org.zkoss.poi.ss.formula.ptg

Class Ptg

java.lang.Object

org.zkoss.poi.ss.formula.ptg.Ptg

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

ArrayPtg, ControlPtg, OperandPtg, OperationPtg, ScalarConstantPtg, UnknownPtg

public abstract class Ptg
extends java.lang.Object
implements java.io.Serializable

Ptg represents a syntactic token in a formula. 'PTG' is an acronym for 'parse thing'. Originally, the name referred to the single byte identifier at the start of the token, but in POI, Ptg encapsulates the whole formula token (initial byte + value data).

Ptgs are logically arranged in a tree representing the structure of the parsed formula. However, in BIFF files Ptgs are written/read in Reverse-Polish Notation order. The RPN ordering also simplifies formula evaluation logic, so POI mostly accesses Ptgs in the same way.

Author:

andy, avik, Jason Height (jheight at chariot dot net dot au)

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static byte	CLASS_ARRAY
static byte	CLASS_REF
static byte	CLASS_VALUE
static Ptg[]	EMPTY_PTG_ARRAY

Constructor Summary

Constructors

Constructor and Description
Ptg()

Method Summary

Modifier and Type	Method and Description
static Ptg	createPtg(LittleEndianInput in)
static boolean	doesFormulaReferToDeletedCell(Ptg[] ptgs)
abstract byte	getDefaultOperandClass()
static int	getEncodedSize(Ptg[] ptgs) This method will return the same result as getEncodedSizeWithoutArrayData(Ptg[]) if there are no array tokens present.
static int	getEncodedSizeWithoutArrayData(Ptg[] ptgs) Used to calculate value that should be encoded at the start of the encoded Ptg token array;
byte	getPtgClass()
char	getRVAType() Debug / diagnostic method to get this token's 'operand class' type.
abstract int	getSize()
abstract boolean	isBaseToken()
static Ptg[]	readTokens(int size, LittleEndianInput in) Reads size bytes of the input stream, to create an array of Ptgs.
static int	serializePtgs(Ptg[] ptgs, byte[] array, int offset) Writes the ptgs to the data buffer, starting at the specified offset.
void	setClass(byte thePtgClass)
abstract java.lang.String	toFormulaString() return a string representation of this token alone
java.lang.String	toString() Overridden toString method to ensure object hash is not printed.
abstract void	write(LittleEndianOutput out)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

EMPTY_PTG_ARRAY

public static final Ptg[] EMPTY_PTG_ARRAY

CLASS_REF

public static final byte CLASS_REF

See Also:

Constant Field Values

CLASS_VALUE

public static final byte CLASS_VALUE

See Also:

Constant Field Values

CLASS_ARRAY

public static final byte CLASS_ARRAY

See Also:

Constant Field Values

Constructor Detail

Ptg

public Ptg()

Method Detail

readTokens

public static Ptg[] readTokens(int size,
                               LittleEndianInput in)

Reads size bytes of the input stream, to create an array of Ptgs. Extra data (beyond size) may be read if and ArrayPtgs are present.

createPtg

public static Ptg createPtg(LittleEndianInput in)

getEncodedSize

public static int getEncodedSize(Ptg[] ptgs)

This method will return the same result as getEncodedSizeWithoutArrayData(Ptg[]) if there are no array tokens present.

Returns:

the full size taken to encode the specified Ptgs

getEncodedSizeWithoutArrayData

public static int getEncodedSizeWithoutArrayData(Ptg[] ptgs)

Used to calculate value that should be encoded at the start of the encoded Ptg token array;

Returns:

the size of the encoded Ptg tokens not including any trailing array data.

serializePtgs

public static int serializePtgs(Ptg[] ptgs,
                                byte[] array,
                                int offset)

Writes the ptgs to the data buffer, starting at the specified offset.
The 2 byte encode length field is not written by this method.

Returns:

number of bytes written

getSize

public abstract int getSize()

Returns:

the encoded length of this Ptg, including the initial Ptg type identifier byte.

write

public abstract void write(LittleEndianOutput out)

toFormulaString

public abstract java.lang.String toFormulaString()

return a string representation of this token alone

toString

public java.lang.String toString()

Overridden toString method to ensure object hash is not printed. This helps get rid of gratuitous diffs when comparing two dumps Subclasses may output more relevant information by overriding this method

Overrides:

toString in class java.lang.Object

setClass

public final void setClass(byte thePtgClass)

getPtgClass

public final byte getPtgClass()

Returns:

the 'operand class' (REF/VALUE/ARRAY) for this Ptg

getRVAType

public final char getRVAType()

Debug / diagnostic method to get this token's 'operand class' type.

Returns:

'R' for 'reference', 'V' for 'value', 'A' for 'array' and '.' for base tokens

getDefaultOperandClass

public abstract byte getDefaultOperandClass()

isBaseToken

public abstract boolean isBaseToken()

Returns:

false if this token is classified as 'reference', 'value', or 'array'

doesFormulaReferToDeletedCell

public static boolean doesFormulaReferToDeletedCell(Ptg[] ptgs)