Class InputBitStream

  • All Implemented Interfaces:
    BooleanIterator, Closeable, Flushable, AutoCloseable, Iterator<Boolean>
    Direct Known Subclasses:
    DebugInputBitStream

    public class InputBitStream
    extends Object
    implements BooleanIterator, Flushable, Closeable
    Bit-level input stream.

    This class wraps any InputStream so that you can treat it as bit stream. Constructors and methods closely resemble those of InputStream. Data can be read from such a stream in several ways: reading a (long) natural number in fixed-width, unary, γ, shifted γ, δ, ζ and (skewed) Golomb coding, or reading a number of bits that will be stored in a vector of bytes. There is limited support for mark(int)/reset() operations.

    This class can also wrap a byte array; this is much more lightweight than wrapping a FastByteArrayInputStream wrapping the array. Overflowing the array will cause an EOFException.

    Note that when reading using a vector of bytes bits are read in the stream format (see OutputBitStream): the first bit is bit 7 of the first byte, the eighth bit is bit 0 of the first byte, the ninth bit is bit 7 of the second byte and so on. When reading natural numbers using some coding, instead, they are stored in the standard way, that is, in the lower bits.

    Additional features:

    • This class provides an internal buffer. By setting a buffer of length 0 at creation time, you can actually bypass the buffering system: Note, however, that several classes providing buffering have synchronised methods, so using a wrapper instead of the internal buffer is likely to lead to a performance drop.
    • To work around the schizophrenic relationship between streams and random access files in java.io, this class provides a flush() method that resets the internal state. At this point, you can safely reposition the underlying stream and read again afterwards. For instance, this is safe and will perform as expected:
       FileInputStream fis = new FileInputStream(...);
       InputBitStream ibs = new InputBitStream(fis);
       ... read operations on ibs ...
       ibs.flush();
       fis.getChannel().position(...);
       ... other read operations on ibs ...
       

      As a commodity, an instance of this class will try to cast the underlying byte stream to a RepositionableStream and to fetch by reflection the FileChannel underlying the given input stream, in this order. If either reference can be successfully fetched, you can use directly the position() method with argument pos with the same semantics of a flush(), followed by a call to position(pos / 8) (where the latter method belongs either to the underlying stream or to its underlying file channel), followed by a skip(pos % 8). However, since the reflective checks are quite heavy they can be disabled using a suitable constructor.

    • Finally, this class implements partially the interface of a boolean iterator. More precisely, nextBoolean() will return the same bit as readBit(), and also the same exceptions, whereas hasNext() will always return true: you must be prepared to catch a RuntimeException wrapping an IOException in case the file ends. It is very difficult to implement completely an eager operator using a input-stream based model.

    This class is not synchronised. If multiple threads access an instance of this class concurrently, they must be synchronised externally.

    Since:
    0.1
    Author:
    Sebastiano Vigna
    See Also:
    InputStream, OutputBitStream
    • Method Summary

