1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/common/unicode/rep.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,261 @@
1.4 +/*
1.5 +**************************************************************************
1.6 +* Copyright (C) 1999-2012, International Business Machines Corporation and
1.7 +* others. All Rights Reserved.
1.8 +**************************************************************************
1.9 +* Date Name Description
1.10 +* 11/17/99 aliu Creation. Ported from java. Modified to
1.11 +* match current UnicodeString API. Forced
1.12 +* to use name "handleReplaceBetween" because
1.13 +* of existing methods in UnicodeString.
1.14 +**************************************************************************
1.15 +*/
1.16 +
1.17 +#ifndef REP_H
1.18 +#define REP_H
1.19 +
1.20 +#include "unicode/uobject.h"
1.21 +
1.22 +/**
1.23 + * \file
1.24 + * \brief C++ API: Replaceable String
1.25 + */
1.26 +
1.27 +U_NAMESPACE_BEGIN
1.28 +
1.29 +class UnicodeString;
1.30 +
1.31 +/**
1.32 + * <code>Replaceable</code> is an abstract base class representing a
1.33 + * string of characters that supports the replacement of a range of
1.34 + * itself with a new string of characters. It is used by APIs that
1.35 + * change a piece of text while retaining metadata. Metadata is data
1.36 + * other than the Unicode characters returned by char32At(). One
1.37 + * example of metadata is style attributes; another is an edit
1.38 + * history, marking each character with an author and revision number.
1.39 + *
1.40 + * <p>An implicit aspect of the <code>Replaceable</code> API is that
1.41 + * during a replace operation, new characters take on the metadata of
1.42 + * the old characters. For example, if the string "the <b>bold</b>
1.43 + * font" has range (4, 8) replaced with "strong", then it becomes "the
1.44 + * <b>strong</b> font".
1.45 + *
1.46 + * <p><code>Replaceable</code> specifies ranges using a start
1.47 + * offset and a limit offset. The range of characters thus specified
1.48 + * includes the characters at offset start..limit-1. That is, the
1.49 + * start offset is inclusive, and the limit offset is exclusive.
1.50 + *
1.51 + * <p><code>Replaceable</code> also includes API to access characters
1.52 + * in the string: <code>length()</code>, <code>charAt()</code>,
1.53 + * <code>char32At()</code>, and <code>extractBetween()</code>.
1.54 + *
1.55 + * <p>For a subclass to support metadata, typical behavior of
1.56 + * <code>replace()</code> is the following:
1.57 + * <ul>
1.58 + * <li>Set the metadata of the new text to the metadata of the first
1.59 + * character replaced</li>
1.60 + * <li>If no characters are replaced, use the metadata of the
1.61 + * previous character</li>
1.62 + * <li>If there is no previous character (i.e. start == 0), use the
1.63 + * following character</li>
1.64 + * <li>If there is no following character (i.e. the replaceable was
1.65 + * empty), use default metadata.<br>
1.66 + * <li>If the code point U+FFFF is seen, it should be interpreted as
1.67 + * a special marker having no metadata<li>
1.68 + * </li>
1.69 + * </ul>
1.70 + * If this is not the behavior, the subclass should document any differences.
1.71 + * @author Alan Liu
1.72 + * @stable ICU 2.0
1.73 + */
1.74 +class U_COMMON_API Replaceable : public UObject {
1.75 +
1.76 +public:
1.77 + /**
1.78 + * Destructor.
1.79 + * @stable ICU 2.0
1.80 + */
1.81 + virtual ~Replaceable();
1.82 +
1.83 + /**
1.84 + * Returns the number of 16-bit code units in the text.
1.85 + * @return number of 16-bit code units in text
1.86 + * @stable ICU 1.8
1.87 + */
1.88 + inline int32_t length() const;
1.89 +
1.90 + /**
1.91 + * Returns the 16-bit code unit at the given offset into the text.
1.92 + * @param offset an integer between 0 and <code>length()</code>-1
1.93 + * inclusive
1.94 + * @return 16-bit code unit of text at given offset
1.95 + * @stable ICU 1.8
1.96 + */
1.97 + inline UChar charAt(int32_t offset) const;
1.98 +
1.99 + /**
1.100 + * Returns the 32-bit code point at the given 16-bit offset into
1.101 + * the text. This assumes the text is stored as 16-bit code units
1.102 + * with surrogate pairs intermixed. If the offset of a leading or
1.103 + * trailing code unit of a surrogate pair is given, return the
1.104 + * code point of the surrogate pair.
1.105 + *
1.106 + * @param offset an integer between 0 and <code>length()</code>-1
1.107 + * inclusive
1.108 + * @return 32-bit code point of text at given offset
1.109 + * @stable ICU 1.8
1.110 + */
1.111 + inline UChar32 char32At(int32_t offset) const;
1.112 +
1.113 + /**
1.114 + * Copies characters in the range [<tt>start</tt>, <tt>limit</tt>)
1.115 + * into the UnicodeString <tt>target</tt>.
1.116 + * @param start offset of first character which will be copied
1.117 + * @param limit offset immediately following the last character to
1.118 + * be copied
1.119 + * @param target UnicodeString into which to copy characters.
1.120 + * @return A reference to <TT>target</TT>
1.121 + * @stable ICU 2.1
1.122 + */
1.123 + virtual void extractBetween(int32_t start,
1.124 + int32_t limit,
1.125 + UnicodeString& target) const = 0;
1.126 +
1.127 + /**
1.128 + * Replaces a substring of this object with the given text. If the
1.129 + * characters being replaced have metadata, the new characters
1.130 + * that replace them should be given the same metadata.
