR_SAFE_STRICT(s, start, i, c)
# define UTF_PREV_CHAR_SAFE_LENIENT(s, start, i, c) UTF16_PREV_CHAR_SAFE_LENIENT(s, start, i, c)
# define UTF_BACK_1_UNSAFE(s, i) UTF16_BACK_1_UNSAFE(s, i) # define UTF_BACK_1_UNSAFE(s, i) UTF16_BACK_1_UNSAFE(s, i)
# define UTF_BACK_1_SAFE(s, start, i) UTF16_BACK_1_SAFE(s, start, i) # define UTF_BACK_1_SAFE(s, start, i) UTF16_BACK_1_SAFE(s, start, i)
@ -198,35 +273,131 @@ typedef int32_t UTextOffset;
#else #else
# error UTF_SIZE must be undefined or one of { 8, 16, 32 } - only 16 is implemented # error UTF_SIZE must be undefined or one of { 8, 16, 32 } - only 16 is implemented
#endif #endif
/*@}*/
/* Define the default macros for handling UTF characters. ------------------- */ /* Define the default macros for handling UTF characters. ------------------- */
/**
* \def UTF_GET_CHAR(s, start, i, length, c)
*
* Set c to the code point that contains the code unit i.
* i could point to the first, the last, or an intermediate code unit.
* i is not modified.
* \pre 0<=i<length
*/
/**
* \def UTF_NEXT_CHAR(s, i, length, c)
*
* Set c to the code point that starts at code unit i
* and advance i to beyond the code units of this code point (post-increment).
* \pre 0<=i<length
* \post 0<i<=length
*/
/**
* \def UTF_APPEND_CHAR(s, i, length, c)
*
* Append the code units of code point c to the string at index i
* and advance i to beyond the new code units (post-increment).
* \pre 0<c<0x10ffff
* \pre 0<=i<length
* \post 0<i<=length
*/
/**
* \def UTF_FWD_1(s, i, length)
*
* Advance i to beyond the code units of the code point that begins at i.
* I.e., advance i by one code point.
* \pre 0<=i<length
* \post 0<i<=length
*/
/**
* \def UTF_FWD_N(s, i, length, n)
*
* Advance i to beyond the code units of the n code points where the first one begins at i.
* I.e., advance i by n code points.
* \pre 0<=i<length
* \post 0<i<=length
*/
/**
* \def UTF_SET_CHAR_START(s, start, i)
*
* Take the random-access index i and adjust it so that it points to the beginning
* of a code point.
* The input index points to any code unit of a code point and is moved to point to
* the first code unit of the same code point. i is never incremented.
* \pre start<=i<length
* \post start<=i<length
*/
/**
* \def UTF_PREV_CHAR(s, start, i, c)
*
* Set c to the code point that has code units before i
* and move i forward to the first code unit of this code point (pre-increment).
* \pre start<i<=length
* \post start<=i<length
*/
/**
* \def UTF_BACK_1(s, start, i)
*
* Move i forward to the first code unit of the code point that has code units before i.
* I.e., move i forward by one code point.
* \pre start<i<=length
* \post start<=i<length
*/
/**
* \def UTF_BACK_N(s, start, i, n)
*
* Move i forward to the first code unit of the n code points that have code units before i.
* I.e., move i forward by n code points.
* \pre start<i<=length
* \post start<=i<length
*/
/**
* \def UTF_SET_CHAR_LIMIT(s, start, i, length)
*
* Take the random-access index i and adjust it so that it points beyond the
* a code point. The input index points beyond any code unit
* of a code point and is moved to point beyond the last code unit of the same
* code point. i is never decremented.