      Modifier and Type Method Description
      void align()
      Aligns the stream.
      long available()
      Returns the number of bits that can be read (or skipped over) from this bit stream without blocking by the next caller of a method.
      static boolean checkLength​(String resource, int[] array, String resouceFullPath)  
      void close()
      Closes the bit stream.
      void copyTo​(OutputBitStream obs, long length)
      Copies a given number of bits from this input bit stream into a given output bit stream.
      void flush()
      Flushes the bit stream.
      boolean hasNext()  
      void mark​(int readLimit)
      Marks the current position in this input stream.
      boolean markSupported()
      Tests if this stream supports the mark(int) and reset() methods.
      boolean nextBoolean()  
      long position()
      Returns this stream bit position.
      void position​(long position)
      Sets this stream bit position, if it is based on a RepositionableStream or on a FileChannel.
      void read​(byte[] bits, int len)
      Reads a sequence of bits.
      int readBit()
      Reads a bit.
      long readBits()
      Returns the number of bits read from this bit stream.
      void readBits​(long readBits)
      Sets the number of bits read from this bit stream.
      int readDelta()
      Reads a natural number in δ coding.
      void readDeltas​(int[] a, int count)
      Reads a given amount of δ-coded natural numbers.
      int readGamma()
      Reads a natural number in γ coding.
      void readGammas​(int[] a, int count)
      Reads a given amount of γ-coded natural numbers.
      int readGolomb​(int b)
      Reads a natural number in Golomb coding.
      int readGolomb​(int b, int log2b)
      Reads a natural number in Golomb coding.
      int readInt​(int len)
      Reads a fixed number of bits into an integer.
      long readLong​(int len)
      Reads a fixed number of bits into a long.
      long readLongDelta()
      Reads a long natural number in δ coding.
      long readLongGamma()
      Reads a long natural number in γ coding.
      long readLongGolomb​(long b)
      Reads a long natural number in Golomb coding.
      long readLongGolomb​(long b, int log2b)
      Reads a long natural number in Golomb coding.
      long readLongMinimalBinary​(long b)
      Reads a long natural number in a limited range using a minimal binary coding.
      long readLongMinimalBinary​(long b, int log2b)
      Reads a long natural number in a limited range using a minimal binary coding.
      long readLongNibble()
      Reads a long natural number in variable-length nibble coding.
      long readLongShiftedGamma()
      Reads a natural number in shifted γ coding.
      long readLongSkewedGolomb​(long b)
      Reads a long natural number in skewed Golomb coding.
      long readLongUnary()
      Reads a long natural number in unary coding.
      long readLongZeta​(int k)
      Reads a long natural number in ζ coding.
      int readMinimalBinary​(int b)
      Reads a natural number in a limited range using a minimal binary coding.
      int readMinimalBinary​(int b, int log2b)
      Reads a natural number in a limited range using a minimal binary coding.
      int readNibble()
      Reads a natural number in variable-length nibble coding.
      int readShiftedGamma()
      Reads a natural number in shifted γ coding.
      void readShiftedGammas​(int[] a, int count)
      Reads a given amount of shifted-γ-coded natural numbers.
      int readSkewedGolomb​(int b)
      Reads a natural number in skewed Golomb coding.
      int readUnary()
      Reads a natural number in unary coding.
      int readZeta​(int k)
      Reads a natural number in ζ coding.
      void readZetas​(int k, int[] a, int count)
      Reads a given amount of γ-coded natural numbers.
      void reset()
      Repositions this bit stream to the position at the time the mark(int) method was last called.
      int skip​(int n)
      Deprecated.
      This method is simply an expensive, try/catch-surrounded version of skip(long) that is made necessary by the interface by BooleanIterator.
      long skip​(long n)
      Skips the given number of bits.
      void skipDeltas​(int n)
      Skips a given amount of δ-coded natural numbers.
      void skipDeltas​(long n)
      Skips a given amount of δ-coded natural numbers.
      void skipGammas​(int n)
      Skips a given amount of γ-coded natural numbers.
      void skipGammas​(long n)
      Skips a given amount of γ-coded natural numbers.
      void skipShiftedGammas​(int n)
      Skips a given amount of shifted-γ-coded natural numbers.
      void skipShiftedGammas​(long n)
      Skips a given amount of shifted-γ-coded natural numbers.
      void skipZetas​(int k, int n)
      Skips a given amount of ζ-coded natural numbers.
      void skipZetas​(int k, long n)
      Skips a given amount of ζ-coded natural numbers.
    • Field Detail

      • GAMMA

        public static final int[] GAMMA
      • DELTA

        public static final int[] DELTA
      • ZETA_3

        public static final int[] ZETA_3
      • SHIFTED_GAMMA

        public static final int[] SHIFTED_GAMMA
      • DEFAULT_BUFFER_SIZE

        public static final int DEFAULT_BUFFER_SIZE
        The default size of the byte buffer in bytes (8Ki).
        See Also:
        Constant Field Values
      • fileChannel

        protected final FileChannel fileChannel
        The cached file channel underlying is, if any.
      • repositionableStream

        protected final RepositionableStream repositionableStream
        is cast to a positionable stream, if possible.
      • wrapping

        protected final boolean wrapping
        True if we are wrapping an array.
      • buffer

        protected byte[] buffer
        The stream buffer.
      • fill

        protected int fill
        Current number of bits in the bit buffer (stored low).
      • pos

        protected int pos
        Current position in the byte buffer.
      • avail

        protected int avail
        Current number of bytes available in the byte buffer.
      • position

        protected long position
        Current position of the first byte in the byte buffer.
    • Constructor Detail

      • InputBitStream

        protected InputBitStream()
        This (non-public) constructor exists just to provide fake initialisation for classes such as DebugInputBitStream.
      • InputBitStream

        public InputBitStream​(InputStream is)
        Creates a new input bit stream wrapping a given input stream using a buffer of size DEFAULT_BUFFER_SIZE.

