ustring.h

linux下开源浏览器WebKit的源码,市面上的很多商用浏览器都是移植自WebKit

/*
**********************************************************************
*   Copyright (C) 1998-2004, International Business Machines
*   Corporation and others.  All Rights Reserved.
**********************************************************************
*
* File ustring.h
*
* Modification History:
*
*   Date        Name        Description
*   12/07/98    bertrand    Creation.
******************************************************************************
*/

#ifndef USTRING_H
#define USTRING_H

#include "unicode/utypes.h"
#include "unicode/putil.h"
#include "unicode/uiter.h"

/** Simple declaration for u_strToTitle() to avoid including unicode/ubrk.h. @stable ICU 2.1*/
#ifndef UBRK_TYPEDEF_UBREAK_ITERATOR
#   define UBRK_TYPEDEF_UBREAK_ITERATOR
    typedef void UBreakIterator;
#endif

/**
 * \file
 * \brief C API: Unicode string handling functions
 *
 * These C API functions provide general Unicode string handling.
 *
 * Some functions are equivalent in name, signature, and behavior to the ANSI C <string.h>
 * functions. (For example, they do not check for bad arguments like NULL string pointers.)
 * In some cases, only the thread-safe variant of such a function is implemented here
 * (see u_strtok_r()).
 *
 * Other functions provide more Unicode-specific functionality like locale-specific
 * upper/lower-casing and string comparison in code point order.
 *
 * ICU uses 16-bit Unicode (UTF-16) in the form of arrays of UChar code units.
 * UTF-16 encodes each Unicode code point with either one or two UChar code units.
 * (This is the default form of Unicode, and a forward-compatible extension of the original,
 * fixed-width form that was known as UCS-2. UTF-16 superseded UCS-2 with Unicode 2.0
 * in 1996.)
 *
 * Some APIs accept a 32-bit UChar32 value for a single code point.
 *
 * ICU also handles 16-bit Unicode text with unpaired surrogates.
 * Such text is not well-formed UTF-16.
 * Code-point-related functions treat unpaired surrogates as surrogate code points,
 * i.e., as separate units.
 *
 * Although UTF-16 is a variable-width encoding form (like some legacy multi-byte encodings),
 * it is much more efficient even for random access because the code unit values
 * for single-unit characters vs. lead units vs. trail units are completely disjoint.
 * This means that it is easy to determine character (code point) boundaries from
 * random offsets in the string.
 *
 * Unicode (UTF-16) string processing is optimized for the single-unit case.
 * Although it is important to support supplementary characters
 * (which use pairs of lead/trail code units called "surrogates"),
 * their occurrence is rare. Almost all characters in modern use require only
 * a single UChar code unit (i.e., their code point values are <=0xffff).
 *
 * For more details see the User Guide Strings chapter (http://oss.software.ibm.com/icu/userguide/strings.html).
 * For a discussion of the handling of unpaired surrogates see also
 * Jitterbug 2145 and its icu mailing list proposal on 2002-sep-18.
 */

/**
 * Determine the length of an array of UChar.
 *
 * @param s The array of UChars, NULL (U+0000) terminated.
 * @return The number of UChars in <code>chars</code>, minus the terminator.
 * @stable ICU 2.0
 */
U_STABLE int32_t U_EXPORT2
u_strlen(const UChar *s);

/**
 * Count Unicode code points in the length UChar code units of the string.
 * A code point may occupy either one or two UChar code units.
 * Counting code points involves reading all code units.
 *
 * This functions is basically the inverse of the U16_FWD_N() macro (see utf.h).
 *
 * @param s The input string.
 * @param length The number of UChar code units to be checked, or -1 to count all
 *               code points before the first NUL (U+0000).
 * @return The number of code points in the specified code units.
 * @stable ICU 2.0
 */
U_STABLE int32_t U_EXPORT2
u_countChar32(const UChar *s, int32_t length);

