ustring.h

khtml在gtk上的移植版本

     */
    UString(const UChar *c, int length);
    /**
     * If copy is false the string data will be adopted.
     * That means that the data will NOT be copied and the pointer will
     * be deleted when the UString object is modified or destroyed.
     * Behaviour defaults to a deep copy if copy is true.
     */
    UString(UChar *c, int length, bool copy);
    /**
     * Copy constructor. Makes a shallow copy only.
     */
    UString(const UString &s) { attach(s.rep); }
    /**
     * Convenience declaration only ! You'll be on your own to write the
     * implementation for a construction from QString.
     *
     * Note: feel free to contact me if you want to see a dummy header for
     * your favourite FooString class here !
     */
    UString(const QString &);
    /**
     * Convenience declaration only ! See @ref UString(const QString&).
     */
    UString(const DOM::DOMString &);
    /**
     * Concatenation constructor. Makes operator+ more efficient.
     */
    UString(const UString &, const UString &);
    /**
     * Destructor. If this handle was the only one holding a reference to the
     * string the data will be freed.
     */
    ~UString() { release(); }

    /**
     * Constructs a string from an int.
     */
    static UString from(int i);
    /**
     * Constructs a string from an unsigned int.
     */
    static UString from(unsigned int u);
    /**
     * Constructs a string from a long.
     */
    static UString from(long l);
    /**
     * Constructs a string from a double.
     */
    static UString from(double d);

    /**
     * Append another string.
     */
    UString &append(const UString &);
    UString &append(const char *);
    UString &append(unsigned short);
    UString &append(char c) { return append(static_cast<unsigned short>(static_cast<unsigned char>(c))); }
    UString &append(UChar c) { return append(c.uc); }

    /**
     * @return The string converted to the 8-bit string type @ref CString().
     */
    CString cstring() const;
    /**
     * Convert the Unicode string to plain ASCII chars chopping of any higher
     * bytes. This method should only be used for *debugging* purposes as it
     * is neither Unicode safe nor free from side effects. In order not to
     * waste any memory the char buffer is static and *shared* by all UString
     * instances.
     */
    char *ascii() const;

    /**
     * Convert the string to UTF-8, assuming it is UTF-16 encoded.
     * Since this function is tolerant of badly formed UTF-16, it can create UTF-8
     * strings that are invalid because they have characters in the range
     * U+D800-U+DDFF, U+FFFE, or U+FFFF, but the UTF-8 string is guaranteed to
     * be otherwise valid.
     */
    CString UTF8String() const;

    /**
     * @see UString(const QString&).
     */
    DOM::DOMString string() const;
    /**
     * @see UString(const QString&).
     */
    QString qstring() const;
    /**
     * @see UString(const QString&).
     */
    QConstString qconststring() const;

    /**
     * Assignment operator.
     */
    UString &operator=(const char *c);
    UString &operator=(const UString &);
    /**
     * Appends the specified string.
     */
    UString &operator+=(const UString &s) { return append(s); }
    UString &operator+=(const char *s) { return append(s); }

    /**
     * @return A pointer to the internal Unicode data.
     */
    const UChar* data() const { return rep->data(); }
    /**
     * @return True if null.
     */
    bool isNull() const { return (rep == &Rep::null); }
    /**
     * @return True if null or zero length.
     */
    bool isEmpty() const { return (!rep->len); }
    /**
     * Use this if you want to make sure that this string is a plain ASCII
     * string. For example, if you don't want to lose any information when
     * using @ref cstring() or @ref ascii().
     *
     * @return True if the string doesn't contain any non-ASCII characters.
     */
    bool is8Bit() const;
    /**
     * @return The length of the string.
     */
    int size() const { return rep->size(); }
    /**
     * Const character at specified position.
     */
    UChar operator[](int pos) const;
    /**
     * Writable reference to character at specified position.
     */
    UCharReference operator[](int pos);

