icuSources/common/unicode/rep.h

   1 // © 2016 and later: Unicode, Inc. and others.
   2 // License & terms of use: http://www.unicode.org/copyright.html
   3 /*
   4 **************************************************************************
   5 * Copyright (C) 1999-2012, International Business Machines Corporation and
   6 * others. All Rights Reserved.
   7 **************************************************************************
   8 *   Date        Name        Description
   9 *   11/17/99    aliu        Creation.  Ported from java.  Modified to
  10 *                           match current UnicodeString API.  Forced
  11 *                           to use name "handleReplaceBetween" because
  12 *                           of existing methods in UnicodeString.
  13 **************************************************************************
  14 */
  15
  16 #ifndef REP_H
  17 #define REP_H
  18
  19 #include "unicode/uobject.h"
  20
  21 /**
  22  * \file
  23  * \brief C++ API: Replaceable String
  24  */
  25
  26 #if U_SHOW_CPLUSPLUS_API
  27 U_NAMESPACE_BEGIN
  28
  29 class UnicodeString;
  30
  31 /**
  32  * <code>Replaceable</code> is an abstract base class representing a
  33  * string of characters that supports the replacement of a range of
  34  * itself with a new string of characters.  It is used by APIs that
  35  * change a piece of text while retaining metadata.  Metadata is data
  36  * other than the Unicode characters returned by char32At().  One
  37  * example of metadata is style attributes; another is an edit
  38  * history, marking each character with an author and revision number.
  39  *
  40  * <p>An implicit aspect of the <code>Replaceable</code> API is that
  41  * during a replace operation, new characters take on the metadata of
  42  * the old characters.  For example, if the string "the <b>bold</b>
  43  * font" has range (4, 8) replaced with "strong", then it becomes "the
  44  * <b>strong</b> font".
  45  *
  46  * <p><code>Replaceable</code> specifies ranges using a start
  47  * offset and a limit offset.  The range of characters thus specified
  48  * includes the characters at offset start..limit-1.  That is, the
  49  * start offset is inclusive, and the limit offset is exclusive.
  50  *
  51  * <p><code>Replaceable</code> also includes API to access characters
  52  * in the string: <code>length()</code>, <code>charAt()</code>,
  53  * <code>char32At()</code>, and <code>extractBetween()</code>.
  54  *
  55  * <p>For a subclass to support metadata, typical behavior of
  56  * <code>replace()</code> is the following:
  57  * <ul>
  58  *   <li>Set the metadata of the new text to the metadata of the first
  59  *   character replaced</li>
  60  *   <li>If no characters are replaced, use the metadata of the
  61  *   previous character</li>
  62  *   <li>If there is no previous character (i.e. start == 0), use the
  63  *   following character</li>
  64  *   <li>If there is no following character (i.e. the replaceable was
  65  *   empty), use default metadata.<br>
  66  *   <li>If the code point U+FFFF is seen, it should be interpreted as
  67  *   a special marker having no metadata<li>
  68  *   </li>
  69  * </ul>
  70  * If this is not the behavior, the subclass should document any differences.
  71  * @author Alan Liu
  72  * @stable ICU 2.0
  73  */
  74 class U_COMMON_API Replaceable : public UObject {
  75
  76 public:
  77     /**
  78      * Destructor.
  79      * @stable ICU 2.0
  80      */
  81     virtual ~Replaceable();
  82
  83     /**
  84      * Returns the number of 16-bit code units in the text.
  85      * @return number of 16-bit code units in text
  86      * @stable ICU 1.8
  87      */
  88     inline int32_t length() const;
  89
  90     /**
  91      * Returns the 16-bit code unit at the given offset into the text.
  92      * @param offset an integer between 0 and <code>length()</code>-1
  93      * inclusive
  94      * @return 16-bit code unit of text at given offset
  95      * @stable ICU 1.8
  96      */
  97     inline char16_t charAt(int32_t offset) const;
  98
  99     /**
 100      * Returns the 32-bit code point at the given 16-bit offset into
 101      * the text.  This assumes the text is stored as 16-bit code units
 102      * with surrogate pairs intermixed.  If the offset of a leading or
 103      * trailing code unit of a surrogate pair is given, return the
 104      * code point of the surrogate pair.
 105      *
 106      * @param offset an integer between 0 and <code>length()</code>-1
 107      * inclusive
 108      * @return 32-bit code point of text at given offset
 109      * @stable ICU 1.8
 110      */
 111     inline UChar32 char32At(int32_t offset) const;
 112
 113     /**
 114      * Copies characters in the range [<tt>start</tt>, <tt>limit</tt>)
 115      * into the UnicodeString <tt>target</tt>.
 116      * @param start offset of first character which will be copied
 117      * @param limit offset immediately following the last character to
 118      * be copied
 119      * @param target UnicodeString into which to copy characters.
 120      * @return A reference to <TT>target</TT>
 121      * @stable ICU 2.1
 122      */
 123     virtual void extractBetween(int32_t start,
 124                                 int32_t limit,
 125                                 UnicodeString& target) const = 0;
 126
 127     /**
 128      * Replaces a substring of this object with the given text.  If the
 129      * characters being replaced have metadata, the new characters
 130      * that replace them should be given the same metadata.