        This constructor performs the reflective tests that are necessary to support position(long).

        Parameters:
        is - the input stream to wrap.
      • InputBitStream

        public InputBitStream​(InputStream is,
                              boolean testForPosition)
        Creates a new input bit stream wrapping a given input stream using a buffer of size DEFAULT_BUFFER_SIZE.
        Parameters:
        is - the input stream to wrap.
        testForPosition - if false, the reflective test that is necessary to support position(long) in case is does not implement RepositionableStream will not be performed.
      • InputBitStream

        public InputBitStream​(InputStream is,
                              int bufSize)
        Creates a new input bit stream wrapping a given input stream with a specified buffer size.

        This constructor performs the reflective tests that are necessary to support position(long).

        Parameters:
        is - the input stream to wrap.
        bufSize - the size in byte of the buffer; it may be 0, denoting no buffering.
      • InputBitStream

        public InputBitStream​(InputStream is,
                              int bufSize,
                              boolean testForPosition)
        Creates a new input bit stream wrapping a given input stream with a specified buffer size.
        Parameters:
        is - the input stream to wrap.
        bufSize - the size in byte of the buffer; it may be 0, denoting no buffering.
        testForPosition - if false, the reflective test that is necessary to support position(long) in case is does not implement RepositionableStream will not be performed.
      • InputBitStream

        public InputBitStream​(FileInputStream is,
                              int bufSize)
        Creates a new input bit stream wrapping a given file input stream with a specified buffer size.

        This constructor invokes directly FileInputStream.getChannel() to support position(long).

        Parameters:
        is - the file input stream to wrap.
        bufSize - the size in byte of the buffer; it may be 0, denoting no buffering.
      • InputBitStream

        public InputBitStream​(byte[] a)
        Creates a new input bit stream wrapping a given byte array.
        Parameters:
        a - the byte array to wrap.
    • Method Detail

      • checkLength

        public static boolean checkLength​(String resource,
                                          int[] array,
                                          String resouceFullPath)
      • flush

        public void flush()
        Flushes the bit stream. All state information associated with the stream is reset. This includes bytes prefetched from the stream, bits in the bit buffer and unget'd bits.

        This method is provided so that users of this class can easily wrap repositionable streams (for instance, file-based streams, which can be repositioned using the underlying FileChannel). It is guaranteed that after calling this method the underlying stream can be repositioned, and that the next read will draw data from the stream.

        Specified by:
        flush in interface Flushable
      • available

        public long available()
                       throws IOException
        Returns the number of bits that can be read (or skipped over) from this bit stream without blocking by the next caller of a method.
        Returns:
        the number of bits that can be read from this bit stream without blocking.
        Throws:
        IOException
      • readBits

        public long readBits()
        Returns the number of bits read from this bit stream.
        Returns:
        the number of bits read so far.
      • readBits

        public void readBits​(long readBits)
        Sets the number of bits read from this bit stream.

        This method is provided so that, for instance, the user can reset via readBits(0) the read-bits count after a flush().

        Parameters:
        readBits - the new value for the number of bits read so far.
      • align

        public void align()
        Aligns the stream. After a call to this function, the stream is byte aligned. Bits that have been read to align are discarded.
      • read

        public void read​(byte[] bits,
                         int len)
                  throws IOException
        Reads a sequence of bits. Bits will be read in the natural way: the first bit is bit 7 of the first byte, the eightth bit is bit 0 of the first byte, the ninth bit is bit 7 of the second byte and so on.
        Parameters:
        bits - an array of bytes to store the result.
        len - the number of bits to read.
        Throws:
        IOException
      • readBit

        public int readBit()
                    throws IOException
        Reads a bit.
        Returns:
        the next bit from the stream.
        Throws:
        IOException
      • readInt