    /**
     * Attempts an conversion to a number. Apart from floating point numbers,
     * the algorithm will recognize hexadecimal representations (as
     * indicated by a 0x or 0X prefix) and +/- Infinity.
     * Returns NaN if the conversion failed.
     * @param tolerateTrailingJunk if true, toDouble can tolerate garbage after the number.
     * @param tolerateEmptyString if false, toDouble will turn an empty string into NaN rather than 0.
     */
    double toDouble(bool tolerateTrailingJunk, bool tolerateEmptyString) const;
    double toDouble(bool tolerateTrailingJunk) const;
    double toDouble() const;
    /**
     * Attempts an conversion to an unsigned long integer. ok will be set
     * according to the success.
     * @param tolerateEmptyString if false, toULong will return false for *ok for an empty string.
     */
    unsigned long toULong(bool *ok, bool tolerateEmptyString) const;
    unsigned long toULong(bool *ok = 0) const;

    uint32_t toUInt32(bool *ok = 0) const;
    uint32_t toStrictUInt32(bool *ok = 0) const;

    /**
     * Attempts an conversion to an array index. The "ok" boolean will be set
     * to true if it is a valid array index according to the rule from
     * ECMA 15.2 about what an array index is. It must exactly match the string
     * form of an unsigned integer, and be less than 2^32 - 1.
     */
    unsigned toArrayIndex(bool *ok = 0) const;

    /**
     * @return Position of first occurrence of f starting at position pos.
     * -1 if the search was not successful.
     */
    int find(const UString &f, int pos = 0) const;
    int find(UChar, int pos = 0) const;
    /**
     * @return Position of first occurrence of f searching backwards from
     * position pos.
     * -1 if the search was not successful.
     */
    int rfind(const UString &f, int pos) const;
    int rfind(UChar, int pos) const;
    /**
     * @return The sub string starting at position pos and length len.
     */
    UString substr(int pos = 0, int len = -1) const;
    /**
     * Static instance of a null string.
     */
    static const UString &null();
#ifdef KJS_DEBUG_MEM
    /**
     * Clear statically allocated resources.
     */
    static void globalClear();
#endif
  private:
    UString(Rep *r) { attach(r); }
    void attach(Rep *r);
    void detach();
    void release();
    int expandedSize(int size, int otherSize) const;
    int usedCapacity() const;
    int usedPreCapacity() const;
    void expandCapacity(int requiredLength);
    void expandPreCapacity(int requiredPreCap);

    Rep *rep;
  };

  inline bool operator==(const UChar &c1, const UChar &c2) {
    return (c1.uc == c2.uc);
  }
  bool operator==(const UString& s1, const UString& s2);
  inline bool operator!=(const UString& s1, const UString& s2) {
    return !KJS::operator==(s1, s2);
  }
  bool operator<(const UString& s1, const UString& s2);
  bool operator==(const UString& s1, const char *s2);
  inline bool operator!=(const UString& s1, const char *s2) {
    return !KJS::operator==(s1, s2);
  }
  inline bool operator==(const char *s1, const UString& s2) {
    return operator==(s2, s1);
  }
  inline bool operator!=(const char *s1, const UString& s2) {
    return !KJS::operator==(s1, s2);
  }
  bool operator==(const CString& s1, const CString& s2);
  inline UString operator+(const UString& s1, const UString& s2) {
    return UString(s1, s2);
  }
  
  int compare(const UString &, const UString &);

  // Given a first byte, gives the length of the UTF-8 sequence it begins.
  // Returns 0 for bytes that are not legal starts of UTF-8 sequences.
  // Only allows sequences of up to 4 bytes, since that works for all Unicode characters (U-00000000 to U-0010FFFF).
  int UTF8SequenceLength(char);

  // Takes a null-terminated C-style string with a UTF-8 sequence in it and converts it to a character.
  // Only allows Unicode characters (U-00000000 to U-0010FFFF).
  // Returns -1 if the sequence is not valid (including presence of extra bytes).
  int decodeUTF8Sequence(const char *);

  // Given a UTF-8 string, converts offsets from the UTF-16 form of the string into offsets into the UTF-8 string.
  // Note: This function can overrun the buffer if the string contains a partial UTF-8 sequence, so it should
  // not be called with strings that might contain such sequences.
  void convertUTF16OffsetsToUTF8Offsets(const char *UTF8String, int *offsets, int numOffsets);

  // Given a UTF-8 string, converts offsets from the UTF-8 string into offsets into the UTF-16 form of the string.
  // Note: This function can overrun the buffer if the string contains a partial UTF-8 sequence, so it should
  // not be called with strings that might contain such sequences.
  void convertUTF8OffsetsToUTF16Offsets(const char *UTF8String, int *offsets, int numOffsets);

}; // namespace

#endif