ncsa.hdf.object.h5
Class H5CompoundDS

java.lang.Object
  ncsa.hdf.object.HObject
      ncsa.hdf.object.Dataset
          ncsa.hdf.object.CompoundDS
              ncsa.hdf.object.h5.H5CompoundDS

All Implemented Interfaces:

java.io.Serializable, DataFormat

public class H5CompoundDS
extends CompoundDS

The H5CompoundDS class defines an HDF5 dataset of compound datatypes.

An HDF5 dataset is an object composed of a collection of data elements, or raw data, and metadata that stores a description of the data elements, data layout, and all other information necessary to write, read, and interpret the stored data.

A HDF5 compound datatype is similar to a struct in C or a common block in Fortran: it is a collection of one or more atomic types or small arrays of such types. Each member of a compound type has a name which is unique within that type, and a byte offset that determines the first byte (smallest byte address) of that member in a compound datum.

For more information on HDF5 datasets and datatypes, read the HDF5 User's Guide.

There are two basic types of compound datasets: simple compound data and nested compound data. Members of a simple compound dataset have atomic datatyes. Members of a nested compound dataset are compound or array of compound data.

Since Java does not understand C structures, we cannot directly read/write compound data values as in the following C example.

 typedef struct s1_t {
         int    a;
         float  b;
         double c; 
         } s1_t;
     s1_t       s1[LENGTH];
     ...
     H5Dwrite(..., s1);
     H5Dread(..., s1);
 

Values of compound data fields are stored in java.util.Vector object. We read and write compound data by fields instead of compound structure. As for the example above, the java.util.Vector object has three elements: int[LENGTH], float[LENGTH] and double[LENGTH]. Since Java understands the primitive datatypes of int, float and double, we will be able to read/write the compound data by field.

Version:

1.1 9/4/2007

Author:

Peter X. Cao

See Also:

Serialized Form

Field Summary

static long

serialVersionUID

Fields inherited from class ncsa.hdf.object.CompoundDS

separator

Constructor Summary

H5CompoundDS(FileFormat theFile, java.lang.String name, java.lang.String path)
Constructs an HDF5 compound dataset with given file, dataset name and path.

H5CompoundDS(FileFormat theFile, java.lang.String name, java.lang.String path, long[] oid)
Deprecated. Not for public use in the future.
Using H5CompoundDS(FileFormat, String, String)

Method Summary

void	clear() Clears memory held by the dataset, such as data buffer.
void	close(int did) Closes access to the object.
static Dataset	create(java.lang.String name, Group pgroup, long[] dims, long[] maxdims, long[] chunks, int gzip, java.lang.String[] memberNames, Datatype[] memberDatatypes, int[] memberRanks, long[][] memberDims, java.lang.Object data) Creates a simple compound dataset in a file with/without chunking and compression
static Dataset	create(java.lang.String name, Group pgroup, long[] dims, java.lang.String[] memberNames, Datatype[] memberDatatypes, int[] memberRanks, long[][] memberDims, java.lang.Object data) Deprecated. Not for public use in the future. Using #create(String, Group, long[], long[], long[], int, String[], Datatype[], int[], int[][], Object)
static Dataset	create(java.lang.String name, Group pgroup, long[] dims, java.lang.String[] memberNames, Datatype[] memberDatatypes, int[] memberSizes, java.lang.Object data) Deprecated. Not for public use in the future. Using #create(String, Group, long[], long[], long[], int, String[], Datatype[], int[], int[][], Object)
Datatype	getDatatype() Returns the datatype object of the dataset.
java.util.List	getMetadata() Retrieves the metadata such as attributes from file.
java.util.List	getMetadata(int... attrPropList)
int	getSize(int tid) Returns the size in bytes of a given datatype.
boolean	hasAttribute() Check if the object has any attributes attached.
void	init() Retrieves datatype and dataspace information from file and sets the dataset in memory.
boolean	isString(int tid) Checks if a given datatype is a string.
int	open() Opens an existing object such as dataset or group for access.
java.lang.Object	read() Reads the data from file.
byte[]	readBytes() Reads the raw data of the dataset from file to a byte array.
void	removeMetadata(java.lang.Object info) Deletes an existing metadata from this data object.
void	setName(java.lang.String newName) Sets the name of the object.
void	write(java.lang.Object buf) Writes the given data buffer into this dataset in a file.
void	writeMetadata(java.lang.Object info) Writes a specific metadata (such as attribute) into file.

