Package edu.rit.mp

Class FloatBuf

java.lang.Object
edu.rit.mp.Buf
edu.rit.mp.FloatBuf
Direct Known Subclasses:
EmptyFloatBuf, FloatArrayBuf, FloatItemBuf, FloatMatrixBuf, SharedFloatArrayBuf, SharedFloatBuf

public abstract class FloatBuf extends Buf
Class FloatBuf is the abstract base class for a buffer of float items sent or received using the Message Protocol (MP). In a message, a float item is represented as four bytes, most significant byte first.

A buffer may be used to send one or more messages at the same time in multiple threads. If a buffer is being used to send a message or messages, the buffer must not be used to receive a message at the same time.

A buffer may be used to receive one message at a time. If a buffer is being used to receive a message, the buffer must not be used to receive another message in a different thread, and the buffer must not be used to send a message or messages.

A buffer is a conduit for retrieving and storing data in some underlying data structure. If the underlying data structure is multiple thread safe, then one thread can be retrieving or storing data via the buffer at the same time as other threads are accessing the data structure. If the underlying data structure is not multiple thread safe, then other threads must not access the data structure while one thread is retrieving or storing data via the buffer.

To create a FloatBuf, call one of the following static factory methods:

  • emptyBuffer()
  • buffer()
  • buffer (float)
  • buffer (float[])
  • sliceBuffer (float[], Range)
  • sliceBuffers (float[], Range[])
  • buffer (float[][])
  • rowSliceBuffer (float[][], Range)
  • rowSliceBuffers (float[][], Range[])
  • colSliceBuffer (float[][], Range)
  • colSliceBuffers (float[][], Range[])
  • patchBuffer (float[][], Range, Range)
  • patchBuffers (float[][], Range[], Range[])
  • buffer (SharedFloat)
  • buffer (SharedFloatArray)
  • sliceBuffer (SharedFloatArray, Range)
  • sliceBuffers (SharedFloatArray, Range[])
Version:
03-May-2008
Author:
Alan Kaminsky
  • Constructor Details

    • FloatBuf

      protected FloatBuf(int theLength)
      Construct a new float buffer.
      Parameters:
      theLength - Number of items.
      Throws:
      IllegalArgumentException - (unchecked exception) Thrown if theLength < 0.
  • Method Details

    • emptyBuffer

      public static FloatBuf emptyBuffer()
      Create an empty buffer. The buffer's length is 0. The buffer's item type is float.
      Returns:
      Empty buffer.
    • buffer

      public static FloatItemBuf buffer()
      Create a buffer for a float item. The item is stored in the item field of the buffer.
      Returns:
      Buffer.
    • buffer

      public static FloatItemBuf buffer(float item)
      Create a buffer for a float item with the given initial value. The item is stored in the item field of the buffer.
      Parameters:
      item - Initial value of the item field.
      Returns:
      Buffer.
    • buffer

      public static FloatBuf buffer(float[] theArray)
      Create a buffer for the entire given float array. The returned buffer encompasses all the elements in theArray.
      Parameters:
      theArray - Array.
      Returns:
      Buffer.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theArray is null.
    • sliceBuffer

      public static FloatBuf sliceBuffer(float[] theArray, Range theRange)
      Create a buffer for one slice of the given float array. The returned buffer encompasses theRange of elements in theArray. The range's stride may be 1 or greater than 1.
      Parameters:
      theArray - Array.
      theRange - Range of elements to include.
      Returns:
      Buffer.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theArray is null or theRange is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theArray does not include all the indexes in theRange.
    • sliceBuffers

      public static FloatBuf[] sliceBuffers(float[] theArray, Range[] theRanges)
      Create an array of buffers for multiple slices of the given float array. The returned buffer array has the same length as theRanges. Each element [i] of the returned buffer array encompasses the elements of theArray specified by theRanges[i]. Each range's stride may be 1 or greater than 1.
      Parameters:
      theArray - Array.
      theRanges - Array of ranges of elements to include.
      Returns:
      Array of buffers.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theArray is null or theRanges or any element thereof is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theArray's allocation does not include any element of theRanges.
    • buffer

      public static FloatBuf buffer(float[][] theMatrix)
      Create a buffer for the entire given float matrix. The returned buffer encompasses all the rows and all the columns in theMatrix.
      Parameters:
      theMatrix - Matrix.
      Returns:
      Buffer.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theMatrix is null.
    • rowSliceBuffer