 131      *
 132      * <p>Subclasses must ensure that if the text between start and
 133      * limit is equal to the replacement text, that replace has no
 134      * effect. That is, any metadata
 135      * should be unaffected. In addition, subclasses are encouraged to
 136      * check for initial and trailing identical characters, and make a
 137      * smaller replacement if possible. This will preserve as much
 138      * metadata as possible.
 139      * @param start the beginning index, inclusive; <code>0 <= start
 140      * <= limit</code>.
 141      * @param limit the ending index, exclusive; <code>start <= limit
 142      * <= length()</code>.
 143      * @param text the text to replace characters <code>start</code>
 144      * to <code>limit - 1</code>
 145      * @stable ICU 2.0
 146      */
 147     virtual void handleReplaceBetween(int32_t start,
 148                                       int32_t limit,
 149                                       const UnicodeString& text) = 0;
 150     // Note: All other methods in this class take the names of
 151     // existing UnicodeString methods.  This method is the exception.
 152     // It is named differently because all replace methods of
 153     // UnicodeString return a UnicodeString&.  The 'between' is
 154     // required in order to conform to the UnicodeString naming
 155     // convention; API taking start/length are named <operation>, and
 156     // those taking start/limit are named <operationBetween>.  The
 157     // 'handle' is added because 'replaceBetween' and
 158     // 'doReplaceBetween' are already taken.
 159
 160     /**
 161      * Copies a substring of this object, retaining metadata.
 162      * This method is used to duplicate or reorder substrings.
 163      * The destination index must not overlap the source range.
 164      *
 165      * @param start the beginning index, inclusive; <code>0 <= start <=
 166      * limit</code>.
 167      * @param limit the ending index, exclusive; <code>start <= limit <=
 168      * length()</code>.
 169      * @param dest the destination index.  The characters from
 170      * <code>start..limit-1</code> will be copied to <code>dest</code>.
 171      * Implementations of this method may assume that <code>dest <= start ||
 172      * dest >= limit</code>.
 173      * @stable ICU 2.0
 174      */
 175     virtual void copy(int32_t start, int32_t limit, int32_t dest) = 0;
 176
 177     /**
 178      * Returns true if this object contains metadata.  If a
 179      * Replaceable object has metadata, calls to the Replaceable API
 180      * must be made so as to preserve metadata.  If it does not, calls
 181      * to the Replaceable API may be optimized to improve performance.
 182      * The default implementation returns true.
 183      * @return true if this object contains metadata
 184      * @stable ICU 2.2
 185      */
 186     virtual UBool hasMetaData() const;
 187
 188     /**
 189      * Clone this object, an instance of a subclass of Replaceable.
 190      * Clones can be used concurrently in multiple threads.
 191      * If a subclass does not implement clone(), or if an error occurs,
 192      * then NULL is returned.
 193      * The clone functions in all subclasses return a pointer to a Replaceable
 194      * because some compilers do not support covariant (same-as-this)
 195      * return types; cast to the appropriate subclass if necessary.
 196      * The caller must delete the clone.
 197      *
 198      * @return a clone of this object
 199      *
 200      * @see getDynamicClassID
 201      * @stable ICU 2.6
 202      */
 203     virtual Replaceable *clone() const;
 204
 205 protected:
 206
 207     /**
 208      * Default constructor.
 209      * @stable ICU 2.4
 210      */
 211     inline Replaceable();
 212
 213     /*
 214      * Assignment operator not declared. The compiler will provide one
 215      * which does nothing since this class does not contain any data members.
 216      * API/code coverage may show the assignment operator as present and
 217      * untested - ignore.
 218      * Subclasses need this assignment operator if they use compiler-provided
 219      * assignment operators of their own. An alternative to not declaring one
 220      * here would be to declare and empty-implement a protected or public one.
 221     Replaceable &Replaceable::operator=(const Replaceable &);
 222      */
 223
 224     /**
 225      * Virtual version of length().
 226      * @stable ICU 2.4
 227      */
 228     virtual int32_t getLength() const = 0;
 229
 230     /**
 231      * Virtual version of charAt().
 232      * @stable ICU 2.4
 233      */
 234     virtual char16_t getCharAt(int32_t offset) const = 0;
 235
 236     /**
 237      * Virtual version of char32At().
 238      * @stable ICU 2.4
 239      */
 240     virtual UChar32 getChar32At(int32_t offset) const = 0;
 241 };
 242
 243 inline Replaceable::Replaceable() {}
 244
 245 inline int32_t
 246 Replaceable::length() const {
 247     return getLength();
 248 }
 249
 250 inline char16_t
 251 Replaceable::charAt(int32_t offset) const {
 252     return getCharAt(offset);
 253 }
 254
 255 inline UChar32
 256 Replaceable::char32At(int32_t offset) const {
 257     return getChar32At(offset);
 258 }
 259
 260 // There is no rep.cpp, see unistr.cpp for Replaceable function implementations.
 261
 262 U_NAMESPACE_END
 263 #endif // U_SHOW_CPLUSPLUS_API
 264
 265 #endif