Methods inherited from class ncsa.hdf.object.CompoundDS

copy, getMemberCount, getMemberNames, getMemberOrders, getMemberTypes, getMemeberDims, getSelectedMemberCount, getSelectedMemberOrders, getSelectedMemberTypes, isMemberSelected, selectMember, setMemberSelection

Methods inherited from class ncsa.hdf.object.Dataset

byteToString, clearData, convertFromUnsignedC, convertFromUnsignedC, convertToUnsignedC, convertToUnsignedC, getChunkSize, getCompression, getConvertByteToString, getData, getDimNames, getDims, getHeight, getMaxDims, getRank, getSelectedDims, getSelectedIndex, getStartDims, getStride, getWidth, isEnumConverted, setConvertByteToString, setData, setEnumConverted, stringToByte, write

Methods inherited from class ncsa.hdf.object.HObject

equalsOID, getFID, getFile, getFileFormat, getFullName, getLinkTargetObjName, getName, getOID, getPath, setLinkTargetObjName, setPath, toString

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

serialVersionUID

public static final long serialVersionUID

See Also:

HObject.serialVersionUID, Constant Field Values

Constructor Detail

H5CompoundDS

public H5CompoundDS(FileFormat theFile,
                    java.lang.String name,
                    java.lang.String path)

Constructs an HDF5 compound dataset with given file, dataset name and path.

The dataset object represents an existing dataset in the file. For example, new H5CompoundDS(file, "dset1", "/g0/") constructs a dataset object that corresponds to the dataset,"dset1", at group "/g0/".

This object is usually constructed at FileFormat.open(), which loads the file structure and object informatoin into tree structure (TreeNode). It is rarely used elsewhere.

Parameters:

theFile - the file that contains the dataset.

name - the name of the CompoundDS, e.g. "compDS".

path - the path of the CompoundDS, e.g. "/g1".

H5CompoundDS

@Deprecated
public H5CompoundDS(FileFormat theFile,
                               java.lang.String name,
                               java.lang.String path,
                               long[] oid)

Deprecated. Not for public use in the future.
Using H5CompoundDS(FileFormat, String, String)

Method Detail

hasAttribute

public boolean hasAttribute()

Description copied from interface: DataFormat

Check if the object has any attributes attached.

Returns:

true if it has any attribute(s), false otherwise.

getDatatype

public Datatype getDatatype()

Description copied from class: Dataset

Returns the datatype object of the dataset.

Specified by:

getDatatype in class Dataset

Returns:

the datatype object of the dataset.

clear

public void clear()

Description copied from class: Dataset

Clears memory held by the dataset, such as data buffer.

Overrides:

clear in class Dataset

readBytes

public byte[] readBytes()
                 throws HDF5Exception

Description copied from class: Dataset

Reads the raw data of the dataset from file to a byte array.

readBytes() reads raw data to an array of bytes instead of array of its datatype. For example, for an one-dimension 32-bit integer dataset of size 5, the readBytes() returns of a byte array of size 20 instead of an int array of 5.

readBytes() can be used to copy data from one dataset to another efficiently because the raw data is not converted to its native type, it saves memory space and CPU time.

Specified by:

readBytes in class Dataset

Returns:

the byte array of the raw data.