        public int readInt​(int len)
                    throws IOException
        Reads a fixed number of bits into an integer.
        Parameters:
        len - a bit length.
        Returns:
        an integer whose lower len bits are taken from the stream; the rest is zeroed.
        Throws:
        IOException
      • readLong

        public long readLong​(int len)
                      throws IOException
        Reads a fixed number of bits into a long.
        Parameters:
        len - a bit length.
        Returns:
        a long whose lower len bits are taken from the stream; the rest is zeroed.
        Throws:
        IOException
      • skip

        public long skip​(long n)
                  throws IOException
        Skips the given number of bits.
        Parameters:
        n - the number of bits to skip.
        Returns:
        the actual number of skipped bits.
        Throws:
        IOException
      • position

        public long position()
        Returns this stream bit position.
        Returns:
        this stream bit position.
      • markSupported

        public boolean markSupported()
        Tests if this stream supports the mark(int) and reset() methods.

        This method will just delegate the test to the underlying InputStream.

        Returns:
        whether this stream supports mark(int)/reset().
      • mark

        public void mark​(int readLimit)
                  throws IOException
        Marks the current position in this input stream. A subsequent call to the reset() method repositions this stream at the last marked position so that subsequent reads re-read the same bits.

        This method will just delegate the mark to the underlying InputStream. Moreover, it will throw an exception if you try to mark outsite byte boundaries.

        Parameters:
        readLimit - the maximum limit of bytes that can be read before the mark position becomes invalid.
        Throws:
        IOException - if you try to mark outside byte boundaries.
      • readLongUnary

        public long readLongUnary()
                           throws IOException
        Reads a long natural number in unary coding. Note that by unary coding we mean that 1 encodes 0, 01 encodes 1 and so on.
        Returns:
        the next unary-encoded long natural number.
        Throws:
        IOException
        See Also:
        OutputBitStream.writeUnary(int)
      • skipGammas

        public void skipGammas​(long n)
                        throws IOException
        Skips a given amount of γ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readGamma() or readLongGamma(), as precomputed tables are used directly, so the number of method calls is greatly reduced, and the result is discarded, so skip(long) can be invoked instead of more specific decoding methods.

        Parameters:
        n - the number of γ-coded natural numbers to be skipped.
        Throws:
        IOException
        See Also:
        readGamma()
      • skipGammas

        public void skipGammas​(int n)
                        throws IOException
        Skips a given amount of γ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readGamma() or readLongGamma(), as precomputed tables are used directly, so the number of method calls is greatly reduced, and the result is discarded, so skip(long) can be invoked instead of more specific decoding methods.

        Parameters:
        n - the number of γ-coded natural numbers to be skipped.
        Throws:
        IOException
        See Also:
        readGamma()
      • readGammas

        public void readGammas​(int[] a,
                               int count)
                        throws IOException
        Reads a given amount of γ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readGamma(), as precomputed tables are used directly, so the number of method calls is greatly reduced.

        Parameters:
        a - an array of at least count integers where the result will be written starting at the first position.
        count - the number of γ-coded natural numbers to be read.
        Throws:
        IOException
        See Also:
        readGamma()
      • skipShiftedGammas

        public void skipShiftedGammas​(long n)
                               throws IOException
        Skips a given amount of shifted-γ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readShiftedGamma() or readLongShiftedGamma(), as precomputed tables are used directly, so the number of method calls is greatly reduced, and the result is discarded, so skip(long) can be invoked instead of more specific decoding methods.

        Parameters:
        n - the number of shifted-γ-coded natural numbers to be skipped.
        Throws:
        IOException
        See Also:
        readShiftedGamma()
      • skipShiftedGammas

        public void skipShiftedGammas​(int n)
                               throws IOException
        Skips a given amount of shifted-γ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readShiftedGamma() or readLongShiftedGamma(), as precomputed tables are used directly, so the number of method calls is greatly reduced, and the result is discarded, so skip(long) can be invoked instead of more specific decoding methods.