      public static FloatBuf rowSliceBuffer(float[][] theMatrix, Range theRowRange)
      Create a buffer for one row slice of the given float matrix. The returned buffer encompasses theRowRange of rows, and all the columns, in theMatrix. The range's stride may be 1 or greater than 1.
      Parameters:
      theMatrix - Matrix.
      theRowRange - Range of rows to include.
      Returns:
      Buffer.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theMatrix is null or theRowRange is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theMatrix's allocation does not include theRowRange.
    • rowSliceBuffers

      public static FloatBuf[] rowSliceBuffers(float[][] theMatrix, Range[] theRowRanges)
      Create an array of buffers for multiple row slices of the given float matrix. The returned buffer array has the same length as theRowRanges. Each element [i] of the returned buffer array encompasses the rows of theMatrix specified by theRowRanges[i] and all the columns of theMatrix. Each range's stride may be 1 or greater than 1.
      Parameters:
      theMatrix - Matrix.
      theRowRanges - Array of ranges of rows to include.
      Returns:
      Array of buffers.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theMatrix is null or theRowRanges or any element thereof is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theMatrix's allocation does not include any element of theRowRanges.
    • colSliceBuffer

      public static FloatBuf colSliceBuffer(float[][] theMatrix, Range theColRange)
      Create a buffer for one column slice of the given float matrix. The returned buffer encompasses all the rows, and theColRange of columns, in theMatrix. The range's stride may be 1 or greater than 1.
      Parameters:
      theMatrix - Matrix.
      theColRange - Range of columns to include.
      Returns:
      Buffer.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theMatrix is null or theColRange is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theMatrix's allocation does not include theColRange.
    • colSliceBuffers

      public static FloatBuf[] colSliceBuffers(float[][] theMatrix, Range[] theColRanges)
      Create an array of buffers for multiple column slices of the given float matrix. The returned buffer array has the same length as theColRanges. Each element [i] of the returned buffer array encompasses all the rows of theMatrix and the columns of theMatrix specified by theColRanges[i]. Each range's stride may be 1 or greater than 1.
      Parameters:
      theMatrix - Matrix.
      theColRanges - Array of ranges of columns to include.
      Returns:
      Array of buffers.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theMatrix is null or theColRanges or any element thereof is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theMatrix's allocation does not include any element of theColRanges.
    • patchBuffer

      public static FloatBuf patchBuffer(float[][] theMatrix, Range theRowRange, Range theColRange)
      Create a buffer for one patch of the given float matrix. The returned buffer encompasses theRowRange of rows, and theColRange of columns, in theMatrix. Each range's stride may be 1 or greater than 1.
      Parameters:
      theMatrix - Matrix.
      theRowRange - Range of rows to include.
      theColRange - Range of columns to include.
      Returns:
      Buffer.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theMatrix is null, theRowRange is null, or theColRange is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theMatrix's allocation does not include theRowRange and theColRange.
    • patchBuffers

      public static FloatBuf[] patchBuffers(float[][] theMatrix, Range[] theRowRanges, Range[] theColRanges)
      Create an array of buffers for multiple patches of the given float matrix. The length of the returned buffer array is equal to the length of theRowRanges times the length of theColRanges. Each element of the returned buffer array encompasses the rows given in one element of theRowRanges array, and the columns given in one element of theColRanges array, in all possible combinations, of theMatrix. Each range's stride may be 1 or greater than 1.
      Parameters:
      theMatrix - Matrix.
      theRowRanges - Array of ranges of rows to include.
      theColRanges - Array of ranges of columns to include.
      Returns:
      Array of buffers.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theMatrix is null, theRowRanges or any element thereof is null, or theColRanges or any element thereof is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theMatrix's allocation does not include any element of theRowRanges or theColRanges.
    • buffer

      public static FloatBuf buffer(SharedFloat item)
      Create a buffer for a shared float item. The item is wrapped in an instance of class SharedFloat. Use the methods of the SharedFloat object to access the actual item.
      Parameters:
      item - SharedFloat object that wraps the item.
      Returns:
      a FloatBuf object.
      Throws:
      NullPointerException - (unchecked exception) Thrown if item is null.
    • buffer