1.131 + *
1.132 + * <p>Subclasses must ensure that if the text between start and
1.133 + * limit is equal to the replacement text, that replace has no
1.134 + * effect. That is, any metadata
1.135 + * should be unaffected. In addition, subclasses are encouraged to
1.136 + * check for initial and trailing identical characters, and make a
1.137 + * smaller replacement if possible. This will preserve as much
1.138 + * metadata as possible.
1.139 + * @param start the beginning index, inclusive; <code>0 <= start
1.140 + * <= limit</code>.
1.141 + * @param limit the ending index, exclusive; <code>start <= limit
1.142 + * <= length()</code>.
1.143 + * @param text the text to replace characters <code>start</code>
1.144 + * to <code>limit - 1</code>
1.145 + * @stable ICU 2.0
1.146 + */
1.147 + virtual void handleReplaceBetween(int32_t start,
1.148 + int32_t limit,
1.149 + const UnicodeString& text) = 0;
1.150 + // Note: All other methods in this class take the names of
1.151 + // existing UnicodeString methods. This method is the exception.
1.152 + // It is named differently because all replace methods of
1.153 + // UnicodeString return a UnicodeString&. The 'between' is
1.154 + // required in order to conform to the UnicodeString naming
1.155 + // convention; API taking start/length are named <operation>, and
1.156 + // those taking start/limit are named <operationBetween>. The
1.157 + // 'handle' is added because 'replaceBetween' and
1.158 + // 'doReplaceBetween' are already taken.
1.159 +
1.160 + /**
1.161 + * Copies a substring of this object, retaining metadata.
1.162 + * This method is used to duplicate or reorder substrings.
1.163 + * The destination index must not overlap the source range.
1.164 + *
1.165 + * @param start the beginning index, inclusive; <code>0 <= start <=
1.166 + * limit</code>.
1.167 + * @param limit the ending index, exclusive; <code>start <= limit <=
1.168 + * length()</code>.
1.169 + * @param dest the destination index. The characters from
1.170 + * <code>start..limit-1</code> will be copied to <code>dest</code>.
1.171 + * Implementations of this method may assume that <code>dest <= start ||
1.172 + * dest >= limit</code>.
1.173 + * @stable ICU 2.0
1.174 + */
1.175 + virtual void copy(int32_t start, int32_t limit, int32_t dest) = 0;
1.176 +
1.177 + /**
1.178 + * Returns true if this object contains metadata. If a
1.179 + * Replaceable object has metadata, calls to the Replaceable API
1.180 + * must be made so as to preserve metadata. If it does not, calls
1.181 + * to the Replaceable API may be optimized to improve performance.
1.182 + * The default implementation returns true.
1.183 + * @return true if this object contains metadata
1.184 + * @stable ICU 2.2
1.185 + */
1.186 + virtual UBool hasMetaData() const;
1.187 +
1.188 + /**
1.189 + * Clone this object, an instance of a subclass of Replaceable.
1.190 + * Clones can be used concurrently in multiple threads.
1.191 + * If a subclass does not implement clone(), or if an error occurs,
1.192 + * then NULL is returned.
1.193 + * The clone functions in all subclasses return a pointer to a Replaceable
1.194 + * because some compilers do not support covariant (same-as-this)
1.195 + * return types; cast to the appropriate subclass if necessary.
1.196 + * The caller must delete the clone.
1.197 + *
1.198 + * @return a clone of this object
1.199 + *
1.200 + * @see getDynamicClassID
1.201 + * @stable ICU 2.6
1.202 + */
1.203 + virtual Replaceable *clone() const;
1.204 +
1.205 +protected:
1.206 +
1.207 + /**
1.208 + * Default constructor.
1.209 + * @stable ICU 2.4
1.210 + */
1.211 + inline Replaceable();
1.212 +
1.213 + /*
1.214 + * Assignment operator not declared. The compiler will provide one
1.215 + * which does nothing since this class does not contain any data members.
1.216 + * API/code coverage may show the assignment operator as present and
1.217 + * untested - ignore.
1.218 + * Subclasses need this assignment operator if they use compiler-provided
1.219 + * assignment operators of their own. An alternative to not declaring one
1.220 + * here would be to declare and empty-implement a protected or public one.
1.221 + Replaceable &Replaceable::operator=(const Replaceable &);
1.222 + */
1.223 +
1.224 + /**
1.225 + * Virtual version of length().
1.226 + * @stable ICU 2.4
1.227 + */
1.228 + virtual int32_t getLength() const = 0;
1.229 +
1.230 + /**
1.231 + * Virtual version of charAt().
1.232 + * @stable ICU 2.4
1.233 + */
1.234 + virtual UChar getCharAt(int32_t offset) const = 0;
1.235 +
1.236 + /**
1.237 + * Virtual version of char32At().
1.238 + * @stable ICU 2.4
1.239 + */
1.240 + virtual UChar32 getChar32At(int32_t offset) const = 0;
1.241 +};
1.242 +
1.243 +inline Replaceable::Replaceable() {}
1.244 +
1.245 +inline int32_t
1.246 +Replaceable::length() const {
1.247 + return getLength();
1.248 +}
1.249 +
1.250 +inline UChar
1.251 +Replaceable::charAt(int32_t offset) const {
1.252 + return getCharAt(offset);
1.253 +}
1.254 +
1.255 +inline UChar32
1.256 +Replaceable::char32At(int32_t offset) const {
1.257 + return getChar32At(offset);
1.258 +}
1.259 +
1.260 +// There is no rep.cpp, see unistr.cpp for Replaceable function implementations.
1.261 +
1.262 +U_NAMESPACE_END
1.263 +
1.264 +#endif
