icuSources/common/unicode/symtable.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) 2000-2005, International Business Machines
   6 *   Corporation and others.  All Rights Reserved.
   7 **********************************************************************
   8 *   Date        Name        Description
   9 *   02/04/00    aliu        Creation.
  10 **********************************************************************
  11 */
  12 #ifndef SYMTABLE_H
  13 #define SYMTABLE_H
  14
  15 #include "unicode/utypes.h"
  16 #include "unicode/uobject.h"
  17
  18 /**
  19  * \file
  20  * \brief C++ API: An interface that defines both lookup protocol and parsing of
  21  * symbolic names.
  22  */
  23
  24 #if U_SHOW_CPLUSPLUS_API
  25 U_NAMESPACE_BEGIN
  26
  27 class ParsePosition;
  28 class UnicodeFunctor;
  29 class UnicodeSet;
  30 class UnicodeString;
  31
  32 /**
  33  * An interface that defines both lookup protocol and parsing of
  34  * symbolic names.
  35  *
  36  * <p>A symbol table maintains two kinds of mappings.  The first is
  37  * between symbolic names and their values.  For example, if the
  38  * variable with the name "start" is set to the value "alpha"
  39  * (perhaps, though not necessarily, through an expression such as
  40  * "$start=alpha"), then the call lookup("start") will return the
  41  * char[] array ['a', 'l', 'p', 'h', 'a'].
  42  *
  43  * <p>The second kind of mapping is between character values and
  44  * UnicodeMatcher objects.  This is used by RuleBasedTransliterator,
  45  * which uses characters in the private use area to represent objects
  46  * such as UnicodeSets.  If U+E015 is mapped to the UnicodeSet [a-z],
  47  * then lookupMatcher(0xE015) will return the UnicodeSet [a-z].
  48  *
  49  * <p>Finally, a symbol table defines parsing behavior for symbolic
  50  * names.  All symbolic names start with the SYMBOL_REF character.
  51  * When a parser encounters this character, it calls parseReference()
  52  * with the position immediately following the SYMBOL_REF.  The symbol
  53  * table parses the name, if there is one, and returns it.
  54  *
  55  * @stable ICU 2.8
  56  */
  57 class U_COMMON_API SymbolTable /* not : public UObject because this is an interface/mixin class */ {
  58 public:
  59
  60     /**
  61      * The character preceding a symbol reference name.
  62      * @stable ICU 2.8
  63      */
  64     enum { SYMBOL_REF = 0x0024 /*$*/ };
  65
  66     /**
  67      * Destructor.
  68      * @stable ICU 2.8
  69      */
  70     virtual ~SymbolTable();
  71
  72     /**
  73      * Lookup the characters associated with this string and return it.
  74      * Return <tt>NULL</tt> if no such name exists.  The resultant
  75      * string may have length zero.
  76      * @param s the symbolic name to lookup
  77      * @return a string containing the name's value, or <tt>NULL</tt> if
  78      * there is no mapping for s.
  79      * @stable ICU 2.8
  80      */
  81     virtual const UnicodeString* lookup(const UnicodeString& s) const = 0;
  82
  83     /**
  84      * Lookup the UnicodeMatcher associated with the given character, and
  85      * return it.  Return <tt>NULL</tt> if not found.
  86      * @param ch a 32-bit code point from 0 to 0x10FFFF inclusive.
  87      * @return the UnicodeMatcher object represented by the given
  88      * character, or NULL if there is no mapping for ch.
  89      * @stable ICU 2.8
  90      */
  91     virtual const UnicodeFunctor* lookupMatcher(UChar32 ch) const = 0;
  92
  93     /**
  94      * Parse a symbol reference name from the given string, starting
  95      * at the given position.  If no valid symbol reference name is
  96      * found, return the empty string and leave pos unchanged.  That is, if the
  97      * character at pos cannot start a name, or if pos is at or after
  98      * text.length(), then return an empty string.  This indicates an
  99      * isolated SYMBOL_REF character.
 100      * @param text the text to parse for the name
 101      * @param pos on entry, the index of the first character to parse.
 102      * This is the character following the SYMBOL_REF character.  On
 103      * exit, the index after the last parsed character.  If the parse
 104      * failed, pos is unchanged on exit.
 105      * @param limit the index after the last character to be parsed.
 106      * @return the parsed name, or an empty string if there is no
 107      * valid symbolic name at the given position.
 108      * @stable ICU 2.8
 109      */
 110     virtual UnicodeString parseReference(const UnicodeString& text,
 111                                          ParsePosition& pos, int32_t limit) const = 0;
 112 };
 113 U_NAMESPACE_END
 114 #endif // U_SHOW_CPLUSPLUS_API
 115
 116 #endif