        Parameters:
        n - the number of shifted-γ-coded natural numbers to be skipped.
        Throws:
        IOException
        See Also:
        readShiftedGamma()
      • readShiftedGammas

        public void readShiftedGammas​(int[] a,
                                      int count)
                               throws IOException
        Reads a given amount of shifted-γ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readShiftedGamma(), as precomputed tables are used directly, so the number of method calls is greatly reduced.

        Parameters:
        a - an array of at least count integers where the result will be written starting at the first position.
        count - the number of shifted-γ-coded natural numbers to be read.
        Throws:
        IOException
        See Also:
        readShiftedGamma()
      • skipDeltas

        public void skipDeltas​(long n)
                        throws IOException
        Skips a given amount of δ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readDelta() or readLongDelta(), as precomputed tables are used directly, so the number of method calls is greatly reduced, and the result is discarded, so skip(long) can be invoked instead of more specific decoding methods.

        Parameters:
        n - the number of δ-coded natural numbers to be skipped.
        Throws:
        IOException
        See Also:
        readDelta()
      • skipDeltas

        public void skipDeltas​(int n)
                        throws IOException
        Skips a given amount of δ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readDelta() or readLongDelta(), as precomputed tables are used directly, so the number of method calls is greatly reduced, and the result is discarded, so skip(long) can be invoked instead of more specific decoding methods.

        Parameters:
        n - the number of δ-coded natural numbers to be skipped.
        Throws:
        IOException
        See Also:
        readDelta()
      • readDeltas

        public void readDeltas​(int[] a,
                               int count)
                        throws IOException
        Reads a given amount of δ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readDelta(), as precomputed tables are used directly, so the number of method calls is greatly reduced.

        Parameters:
        a - an array of at least count integers where the result will be written starting at the first position.
        count - the number of δ-coded natural numbers to be read.
        Throws:
        IOException
        See Also:
        readDelta()
      • readMinimalBinary

        public int readMinimalBinary​(int b,
                                     int log2b)
                              throws IOException
        Reads a natural number in a limited range using a minimal binary coding. This method is faster than readMinimalBinary(int) because it does not have to compute log2b.
        Parameters:
        b - a strict upper bound.
        log2b - the floor of the base-2 logarithm of the bound.
        Returns:
        the next minimally binary encoded natural number.
        Throws:
        IllegalArgumentException - if you try to read a negative number or use a nonpositive base.
        IOException
        See Also:
        OutputBitStream.writeMinimalBinary(int, int)
      • readLongMinimalBinary

        public long readLongMinimalBinary​(long b,
                                          int log2b)
                                   throws IOException
        Reads a long natural number in a limited range using a minimal binary coding. This method is faster than readLongMinimalBinary(long) because it does not have to compute log2b.
        Parameters:
        b - a strict upper bound.
        log2b - the floor of the base-2 logarithm of the bound.
        Returns:
        the next minimally binary encoded long natural number.
        Throws:
        IllegalArgumentException - if you try to read a negative number or use a nonpositive base.
        IOException
        See Also:
        OutputBitStream.writeMinimalBinary(int, int)
      • readGolomb

        public int readGolomb​(int b)
                       throws IOException
        Reads a natural number in Golomb coding.

        This method implements also the case in which b is 0: in this case, nothing will be read, and 0 will be returned.

        Parameters:
        b - the modulus for the coding.
        Returns:
        the next Golomb-encoded natural number.
        Throws:
        IllegalArgumentException - if you use a nonpositive modulus.
        IOException
        See Also:
        OutputBitStream.writeGolomb(int, int)
      • readGolomb

        public int readGolomb​(int b,
                              int log2b)
                       throws IOException
        Reads a natural number in Golomb coding. This method is faster than readGolomb(int) because it does not have to compute log2b.

        This method implements also the case in which b is 0: in this case, nothing will be read, and 0 will be returned.

        Parameters:
        b - the modulus for the coding.
        log2b - the floor of the base-2 logarithm of the coding modulus.
        Returns:
        the next Golomb-encoded natural number.
        Throws:
        IllegalArgumentException - if you use a nonpositive modulus.
        IOException
        See Also:
        OutputBitStream.writeGolomb(int, int)
      • readLongGolomb

        public long readLongGolomb​(long b)
                            throws IOException
        Reads a long natural number in Golomb coding.

        This method implements also the case in which b is 0: in this case, nothing will be read, and 0 will be returned.