/**
 * Check if the string contains more Unicode code points than a certain number.
 * This is more efficient than counting all code points in the entire string
 * and comparing that number with a threshold.
 * This function may not need to scan the string at all if the length is known
 * (not -1 for NUL-termination) and falls within a certain range, and
 * never needs to count more than 'number+1' code points.
 * Logically equivalent to (u_countChar32(s, length)>number).
 * A Unicode code point may occupy either one or two UChar code units.
 *
 * @param s The input string.
 * @param length The length of the string, or -1 if it is NUL-terminated.
 * @param number The number of code points in the string is compared against
 *               the 'number' parameter.
 * @return Boolean value for whether the string contains more Unicode code points
 *         than 'number'. Same as (u_countChar32(s, length)>number).
 * @stable ICU 2.4
 */
U_STABLE UBool U_EXPORT2
u_strHasMoreChar32Than(const UChar *s, int32_t length, int32_t number);

/**
 * Concatenate two ustrings.  Appends a copy of <code>src</code>,
 * including the null terminator, to <code>dst</code>. The initial copied
 * character from <code>src</code> overwrites the null terminator in <code>dst</code>.
 *
 * @param dst The destination string.
 * @param src The source string.
 * @return A pointer to <code>dst</code>.
 * @stable ICU 2.0
 */
U_STABLE UChar* U_EXPORT2
u_strcat(UChar     *dst, 
    const UChar     *src);

/**
 * Concatenate two ustrings.  
 * Appends at most <code>n</code> characters from <code>src</code> to <code>dst</code>.
 * Adds a terminating NUL.
 * If src is too long, then only <code>n-1</code> characters will be copied
 * before the terminating NUL.
 * If <code>n&lt;=0</code> then dst is not modified.
 *
 * @param dst The destination string.
 * @param src The source string.
 * @param n The maximum number of characters to compare.
 * @return A pointer to <code>dst</code>.
 * @stable ICU 2.0
 */
U_STABLE UChar* U_EXPORT2
u_strncat(UChar     *dst, 
     const UChar     *src, 
     int32_t     n);

/**
 * Find the first occurrence of a substring in a string.
 * The substring is found at code point boundaries.
 * That means that if the substring begins with
 * a trail surrogate or ends with a lead surrogate,
 * then it is found only if these surrogates stand alone in the text.
 * Otherwise, the substring edge units would be matched against
 * halves of surrogate pairs.
 *
 * @param s The string to search (NUL-terminated).
 * @param substring The substring to find (NUL-terminated).
 * @return A pointer to the first occurrence of <code>substring</code> in <code>s</code>,
 *         or <code>s</code> itself if the <code>substring</code> is empty,
 *         or <code>NULL</code> if <code>substring</code> is not in <code>s</code>.
 * @stable ICU 2.0
 *
 * @see u_strrstr
 * @see u_strFindFirst
 * @see u_strFindLast
 */
U_STABLE UChar * U_EXPORT2
u_strstr(const UChar *s, const UChar *substring);

/**
 * Find the first occurrence of a substring in a string.
 * The substring is found at code point boundaries.
 * That means that if the substring begins with
 * a trail surrogate or ends with a lead surrogate,
 * then it is found only if these surrogates stand alone in the text.
 * Otherwise, the substring edge units would be matched against
 * halves of surrogate pairs.
 *
 * @param s The string to search.
 * @param length The length of s (number of UChars), or -1 if it is NUL-terminated.
 * @param substring The substring to find (NUL-terminated).
 * @param subLength The length of substring (number of UChars), or -1 if it is NUL-terminated.
 * @return A pointer to the first occurrence of <code>substring</code> in <code>s</code>,
 *         or <code>s</code> itself if the <code>substring</code> is empty,
 *         or <code>NULL</code> if <code>substring</code> is not in <code>s</code>.
 * @stable ICU 2.4
 *
 * @see u_strstr
 * @see u_strFindLast
 */
U_STABLE UChar * U_EXPORT2
u_strFindFirst(const UChar *s, int32_t length, const UChar *substring, int32_t subLength);

/**
 * Find the first occurrence of a BMP code point in a string.
 * A surrogate code point is found only if its match in the text is not
 * part of a surrogate pair.
 * A NUL character is found at the string terminator.
 *
 * @param s The string to search (NUL-terminated).
 * @param c The BMP code point to find.
 * @return A pointer to the first occurrence of <code>c</code> in <code>s</code>
 *         or <code>NULL</code> if <code>c</code> is not in <code>s</code>.
 * @stable ICU 2.0
 *
 * @see u_strchr32
 * @see u_memchr
 * @see u_strstr
 * @see u_strFindFirst
 */
U_STABLE UChar * U_EXPORT2
u_strchr(const UChar *s, UChar c);