* \pre start<i<=length
* \post start<i<=length
*/
#ifdef UTF_SAFE #ifdef UTF_SAFE
# define UTF_GET_CHAR(s, start, i, length, c) UTF_GET_CHAR_SAFE_LENIENT(s, start, i, length, c) # define UTF_GET_CHAR(s, start, i, length, c) UTF_GET_CHAR_SAFE(s, start, i, length, c, FALSE)
# define UTF_NEXT_CHAR(s, i, length, c) UTF_NEXT_CHAR_SAFE_LENIENT(s, i, length, c) # define UTF_NEXT_CHAR(s, i, length, c) UTF_NEXT_CHAR_SAFE(s, i, length, c, FALSE)
# define UTF_APPEND_CHAR(s, i, length, c) UTF_APPEND_CHAR_SAFE(s, i, length, c) # define UTF_APPEND_CHAR(s, i, length, c) UTF_APPEND_CHAR_SAFE(s, i, length, c)
# define UTF_FWD_1(s, i, length) UTF_FWD_1_SAFE(s, i, length) # define UTF_FWD_1(s, i, length) UTF_FWD_1_SAFE(s, i, length)
# define UTF_FWD_N(s, i, length, n) UTF_FWD_N_SAFE(s, i, length, n) # define UTF_FWD_N(s, i, length, n) UTF_FWD_N_SAFE(s, i, length, n)
# define UTF_SET_CHAR_START(s, start, i) UTF_SET_CHAR_START_SAFE(s, start, i) # define UTF_SET_CHAR_START(s, start, i) UTF_SET_CHAR_START_SAFE(s, start, i)
# define UTF_PREV_CHAR(s, start, i, c) UTF_PREV_CHAR_SAFE_LENIENT(s, start, i, c) # define UTF_PREV_CHAR(s, start, i, c) UTF_PREV_CHAR_SAFE(s, start, i, c, FALSE)
# define UTF_BACK_1(s, start, i) UTF_BACK_1_SAFE(s, start, i) # define UTF_BACK_1(s, start, i) UTF_BACK_1_SAFE(s, start, i)
# define UTF_BACK_N(s, start, i, n) UTF_BACK_N_SAFE(s, start, i, n) # define UTF_BACK_N(s, start, i, n) UTF_BACK_N_SAFE(s, start, i, n)
# define UTF_SET_CHAR_LIMIT(s, start, i, length) UTF_SET_CHAR_LIMIT_SAFE(s, start, i, length) # define UTF_SET_CHAR_LIMIT(s, start, i, length) UTF_SET_CHAR_LIMIT_SAFE(s, start, i, length)
#elif defined(UTF_STRICT) #elif defined(UTF_STRICT)
# define UTF_GET_CHAR(s, start, i, length, c) UTF_GET_CHAR_SAFE_STRICT(s, start, i, length, c) # define UTF_GET_CHAR(s, start, i, length, c) UTF_GET_CHAR_SAFE(s, start, i, length, c, TRUE)
# define UTF_NEXT_CHAR(s, i, length, c) UTF_NEXT_CHAR_SAFE_STRICT(s, i, length, c) # define UTF_NEXT_CHAR(s, i, length, c) UTF_NEXT_CHAR_SAFE(s, i, length, c, TRUE)
# define UTF_APPEND_CHAR(s, i, length, c) UTF_APPEND_CHAR_SAFE(s, i, length, c) # define UTF_APPEND_CHAR(s, i, length, c) UTF_APPEND_CHAR_SAFE(s, i, length, c)
# define UTF_FWD_1(s, i, length) UTF_FWD_1_SAFE(s, i, length) # define UTF_FWD_1(s, i, length) UTF_FWD_1_SAFE(s, i, length)
# define UTF_FWD_N(s, i, length, n) UTF_FWD_N_SAFE(s, i, length, n) # define UTF_FWD_N(s, i, length, n) UTF_FWD_N_SAFE(s, i, length, n)
# define UTF_SET_CHAR_START(s, start, i) UTF_SET_CHAR_START_SAFE(s, start, i) # define UTF_SET_CHAR_START(s, start, i) UTF_SET_CHAR_START_SAFE(s, start, i)
# define UTF_PREV_CHAR(s, start, i, c) UTF_PREV_CHAR_SAFE_STRICT(s, start, i, c) # define UTF_PREV_CHAR(s, start, i, c) UTF_PREV_CHAR_SAFE(s, start, i, c, TRUE)
# define UTF_BACK_1(s, start, i) UTF_BACK_1_SAFE(s, start, i) # define UTF_BACK_1(s, start, i) UTF_BACK_1_SAFE(s, start, i)
# define UTF_BACK_N(s, start, i, n) UTF_BACK_N_SAFE(s, start, i, n) # define UTF_BACK_N(s, start, i, n) UTF_BACK_N_SAFE(s, start, i, n)
# define UTF_SET_CHAR_LIMIT(s, start, i, length) UTF_SET_CHAR_LIMIT_SAFE(s, start, i, length) # define UTF_SET_CHAR_LIMIT(s, start, i, length) UTF_SET_CHAR_LIMIT_SAFE(s, start, i, length)

2