Throws:

HDF5Exception

read

public java.lang.Object read()
                      throws HDF5Exception

Description copied from class: Dataset

Reads the data from file.

read() reads the data from file to a memory buffer and returns the memory buffer. The dataset object does not hold the memobry buffer. To store the memory buffer in the dataset object, one must call getData().

By default, the whole dataset is read into memory. Users can also select subset to read. Subsetting is done in an implicit way.

How to Select a Subset

A selection is specified by three arrays: start, stride and count.

1. start: offset of a selection
2. stride: determining how many elements to move in each dimension
3. count: number of elements to select in each dimension

getStartDims(), getStartDims() and getSelectedDims() returns the start, stride and count arrays respectively. Applications can make a selection by changing the values of the arrays.

The following example shows how to make a subset. In the example, the dataset is a 4-dimensional array of [200][100][50][10], i.e. dims[0]=200; dims[1]=100; dims[2]=50; dims[3]=10;
We want to select every other data point in dims[1] and dims[2]

 int rank = dataset.getRank(); // number of dimension of the dataset
 long[] dims = dataset.getDims(); // the dimension sizes of the dataset
 long[] selected = dataset.getSelectedDims(); // the selected size of the dataet
 long[] start = dataset.getStartDims(); // the off set of the selection
 long[] stride = dataset.getStride(); // the stride of the dataset
 int[] selectedIndex = dataset.getSelectedIndex(); // the selected dimensions for display
 
 // select dim1 and dim2 as 2D data for display,and slice through dim0
 selectedIndex[0] = 1;
 selectedIndex[1] = 2;
 selectedIndex[1] = 0;
 
 // reset the selection arrays
 for (int i = 0; i < rank; i++) {
     start[i] = 0;
     selected[i] = 1;
     stride[i] = 1;
 }
 
 // set stride to 2 on dim1 and dim2 so that every other data points are selected.
 stride[1] = 2;
 stride[2] = 2;
 
 // set the selection size of dim1 and dim2
 selected[1] = dims[1] / stride[1];
 selected[2] = dims[1] / stride[2];
 
 // when dataset.getData() is called, the slection above will be used since
 // the dimension arrays are passed by reference. Changes of these arrays
 // outside the dataset object directly change the values of these array
 // in the dataset object.
 

For ScalarDS, the memory data buffer is an one-dimensional array of byte, short, int, float, double or String type based on the datatype of the dataset.

For CompoundDS, the meory data object is an java.util.List object. Each element of the list is a data array that corresponds to a compound field.

For example, if compound dataset "comp" has the following nested structure, and memeber datatypes

 comp --> m01 (int)
 comp --> m02 (float)
 comp --> nest1 --> m11 (char)
 comp --> nest1 --> m12 (String)
 comp --> nest1 --> nest2 --> m21 (long)
 comp --> nest1 --> nest2 --> m22 (double)
 

getData() returns a list of six arrays: {int[], float[], char[], Stirng[], long[] and double[]}.

Specified by:

read in class Dataset

Returns:

the data read from file.

Throws:

HDF5Exception

See Also:

#getData()}

write

public void write(java.lang.Object buf)
           throws HDF5Exception

Writes the given data buffer into this dataset in a file.

The data buffer is a vector that contains the data values of compound fields. The data is written into file field by field.

Specified by:

write in class Dataset

Parameters:

buf - The vector that contains the data values of compound fields.

Throws:

HDF5Exception

getMetadata

public java.util.List getMetadata()
                           throws HDF5Exception

Description copied from interface: DataFormat

Retrieves the metadata such as attributes from file.

Metadata such as attributes are stored in a List.

Returns:

the list of metadata objects.

Throws:

HDF5Exception

getMetadata

public java.util.List getMetadata(int... attrPropList)
                           throws HDF5Exception

Throws:

HDF5Exception

writeMetadata

public void writeMetadata(java.lang.Object info)
                   throws java.lang.Exception

Description copied from interface: DataFormat

Writes a specific metadata (such as attribute) into file.

