icuSources/common/unicode/symtable.h

   1 /*
   2 **********************************************************************
   3 *   Copyright (c) 2000-2004, International Business Machines
   4 *   Corporation and others.  All Rights Reserved.
   5 **********************************************************************
   6 *   Date        Name        Description
   7 *   02/04/00    aliu        Creation.
   8 **********************************************************************
   9 */
  10 #ifndef SYMTABLE_H
  11 #define SYMTABLE_H
  12
  13 #include "unicode/utypes.h"
  14 #include "unicode/uobject.h"
  15
  16 U_NAMESPACE_BEGIN
  17
  18 class ParsePosition;
  19 class UnicodeFunctor;
  20 class UnicodeSet;
  21 class UnicodeString;
  22
  23 /**
  24  * An interface that defines both lookup protocol and parsing of
  25  * symbolic names.
  26  *
  27  * <p>A symbol table maintains two kinds of mappings.  The first is
  28  * between symbolic names and their values.  For example, if the
  29  * variable with the name "start" is set to the value "alpha"
  30  * (perhaps, though not necessarily, through an expression such as
  31  * "$start=alpha"), then the call lookup("start") will return the
  32  * char[] array ['a', 'l', 'p', 'h', 'a'].
  33  *
  34  * <p>The second kind of mapping is between character values and
  35  * UnicodeMatcher objects.  This is used by RuleBasedTransliterator,
  36  * which uses characters in the private use area to represent objects
  37  * such as UnicodeSets.  If U+E015 is mapped to the UnicodeSet [a-z],
  38  * then lookupMatcher(0xE015) will return the UnicodeSet [a-z].
  39  *
  40  * <p>Finally, a symbol table defines parsing behavior for symbolic
  41  * names.  All symbolic names start with the SYMBOL_REF character.
  42  * When a parser encounters this character, it calls parseReference()
  43  * with the position immediately following the SYMBOL_REF.  The symbol
  44  * table parses the name, if there is one, and returns it.
  45  *
  46  * @draft ICU 2.8
  47  */
  48 class U_COMMON_API SymbolTable /* not : public UObject because this is an interface/mixin class */ {
  49 public:
  50
  51     /**
  52      * The character preceding a symbol reference name.
  53      * @draft ICU 2.8
  54      */
  55     enum { SYMBOL_REF = 0x0024 /*$*/ };
  56
  57     /**
  58      * Destructor.
  59      * @draft ICU 2.8
  60      */
  61     virtual ~SymbolTable();
  62
  63     /**
  64      * Lookup the characters associated with this string and return it.
  65      * Return <tt>NULL</tt> if no such name exists.  The resultant
  66      * string may have length zero.
  67      * @param s the symbolic name to lookup
  68      * @return a string containing the name's value, or <tt>NULL</tt> if
  69      * there is no mapping for s.
  70      * @draft ICU 2.8
  71      */
  72     virtual const UnicodeString* lookup(const UnicodeString& s) const = 0;
  73
  74     /**
  75      * Lookup the UnicodeMatcher associated with the given character, and
  76      * return it.  Return <tt>NULL</tt> if not found.
  77      * @param ch a 32-bit code point from 0 to 0x10FFFF inclusive.
  78      * @return the UnicodeMatcher object represented by the given
  79      * character, or NULL if there is no mapping for ch.
  80      * @draft ICU 2.8
  81      */
  82     virtual const UnicodeFunctor* lookupMatcher(UChar32 ch) const = 0;
  83
  84     /**
  85      * Parse a symbol reference name from the given string, starting
  86      * at the given position.  If no valid symbol reference name is
  87      * found, return the empty string and leave pos unchanged.  That is, if the
  88      * character at pos cannot start a name, or if pos is at or after
  89      * text.length(), then return an empty string.  This indicates an
  90      * isolated SYMBOL_REF character.
  91      * @param text the text to parse for the name
  92      * @param pos on entry, the index of the first character to parse.
  93      * This is the character following the SYMBOL_REF character.  On
  94      * exit, the index after the last parsed character.  If the parse
  95      * failed, pos is unchanged on exit.
  96      * @param limit the index after the last character to be parsed.
  97      * @return the parsed name, or an empty string if there is no
  98      * valid symbolic name at the given position.
  99      * @draft ICU 2.8
 100      */
 101     virtual UnicodeString parseReference(const UnicodeString& text,
 102                                          ParsePosition& pos, int32_t limit) const = 0;
 103 };
 104 U_NAMESPACE_END
 105
 106 #endif