icu4c/source/common/unicode/utf32.h
View File

@ -15,7 +15,7 @@
*/ */
/** /**
* \file * \file
* \brief Description of UTF-32 macros * \brief C API: UTF-32 macros
* *
* This file defines macros to deal with UTF-32 code units and code points. * This file defines macros to deal with UTF-32 code units and code points.
* Signatures and semantics are the same as for the similarly named macros * Signatures and semantics are the same as for the similarly named macros

2
icu4c/source/common/unicode/utf8.h
View File

@ -16,7 +16,7 @@
/** /**
* \file * \file
* \brief Description of UTF-8 macros * \brief C API: UTF-8 macros
* *
* This file defines macros to deal with UTF-8 code units and code points. * This file defines macros to deal with UTF-8 code units and code points.
* Signatures and semantics are the same as for the similarly named macros * Signatures and semantics are the same as for the similarly named macros

197
icu4c/source/common/unicode/utypes.h
View File

@ -35,28 +35,41 @@
#include "unicode/umachine.h" #include "unicode/umachine.h"
#include "unicode/utf.h" #include "unicode/utf.h"
/*!
* \file
* \brief Basic definitions for ICU, for both C and C++ APIs
*
* This file defines basic types, constants, and enumerations directly or
* indirectly by including other header files, especially utf.h for the
* basic character and string definitions and umachine.h for consistent
* integer and other types.
*/
/*===========================================================================*/ /*===========================================================================*/
/* char Character set family */ /* char Character set family */
/*===========================================================================*/ /*===========================================================================*/
/* /**
* These definitions allow to specify the encoding of text * \def U_CHARSET_FAMILY
*
* <p>These definitions allow to specify the encoding of text
* in the char data type as defined by the platform and the compiler. * in the char data type as defined by the platform and the compiler.
* It is enough to determine the code point values of "invariant characters", * It is enough to determine the code point values of "invariant characters",
* which are the ones shared by all encodings that are in use * which are the ones shared by all encodings that are in use
* on a given platform. * on a given platform.</p>
* *
* Those "invariant characters" should be all the uppercase and lowercase * <p>Those "invariant characters" should be all the uppercase and lowercase
* latin letters, the digits, the space, and "basic punctuation". * latin letters, the digits, the space, and "basic punctuation".