If an HDF(4&5) attribute exists in file, the method updates its value. If the attribute does not exists in file, it creates the attribute in file and attaches it to the object. It will fail to write a new attribute to the object where an attribute with the same name already exists. To update the value of an existing attribute in file, one needs to get the instance of the attribute by getMetadata(), change its values, and use writeMetadata() to write the value.

Parameters:

info - the metadata to write.

Throws:

java.lang.Exception

removeMetadata

public void removeMetadata(java.lang.Object info)
                    throws HDF5Exception

Description copied from interface: DataFormat

Deletes an existing metadata from this data object.

Parameters:

info - the metadata to delete.

Throws:

HDF5Exception

open

public int open()

Description copied from class: HObject

Opens an existing object such as dataset or group for access. The return value is an object identifier obtained by implementing classes such as H5.H5Dopen(). This function is needed to allow other objects to be able to access the object. For instance, H5File class uses the open() function to obtain object identifier for copyAttributes(int src_id, int dst_id) and other purposes. The open() function should be used in pair with close(int) function.

Specified by:

open in class HObject

Returns:

the object identifier if successful; otherwise returns a negative value.

See Also:

HObject.close(int)

close

public void close(int did)

Description copied from class: HObject

Closes access to the object.

Sub-classes must implement this interface because different data objects have their own ways of how the data resources are closed.

For example, H5Group.close() calls the ncsa.hdf.hdf5lib.H5.H5Gclose() method and closes the group resource specified by the group id.

Specified by:

close in class HObject

Parameters:

did - The object identifier.

init

public void init()

Description copied from class: Dataset

Retrieves datatype and dataspace information from file and sets the dataset in memory.

The init() is designed to support lazy operation in dataset object. When a data object is retrieved from file, the datatype, dataspace and raw data are not loaded into memory. When it is asked to read the raw data from file, init() is first called to get the datatype and dataspace information, then load the raw data from file.

init() is also used to reset selection of a dataset (start, stride and count) to the default, which is the entire dataset for 1D or 2D datasets. In the following example, init() at step 1) retrieve datatype and dataspace information from file. getData() at step 3) read only one data point. init() at step 4) reset the selection to the whole dataset. getData() at step 4) reads the values of whole dataset into memory.

 dset = (Dataset) file.get(NAME_DATASET);
 
 // 1) get datatype and dataspace information from file
 dset.init();
 rank = dset.getRank(); // rank = 2, a 2D dataset
 count = dset.getSelectedDims();
 start = dset.getStartDims();
 dims = dset.getDims();
 
 // 2) select only one data point
 for (int i = 0; i < rank; i++) {
     start[0] = 0;
     count[i] = 1;
 }
 
 // 3)  read one data point
 data = dset.getData();
 
 // 4)  reset to select the whole dataset
 dset.init();
 
 // 5) clean the memory data buffer
 dset.clearData();
 
 // 6) Read the whole dataset
 data = dset.getData();
 

Specified by:

init in class Dataset

setName

public void setName(java.lang.String newName)
             throws java.lang.Exception

Description copied from class: HObject

Sets the name of the object.

setName (String newName) changes the name of the object in the file.

Overrides:

setName in class HObject

Parameters:

newName - The new name of the object.

Throws:

java.lang.Exception

create

@Deprecated
public static Dataset create(java.lang.String name,
                                        Group pgroup,
                                        long[] dims,
                                        java.lang.String[] memberNames,
                                        Datatype[] memberDatatypes,
                                        int[] memberSizes,
                                        java.lang.Object data)
                      throws java.lang.Exception

Deprecated. Not for public use in the future.
Using #create(String, Group, long[], long[], long[], int, String[], Datatype[], int[], int[][], Object)

Throws:

java.lang.Exception

create

@Deprecated
public static Dataset create(java.lang.String name,
                                        Group pgroup,
                                        long[] dims,
                                        java.lang.String[] memberNames,
                                        Datatype[] memberDatatypes,
                                        int[] memberRanks,
                                        long[][] memberDims,
                                        java.lang.Object data)
                      throws java.lang.Exception