/**
 * Find the first occurrence of a code point in a string.
 * A surrogate code point is found only if its match in the text is not
 * part of a surrogate pair.
 * A NUL character is found at the string terminator.
 *
 * @param s The string to search (NUL-terminated).
 * @param c The code point to find.
 * @return A pointer to the first occurrence of <code>c</code> in <code>s</code>
 *         or <code>NULL</code> if <code>c</code> is not in <code>s</code>.
 * @stable ICU 2.0
 *
 * @see u_strchr
 * @see u_memchr32
 * @see u_strstr
 * @see u_strFindFirst
 */
U_STABLE UChar * U_EXPORT2
u_strchr32(const UChar *s, UChar32 c);

/**
 * Find the last occurrence of a substring in a string.
 * The substring is found at code point boundaries.
 * That means that if the substring begins with
 * a trail surrogate or ends with a lead surrogate,
 * then it is found only if these surrogates stand alone in the text.
 * Otherwise, the substring edge units would be matched against
 * halves of surrogate pairs.
 *
 * @param s The string to search (NUL-terminated).
 * @param substring The substring to find (NUL-terminated).
 * @return A pointer to the last occurrence of <code>substring</code> in <code>s</code>,
 *         or <code>s</code> itself if the <code>substring</code> is empty,
 *         or <code>NULL</code> if <code>substring</code> is not in <code>s</code>.
 * @stable ICU 2.4
 *
 * @see u_strstr
 * @see u_strFindFirst
 * @see u_strFindLast
 */
U_STABLE UChar * U_EXPORT2
u_strrstr(const UChar *s, const UChar *substring);

/**
 * Find the last occurrence of a substring in a string.
 * The substring is found at code point boundaries.
 * That means that if the substring begins with
 * a trail surrogate or ends with a lead surrogate,
 * then it is found only if these surrogates stand alone in the text.
 * Otherwise, the substring edge units would be matched against
 * halves of surrogate pairs.
 *
 * @param s The string to search.
 * @param length The length of s (number of UChars), or -1 if it is NUL-terminated.
 * @param substring The substring to find (NUL-terminated).
 * @param subLength The length of substring (number of UChars), or -1 if it is NUL-terminated.
 * @return A pointer to the last occurrence of <code>substring</code> in <code>s</code>,
 *         or <code>s</code> itself if the <code>substring</code> is empty,
 *         or <code>NULL</code> if <code>substring</code> is not in <code>s</code>.
 * @stable ICU 2.4
 *
 * @see u_strstr
 * @see u_strFindLast
 */
U_STABLE UChar * U_EXPORT2
u_strFindLast(const UChar *s, int32_t length, const UChar *substring, int32_t subLength);

/**
 * Find the last occurrence of a BMP code point in a string.
 * A surrogate code point is found only if its match in the text is not
 * part of a surrogate pair.
 * A NUL character is found at the string terminator.
 *
 * @param s The string to search (NUL-terminated).
 * @param c The BMP code point to find.
 * @return A pointer to the last occurrence of <code>c</code> in <code>s</code>
 *         or <code>NULL</code> if <code>c</code> is not in <code>s</code>.
 * @stable ICU 2.4
 *
 * @see u_strrchr32
 * @see u_memrchr
 * @see u_strrstr
 * @see u_strFindLast
 */
U_STABLE UChar * U_EXPORT2
u_strrchr(const UChar *s, UChar c);

/**
 * Find the last occurrence of a code point in a string.
 * A surrogate code point is found only if its match in the text is not
 * part of a surrogate pair.
 * A NUL character is found at the string terminator.
 *
 * @param s The string to search (NUL-terminated).
 * @param c The code point to find.
 * @return A pointer to the last occurrence of <code>c</code> in <code>s</code>
 *         or <code>NULL</code> if <code>c</code> is not in <code>s</code>.
 * @stable ICU 2.4
 *
 * @see u_strrchr
 * @see u_memchr32
 * @see u_strrstr
 * @see u_strFindLast
 */
U_STABLE UChar * U_EXPORT2
u_strrchr32(const UChar *s, UChar32 c);

/**
 * Locates the first occurrence in the string <code>string</code> of any of the characters
 * in the string <code>matchSet</code>.
 * Works just like C's strpbrk but with Unicode.
 *