* Also, '\n', '\r', '\t' should be available. * Also, '\n', '\r', '\t' should be available.</p>
* *
* The list of "invariant characters" is: * <p>The list of "invariant characters" is:<br>
* A-Z a-z 0-9 SPACE " % & ' ( ) * + , - . / : ; < = > ? _ * A-Z a-z 0-9 SPACE " % &amp; ' ( ) * + , - . / : ; &lt; = > ? _<br>
* (52 letters + 10 numbers + 20 punc/sym = 82 total) * (52 letters + 10 numbers + 20 punc/sym = 82 total)</p>
* *
* In other words, all the graphic characters in 7-bit ASCII should * <p>In other words, all the graphic characters in 7-bit ASCII should
* be safely accessible except the following: * be safely accessible except the following:</p>
* *
* \code
* '\' <backslash> * '\' <backslash>
* '[' <left bracket> * '[' <left bracket>
* ']' <right bracket> * ']' <right bracket>
@ -70,6 +83,7 @@
* '$' <dollar sign> * '$' <dollar sign>
* '@' <commercial at> * '@' <commercial at>
* '`' <grave accent> * '`' <grave accent>
* \endcode
*/ */
#define U_ASCII_FAMILY 0 #define U_ASCII_FAMILY 0
@ -82,19 +96,38 @@
/*===========================================================================*/ /*===========================================================================*/
/* Related version information */ /* Related version information */
/*===========================================================================*/ /*===========================================================================*/
/** The current ICU library version as a dotted-decimal string. */
#define U_ICU_VERSION "1.7" #define U_ICU_VERSION "1.7"
/** The current ICU library major/minor version as a string without dots, for library name suffixes. */
#define U_ICU_VERSION_SHORT "17" #define U_ICU_VERSION_SHORT "17"
/** An ICU version consists of up to 4 numbers from 0..255. */
#define U_MAX_VERSION_LENGTH 4 #define U_MAX_VERSION_LENGTH 4
/** In a string, ICU version fields are delimited by dots. */
#define U_VERSION_DELIMITER '.' #define U_VERSION_DELIMITER '.'
/** The maximum length of an ICU version string. */
#define U_MAX_VERSION_STRING_LENGTH 20 #define U_MAX_VERSION_STRING_LENGTH 20
/** The binary form of a version on ICU APIs is an array of 4 uint8_t. */
typedef uint8_t UVersionInfo[U_MAX_VERSION_LENGTH]; typedef uint8_t UVersionInfo[U_MAX_VERSION_LENGTH];
/*===========================================================================*/ /*===========================================================================*/
/* ICUDATA naming scheme */ /* ICUDATA naming scheme */
/*===========================================================================*/ /*===========================================================================*/
/**
* \def U_ICUDATA_TYPE_LETTER
*
* This is a platform-dependent string containing one letter:
* - b for big-endian, ASCII-family platforms
* - l for little-endian, ASCII-family platforms
* - e for big-endian, EBCDIC-family platforms
* This letter is part of the common data file name.
*/
#if U_CHARSET_FAMILY #if U_CHARSET_FAMILY
# if U_IS_BIG_ENDIAN # if U_IS_BIG_ENDIAN
/* EBCDIC - should always be BE */ /* EBCDIC - should always be BE */
@ -113,19 +146,25 @@ typedef uint8_t UVersionInfo[U_MAX_VERSION_LENGTH];
# endif # endif
#endif #endif
/* A single string literal containing the icudata stub name, i.e. 'icudt18e' for /** A single string literal containing the icudata stub name, i.e. 'icudt18e' for
ICU 1.8.x on EBCDIC, etc.. */ ICU 1.8.x on EBCDIC, etc.. */
#define U_ICUDATA_NAME "icudt" U_ICU_VERSION_SHORT U_ICUDATA_TYPE_LETTER #define U_ICUDATA_NAME "icudt" U_ICU_VERSION_SHORT U_ICUDATA_TYPE_LETTER
/* Work around the OS390 compiler issue, to be removed when the compiler /**
updates come out. */ * \def U_CALLCONV
* Work around the OS390 compiler issue, to be removed when the compiler
* updates come out.
*/
#if defined(OS390) && defined(XP_CPLUSPLUS) #if defined(OS390) && defined(XP_CPLUSPLUS)
# define U_CALLCONV __cdecl # define U_CALLCONV __cdecl
#else #else
# define U_CALLCONV # define U_CALLCONV
#endif #endif
/* Define NULL (the wrong way, cast to void *) if it does not exist. */ /**
* \def NULL
* Define NULL if necessary, to 0 for C++ and to ((void *)0) for C.