Deprecated. Not for public use in the future.
Using #create(String, Group, long[], long[], long[], int, String[], Datatype[], int[], int[][], Object)

Throws:

java.lang.Exception

create

public static Dataset create(java.lang.String name,
                             Group pgroup,
                             long[] dims,
                             long[] maxdims,
                             long[] chunks,
                             int gzip,
                             java.lang.String[] memberNames,
                             Datatype[] memberDatatypes,
                             int[] memberRanks,
                             long[][] memberDims,
                             java.lang.Object data)
                      throws java.lang.Exception

Creates a simple compound dataset in a file with/without chunking and compression

This function provides an easy way to create a simple compound dataset in file by hiding tedious details of creating a compound dataset from users.

This functoin calls H5.H5Dcreate() to create a simple compound dataset in file. Nested compound dataset is not supported. The required information to create a compound dataset includes the name, the parent group and data space of the dataset, the names, datatypes and data spaces of the compound fields. Other information such as chunks, compression and the data buffer is optional.

The following example shows how to use this function to create a compound dataset in file.

 H5File file = null;
 String message = "";
 Group pgroup = null;
 int[] DATA_INT = new int[DIM_SIZE];
 float[] DATA_FLOAT = new float[DIM_SIZE];
 String[] DATA_STR = new String[DIM_SIZE];
 long[] DIMs = { 50, 10 };
 long[] CHUNKs = { 25, 5 };
 
 try {
     file = (H5File) H5FILE.open(fname, H5File.CREATE);
     file.open();
     pgroup = (Group) file.get("/");
 }
 catch (Exception ex) {
 }
 
 Vector data = new Vector();
 data.add(0, DATA_INT);
 data.add(1, DATA_FLOAT);
 data.add(2, DATA_STR);
 
 // create groups
 Datatype[] mdtypes = new H5Datatype[3];
 String[] mnames = { "int", "float", "string" };
 Dataset dset = null;
 try {
     mdtypes[0] = new H5Datatype(Datatype.CLASS_INTEGER, 4, -1, -1);
     mdtypes[1] = new H5Datatype(Datatype.CLASS_FLOAT, 4, -1, -1);
     mdtypes[2] = new H5Datatype(Datatype.CLASS_STRING, STR_LEN, -1, -1);
     dset = file.createCompoundDS("/CompoundDS", pgroup, DIMs, null, CHUNKs, 9,
             mnames, mdtypes, null, data);
 }
 catch (Exception ex) {
     failed(message, ex, file);
     return 1;
 }
 

Parameters:

name - the name of the new dataset

pgroup - parent group where the new dataset is created.

dims - the dimension size

maxdims - maximum dimension sizes of the new dataset, null if maxdims is the same as dims.

chunks - chunk sizes of the new dataset, null if no chunking

gzip - GZIP compression level (1 to 9), 0 or negative values if no compression.

memberNames - the names of compound datatype

memberDatatypes - the datatypes of the compound datatype

memberRanks - the ranks of the members

memberDims - the dim sizes of the members

data - list of data arrays written to the new dataset, null if no data is written to the new dataset.

Returns:

the new compound dataset if successful; otherwise returns null

Throws:

java.lang.Exception

isString

public boolean isString(int tid)

Description copied from class: Dataset

Checks if a given datatype is a string. Sub-classes must replace this default implementation.

Overrides:

isString in class Dataset

Parameters:

tid - The data type identifier.

Returns:

true if the datatype is a string; otherwise returns false.

getSize

public int getSize(int tid)

Description copied from class: Dataset

Returns the size in bytes of a given datatype. Sub-classes must replace this default implementation.

Overrides:

getSize in class Dataset

Parameters:

tid - The data type identifier.

Returns:

The size of the datatype