      public static FloatBuf buffer(SharedFloatArray theArray)
      Create a buffer for the entire given shared float array. The returned buffer encompasses all the elements in theArray.
      Parameters:
      theArray - Array.
      Returns:
      Buffer.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theArray is null.
    • sliceBuffer

      public static FloatBuf sliceBuffer(SharedFloatArray theArray, Range theRange)
      Create a buffer for one slice of the given shared float array. The returned buffer encompasses theRange of elements in theArray. The range's stride may be 1 or greater than 1.
      Parameters:
      theArray - Array.
      theRange - Range of elements to include.
      Returns:
      Buffer.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theArray is null or theRange is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theArray does not include all the indexes in theRange.
    • sliceBuffers

      public static FloatBuf[] sliceBuffers(SharedFloatArray theArray, Range[] theRanges)
      Create an array of buffers for multiple slices of the given shared float array. The returned buffer array has the same length as theRanges. Each element [i] of the returned buffer array encompasses the elements of theArray specified by theRanges[i]. Each range's stride may be 1 or greater than 1.
      Parameters:
      theArray - Array.
      theRanges - Array of ranges of elements to include.
      Returns:
      Array of buffers.
      Throws:
      NullPointerException - (unchecked exception) Thrown if theArray is null or theRanges or any element thereof is null.
      IndexOutOfBoundsException - (unchecked exception) Thrown if theArray's allocation does not include any element of theRanges.
    • get

      public abstract float get(int i)
      Obtain the given item from this buffer.

      The get() method must not block the calling thread; if it does, all message I/O in MP will be blocked.

      Parameters:
      i - Item index in the range 0 .. length()-1.
      Returns:
      Item at index i.
    • put

      public abstract void put(int i, float item)
      Store the given item in this buffer.

      The put() method must not block the calling thread; if it does, all message I/O in MP will be blocked.

      Parameters:
      i - Item index in the range 0 .. length()-1.
      item - Item to be stored at index i.
    • copy

      public void copy(Buf theSrc)
      Copy items from the given buffer to this buffer. The number of items copied is this buffer's length or theSrc's length, whichever is smaller. If theSrc is this buffer, the copy() method does nothing. Copy items from the given buffer to this buffer. The number of items copied is this buffer's length or theSrc's length, whichever is smaller. If theSrc is this buffer, the copy() method does nothing.

      The default implementation of the copy() method calls the defaultCopy() method. A subclass can override the copy() method to use a more efficient algorithm.

      Specified by:
      copy in class Buf
      Parameters:
      theSrc - Source of items to copy into this buffer.
      Throws:
      ClassCastException - (unchecked exception) Thrown if theSrc's item data type is not the same as this buffer's item data type.
    • fill

      public void fill(Object item)
      Fill this buffer with the given item. The item is assigned to each element in this buffer.

      If this buffer's item data type is a primitive type, the item must be an instance of the corresponding primitive wrapper class -- class Integer for type int, class Double for type double, and so on. If the item is null, the item data type's default initial value is assigned to each element in this buffer.

      If this buffer's item data type is a nonprimitive type, the item must be an instance of the item class or a subclass thereof. The item may be null. Note that since item is assigned to every buffer element, every buffer element ends up referring to the same item. Fill this buffer with the given item. The item is assigned to each element in this buffer.

      The item must be an instance of class Float. If the item is null, 0.0f is assigned to each element in this buffer.

      Specified by:
      fill in class Buf
      Parameters:
      item - Item.
      Throws:
      ClassCastException - (unchecked exception) Thrown if the item's data type is not the same as this buffer's item data type.
    • getTemporaryBuf

      public Buf getTemporaryBuf()
      Create a temporary buffer with the same type of items and the same length as this buffer. The new buffer items are stored in a newly created array, separate from the storage for this buffer's items.
      Specified by:
      getTemporaryBuf in class Buf
      Returns:
      a Buf object.
    • defaultCopy

      protected static void defaultCopy(FloatBuf theSrc, FloatBuf theDst)
      Copy items from the given source buffer to the given destination buffer. The number of items copied is theSrc's length or theDst's length, whichever is smaller. Each item is copied individually using the get() and put() methods. It is assumed that theSrc is not the same as theDst.
      Parameters:
      theSrc - Source of items to copy.
      theDst - Destination of items to copy.