*/
#ifndef NULL #ifndef NULL
#ifdef XP_CPLUSPLUS #ifdef XP_CPLUSPLUS
#define NULL 0 #define NULL 0
@ -134,8 +173,10 @@ updates come out. */
#endif #endif
#endif #endif
/* Maximum value of a (void*) - use to indicate the limit of /**
an 'infinite' buffer. */ * \def U_MAX_PTR
* Maximum value of a (void*) - use to indicate the limit of an 'infinite' buffer.
*/
#ifndef U_MAX_PTR #ifndef U_MAX_PTR
#define U_MAX_PTR ((void*)-1) #define U_MAX_PTR ((void*)-1)
#endif #endif
@ -170,9 +211,11 @@ typedef double UDate;
* described in detail below. UClassID values can be compared using * described in detail below. UClassID values can be compared using
* operator==(). Nothing else should be done with them. * operator==(). Nothing else should be done with them.
* *
* \par
* getDynamicClassID() is declared in the base class of the hierarchy as * getDynamicClassID() is declared in the base class of the hierarchy as
* a pure virtual. Each concrete subclass implements it in the same way: * a pure virtual. Each concrete subclass implements it in the same way:
* *
* \code
* class Base { * class Base {
* public: * public:
* virtual UClassID getDynamicClassID() const = 0; * virtual UClassID getDynamicClassID() const = 0;
@ -183,10 +226,12 @@ typedef double UDate;
* virtual UClassID getDynamicClassID() const * virtual UClassID getDynamicClassID() const
* { return Derived::getStaticClassID(); } * { return Derived::getStaticClassID(); }
* } * }
* \endcode
* *
* Each concrete class implements getStaticClassID() as well, which allows * Each concrete class implements getStaticClassID() as well, which allows
* clients to test for a specific type. * clients to test for a specific type.
* *
* \code
* class Derived { * class Derived {
* public: * public:
* static UClassID getStaticClassID(); * static UClassID getStaticClassID();
@ -198,6 +243,7 @@ typedef double UDate;
* UClassID Derived::getStaticClassID() * UClassID Derived::getStaticClassID()
* { return (UClassID)&Derived::fgClassID; } * { return (UClassID)&Derived::fgClassID; }
* char Derived::fgClassID = 0; // Value is irrelevant * char Derived::fgClassID = 0; // Value is irrelevant
* \endcode
*/ */
typedef void* UClassID; typedef void* UClassID;
@ -206,11 +252,29 @@ typedef void* UClassID;
/* Shared library/DLL import-export API control */ /* Shared library/DLL import-export API control */
/*===========================================================================*/ /*===========================================================================*/
/** /*
* Control of symbol import/export. * Control of symbol import/export.
* The ICU is separated into two libraries. * The ICU is separated into two libraries.
*/ */
/**
* \def U_COMMON_API
* Set to export library symbols from inside the common library,
* and to import them from outside.
*/
/**
* \def U_I18N_API
* Set to export library symbols from inside the i18n library,
* and to import them from outside.
*/
/**
* \def U_LAYOUT_API
* Set to export library symbols from inside the layout engine library,
* and to import them from outside.
*/
#ifdef U_COMMON_IMPLEMENTATION #ifdef U_COMMON_IMPLEMENTATION
#define U_COMMON_API U_EXPORT #define U_COMMON_API U_EXPORT
#define U_I18N_API U_IMPORT #define U_I18N_API U_IMPORT
@ -230,6 +294,7 @@ typedef void* UClassID;
#endif #endif
/** /**
* \def U_STANDARD_CPP_NAMESPACE
* Control of C++ Namespace * Control of C++ Namespace
*/ */
#ifdef __cplusplus #ifdef __cplusplus
@ -242,38 +307,47 @@ typedef void* UClassID;
/* UErrorCode */ /* UErrorCode */
/*===========================================================================*/ /*===========================================================================*/
/** Error code to replace exception handling. /**
* So that the code is compatible with all C++ compilers. * Error code to replace exception handling, so that the code is compatible with all C++ compilers,
* and to use the same mechanism for C and C++.