        Parameters:
        b - the modulus for the coding.
        Returns:
        the next Golomb-encoded long natural number.
        Throws:
        IllegalArgumentException - if you use a nonpositive modulus.
        IOException
        See Also:
        OutputBitStream.writeGolomb(int, int)
      • readLongGolomb

        public long readLongGolomb​(long b,
                                   int log2b)
                            throws IOException
        Reads a long natural number in Golomb coding. This method is faster than readLongGolomb(long) because it does not have to compute log2b.

        This method implements also the case in which b is 0: in this case, nothing will be read, and 0 will be returned.

        Parameters:
        b - the modulus for the coding.
        log2b - the floor of the base-2 logarithm of the coding modulus.
        Returns:
        the next Golomb-encoded long natural number.
        Throws:
        IllegalArgumentException - if you use a nonpositive modulus.
        IOException
        See Also:
        OutputBitStream.writeGolomb(int, int)
      • readSkewedGolomb

        public int readSkewedGolomb​(int b)
                             throws IOException
        Reads a natural number in skewed Golomb coding.

        This method implements also the case in which b is 0: in this case, nothing will be read, and 0 will be returned.

        Parameters:
        b - the modulus for the coding.
        Returns:
        the next skewed Golomb-encoded natural number.
        Throws:
        IllegalArgumentException - if you use a negative modulus.
        IOException
        See Also:
        OutputBitStream.writeSkewedGolomb(int, int)
      • readLongSkewedGolomb

        public long readLongSkewedGolomb​(long b)
                                  throws IOException
        Reads a long natural number in skewed Golomb coding.

        This method implements also the case in which b is 0: in this case, nothing will be read, and 0 will be returned.

        Parameters:
        b - the modulus for the coding.
        Returns:
        the next skewed Golomb-encoded long natural number.
        Throws:
        IllegalArgumentException - if you use a negative modulus.
        IOException
        See Also:
        OutputBitStream.writeSkewedGolomb(int, int)
      • skipZetas

        public void skipZetas​(int k,
                              long n)
                       throws IOException
        Skips a given amount of ζ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readZeta(int) or readLongZeta(int), as precomputed tables are used directly, so the number of method calls is greatly reduced, and the result is discarded, so skip(long) can be invoked instead of more specific decoding methods.

        Parameters:
        k - the shrinking factor.
        n - the number of ζ-coded natural numbers to be skipped.
        Throws:
        IOException
        See Also:
        readZeta(int)
      • skipZetas

        public void skipZetas​(int k,
                              int n)
                       throws IOException
        Skips a given amount of ζ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readZeta(int) or readLongZeta(int), as precomputed tables are used directly, so the number of method calls is greatly reduced, and the result is discarded, so skip(long) can be invoked instead of more specific decoding methods.

        Parameters:
        k - the shrinking factor.
        n - the number of ζ-coded natural numbers to be skipped.
        Throws:
        IOException
        See Also:
        readZeta(int)
      • readZetas

        public void readZetas​(int k,
                              int[] a,
                              int count)
                       throws IOException
        Reads a given amount of γ-coded natural numbers.

        This method should be significantly quicker than iterating n times on readGamma(), as precomputed tables are used directly, so the number of method calls is greatly reduced.

        Parameters:
        k - the shrinking factor.
        a - an array of at least count integers where the result will be written starting at the first position.
        count - the number of ζ-coded natural numbers to be read.
        Throws:
        IOException
        See Also:
        readGamma()
      • skip

        @Deprecated
        public int skip​(int n)
        Deprecated.
        This method is simply an expensive, try/catch-surrounded version of skip(long) that is made necessary by the interface by BooleanIterator.
        Skips over the given number of bits.
        Specified by:
        skip in interface BooleanIterator
        Parameters:
        n - the number of bits to skip.
        Returns:
        the number of bits actually skipped.
      • copyTo

        public void copyTo​(OutputBitStream obs,
                           long length)
                    throws IOException
        Copies a given number of bits from this input bit stream into a given output bit stream.
        Parameters:
        obs - an output bit stream.
        length - the number of bits to copy.
        Throws:
        EOFException - if there are not enough bits to copy.
        IOException