uk.ac.bristol.star.cdf

Class Variable

java.lang.Object

uk.ac.bristol.star.cdf.Variable

public class Variable
extends java.lang.Object

Provides the metadata and record data for a CDF variable.

At construction time, a map of where the records are stored is constructed, but the record data itself is not read unless or until one of the read methods is called.

This interface does not currently support data reading in such a flexible way as the official CDF interface. You can read a record's worth of data at a time using either readRawRecord (which should be fairly efficient) or readShapedRecord (which may have to copy and possibly re-order the array, and may not be so efficient).

Since:

20 Jun 2013

Constructor Summary

Constructors

Constructor and Description
Variable(VariableDescriptorRecord vdr, CdfInfo cdfInfo, RecordFactory recFact) Constructor.

Method Summary

Modifier and Type	Method and Description
java.lang.Object	createRawValueArray() Creates a workspace array suitable for use with this variable's reading methods.
DataType	getDataType() Returns the data type of this variable.
VariableDescriptorRecord	getDescriptor() Returns the VariableDescriptorRecord on which this Variable instance is based.
java.lang.String	getName() Returns this variable's name.
int	getNum() Returns the index number within the CDF of this variable.
int	getRecordCount() Returns the upper limit of records that may have values.
boolean	getRecordVariance() Indicates whether this variable has a value which is fixed for all records or can vary per record.
Shaper	getShaper() Returns an object that knows about the array dimensions of the data values.
java.lang.String	getSummary() Returns a short text string describing the type, shape and variance of this variable.
boolean	hasRecord(int irec) Indicates whether a real distinct file-based record exists for the given index.
boolean	isZVariable() Indicates whether this variable is a zVariable or rVariable.
void	readRawRecord(int irec, java.lang.Object rawValueArray) Reads the data from a single record into a supplied raw value array.
java.lang.Object	readShapedRecord(int irec, boolean rowMajor, java.lang.Object rawValueArrayWorkspace) Reads the data from a single record and returns it as an object of a suitable type for this variable.

Methods inherited from class java.lang.Object

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

Constructor Detail

Variable

public Variable(VariableDescriptorRecord vdr,
                CdfInfo cdfInfo,
                RecordFactory recFact)
         throws java.io.IOException

Constructor.

Parameters:

vdr - variable descriptor record for the variable

cdfInfo - global CDF information

recFact - record factory

Throws:

java.io.IOException

Method Detail

getName

public java.lang.String getName()

Returns this variable's name.

Returns:

variable name

getNum

public int getNum()

Returns the index number within the CDF of this variable.

Returns:

variable num

isZVariable

public boolean isZVariable()

Indicates whether this variable is a zVariable or rVariable.

Returns:

true for zVariable, false for rVariable

getRecordCount

public int getRecordCount()

Returns the upper limit of records that may have values. The actual number of records may be lower than this in case of sparsity.

Returns:

maximum record count

getDataType

public DataType getDataType()

Returns the data type of this variable.

Returns:

data type

getShaper

public Shaper getShaper()

Returns an object that knows about the array dimensions of the data values.

Returns:

shaper

getRecordVariance

public boolean getRecordVariance()

Indicates whether this variable has a value which is fixed for all records or can vary per record.

Returns:

false for fixed, true for varying

getSummary

public java.lang.String getSummary()

Returns a short text string describing the type, shape and variance of this variable.

Returns:

text summary of variable characteristics

getDescriptor

public VariableDescriptorRecord getDescriptor()

Returns the VariableDescriptorRecord on which this Variable instance is based.

Returns:

variable descriptor record (rVDR or zVDR)

createRawValueArray

public java.lang.Object createRawValueArray()

Creates a workspace array suitable for use with this variable's reading methods. The returned array is a 1-dimensional array of a primitive type or of String.

Returns:

workspace array for data reading

hasRecord

public boolean hasRecord(int irec)
                  throws java.io.IOException

Indicates whether a real distinct file-based record exists for the given index. Reading a record will give you a result in any case, but if this returns false it will be some kind of fixed or default value.

Parameters:

irec - record index

Returns:

true iff a file-based record exists for irec

Throws:

java.io.IOException

readRawRecord

public void readRawRecord(int irec,
                          java.lang.Object rawValueArray)
                   throws java.io.IOException

Reads the data from a single record into a supplied raw value array. The values are read into the supplied array in the order in which they are stored in the data stream, that is depending on the row/column majority of the CDF.

The raw value array is as obtained from createRawValueArray().

Parameters:

irec - record index

rawValueArray - workspace array, as created by the createRawValueArray method

Throws:

java.io.IOException

readShapedRecord

public java.lang.Object readShapedRecord(int irec,
                                         boolean rowMajor,
                                         java.lang.Object rawValueArrayWorkspace)
                                  throws java.io.IOException

Reads the data from a single record and returns it as an object of a suitable type for this variable. If the variable type a scalar, then the return value will be one of the primitive wrapper types (Integer etc), otherwise it will be an array of primitive or String values. If the majority of the stored data does not match the rowMajor argument, the array elements will be rordered appropriately. If some of the dimension variances are false, the values will be duplicated accordingly. The Shaper returned from the getShaper() method can provide more information on the return value from this method.

The workspace is as obtained from createRawValueArray().

Parameters:

irec - record index

rowMajor - required majority of output array; true for row major, false for column major; only has an effect for dimensionality >=2

rawValueArrayWorkspace - workspace array, as created by the createRawValueArray method

Returns:

a new object containing the shaped result (not the same object as rawValueArray

Throws:

java.io.IOException