*
* \par
* ICU functions that take a reference (C++) or a pointer (C) to a UErrorCode
* first test if(U_FAILURE(errorCode)) { return immediately; }
* so that in a chain of such functions the first one that sets an error code
* causes the following ones to not perform any operations.
*
* \par
* Error codes should be tested using U_FAILURE() and U_SUCCESS().
*/ */
enum UErrorCode { enum UErrorCode {
U_ERROR_INFO_START = -128, /* Start of information results (semantically successful) */ U_ERROR_INFO_START = -128, /**< Start of information results (semantically successful) */
U_USING_FALLBACK_ERROR = -128, U_USING_FALLBACK_ERROR = -128, /**< A resource bundle lookup returned a fallback result (not an error) */
U_USING_DEFAULT_ERROR = -127, U_USING_DEFAULT_ERROR = -127, /**< A reousrce bundle lookup returned a result from the root locale (not an error) */
U_ERROR_INFO_LIMIT, U_ERROR_INFO_LIMIT, /**< This must always be the last warning value to indicate the limit for UErrorCode warnings (last warning code +1) */
/** success */ U_ZERO_ERROR = 0, /**< No error, no warning. */
U_ZERO_ERROR = 0,
U_ILLEGAL_ARGUMENT_ERROR = 1, /* Start of codes indicating failure */ U_ILLEGAL_ARGUMENT_ERROR = 1, /**< Start of codes indicating failure */
U_MISSING_RESOURCE_ERROR = 2, U_MISSING_RESOURCE_ERROR = 2,
U_INVALID_FORMAT_ERROR = 3, U_INVALID_FORMAT_ERROR = 3,
U_FILE_ACCESS_ERROR = 4, U_FILE_ACCESS_ERROR = 4,
U_INTERNAL_PROGRAM_ERROR = 5, /* Indicates a bug in the library code */ U_INTERNAL_PROGRAM_ERROR = 5, /**< Indicates a bug in the library code */
U_MESSAGE_PARSE_ERROR = 6, U_MESSAGE_PARSE_ERROR = 6,
U_MEMORY_ALLOCATION_ERROR = 7, /* Memory allocation error */ U_MEMORY_ALLOCATION_ERROR = 7, /**< Memory allocation error */
U_INDEX_OUTOFBOUNDS_ERROR = 8, U_INDEX_OUTOFBOUNDS_ERROR = 8,
U_PARSE_ERROR = 9, /* Equivalent to Java ParseException */ U_PARSE_ERROR = 9, /**< Equivalent to Java ParseException */
U_INVALID_CHAR_FOUND = 10, /* In the Character conversion routines: Invalid character or sequence was encountered*/ U_INVALID_CHAR_FOUND = 10, /**< In the Character conversion routines: Invalid character or sequence was encountered */
U_TRUNCATED_CHAR_FOUND = 11, /* In the Character conversion routines: More bytes are required to complete the conversion successfully*/ U_TRUNCATED_CHAR_FOUND = 11, /**< In the Character conversion routines: More bytes are required to complete the conversion successfully */
U_ILLEGAL_CHAR_FOUND = 12, /* In codeset conversion: a sequence that does NOT belong in the codepage has been encountered*/ U_ILLEGAL_CHAR_FOUND = 12, /**< In codeset conversion: a sequence that does NOT belong in the codepage has been encountered */
U_INVALID_TABLE_FORMAT = 13, /* Conversion table file found, but corrupted*/ U_INVALID_TABLE_FORMAT = 13, /**< Conversion table file found, but corrupted */
U_INVALID_TABLE_FILE = 14, /* Conversion table file not found*/ U_INVALID_TABLE_FILE = 14, /**< Conversion table file not found */
U_BUFFER_OVERFLOW_ERROR = 15, /* A result would not fit in the supplied buffer */ U_BUFFER_OVERFLOW_ERROR = 15, /**< A result would not fit in the supplied buffer */
U_UNSUPPORTED_ERROR = 16, /* Requested operation not supported in current context */ U_UNSUPPORTED_ERROR = 16, /**< Requested operation not supported in current context */
U_RESOURCE_TYPE_MISMATCH = 17, /* an operation is requested over a resource that does not support it*/ U_RESOURCE_TYPE_MISMATCH = 17, /**< an operation is requested over a resource that does not support it */
U_ILLEGAL_ESCAPE_SEQUENCE = 18, /* ISO-2022 illlegal escape sequence*/ U_ILLEGAL_ESCAPE_SEQUENCE = 18, /**< ISO-2022 illlegal escape sequence */
U_UNSUPPORTED_ESCAPE_SEQUENCE = 19, /* ISO-2022 unsupported escape sequence*/ U_UNSUPPORTED_ESCAPE_SEQUENCE = 19, /**< ISO-2022 unsupported escape sequence */
U_ERROR_LIMIT U_ERROR_LIMIT /**< This must always be the last value to indicate the limit for UErrorCode (last error code +1) */
}; };
#ifndef XP_CPLUSPLUS #ifndef XP_CPLUSPLUS
@ -282,22 +356,35 @@ typedef enum UErrorCode UErrorCode;
/* Use the following to determine if an UErrorCode represents */ /* Use the following to determine if an UErrorCode represents */
/* operational success or failure. */ /* operational success or failure. */
#ifdef XP_CPLUSPLUS #ifdef XP_CPLUSPLUS
/** @stable */ /**
inline UBool U_SUCCESS(UErrorCode code) { return (UBool)(code<=U_ZERO_ERROR); } * Does the error code indicate success?
/** @stable */ * @stable
inline UBool U_FAILURE(UErrorCode code) { return (UBool)(code>U_ZERO_ERROR); } */
inline UBool U_SUCCESS(UErrorCode code) { return (UBool)(code<=U_ZERO_ERROR); }
/**
* Does the error code indicate a failure?
* @stable
*/
inline UBool U_FAILURE(UErrorCode code) { return (UBool)(code>U_ZERO_ERROR); }
#else #else
/** @stable */ /**
#define U_SUCCESS(x) ((x)<=U_ZERO_ERROR) * Does the error code indicate success?
/** @stable */ * @stable
#define U_FAILURE(x) ((x)>U_ZERO_ERROR) */
# define U_SUCCESS(x) ((x)<=U_ZERO_ERROR)
/**
* Does the error code indicate a failure?
* @stable
*/
# define U_FAILURE(x) ((x)>U_ZERO_ERROR)
#endif #endif
/** /**
* Return a string for a UErrorCode value. * Return a string for a UErrorCode value.
* The string will be the same as the name of the error code constant * The string will be the same as the name of the error code constant
* in the enum above. * in the UErrorCode enum above.
*/ */
U_CAPI const char * U_EXPORT2 U_CAPI const char * U_EXPORT2
u_errorName(UErrorCode code); u_errorName(UErrorCode code);
@ -307,16 +394,16 @@ u_errorName(UErrorCode code);
#define U_COPYRIGHT_STRING_LENGTH 160 #define U_COPYRIGHT_STRING_LENGTH 160
/**
* Mutex data type.
* @internal
*/
typedef void *UMTX;
/*===========================================================================*/ /*===========================================================================*/
/* Include header for platform utilies */ /* Include header for platform utilies */
/*===========================================================================*/ /*===========================================================================*/
/* Mutex data type. INTERNAL.*/
typedef void *UMTX;
#include "unicode/putil.h" #include "unicode/putil.h"
#endif /* _UTYPES */ #endif /* _UTYPES */