Public

Pulls bytes from the provided IndexInput.

<init>

copyBytes

finish

getPosition

getReverseReader

ramBytesUsed

Reverse from srcPos, inclusive, to destPos, inclusive.

reverse

skipBytes

Absolute write byte; you must ensure dest is < max position written so far.

writeByte

writeBytes

Writes all of our bytes to the target DataOutput.

writeTo

writeVLong

getBlockBits

getForwardReader

Pos must be less than the max position written so far! Ie, you cannot "grow" the
file with this!

truncate

writeVInt

Running tasks concurrently on multiple threads

Provides access to system-related information and resources including standard
input and output. Ena

java.lang.System

A controller for the selection of SelectableChannel objects. Selectable channels
can be registered w

java.nio.channels.Selector

A Java representation of the SQL TIMESTAMP type. It provides the capability of
representing the SQL 

java.sql.Timestamp

This class provides methods that return pseudo-random values.It is dangerous to
seed Random with the

java.util.Random

An Executor that provides methods to manage termination and methods that can
produce a Future for tr

java.util.concurrent.ExecutorService

A reentrant mutual exclusion Lock with the same basic behavior and semantics as
the implicit monitor

java.util.concurrent.locks.ReentrantLock

startActivity

getSystemService

getContentResolver

/** Expert: creates an FST by packing this one.  This
   *  process requires substantial additional RAM (currently
   *  up to ~8 bytes per node depending on
   *  <code>acceptableOverheadRatio</code>), but then should
   *  produce a smaller FST.
   *
   *  <p>The implementation of this method uses ideas from
   *  <a target="_blank" href="http://www.cs.put.poznan.pl/dweiss/site/publications/download/fsacomp.pdf">Smaller Representation of Finite State Automata</a>,
   *  which describes techniques to reduce the size of a FST.
   *  However, this is not a strict implementation of the
   *  algorithms described in this paper.
   */

// NOTE: maxDerefNodes is intentionally int: we cannot

//   - renumber the nodes to get more next / better locality?

//   - allow multiple input labels on an arc, so

//     singular chain of inputs can take one arc (on

//     wikipedia terms this could save another ~6%)

//   - in the ord case, the output '1' is presumably

//     very common (after NO_OUTPUT)... maybe use a bit

//   - use spare bits in flags.... for top few labels /

"this FST was not built with willPackFST=true"

// Find top nodes with highest number of incoming arcs:

// TODO: we could use more RAM efficient selection algo here...

//System.out.println("map node=" + n.node + " inCount=" + n.count + " to newID=" + downTo);

// +1 because node ords start at 1 (0 is reserved as stop node):

// Skip 0 byte since 0 is reserved target:

// Since we re-reverse the bytes, we now write the

// nodes backwards, so that BIT_TARGET_NEXT is

//System.out.println("  node: " + node + " address=" + address);

//System.out.println("    change: " + (address - newNodeAddress[node]));

How to use finishmethodin org.apache.lucene.util.fst.BytesStore

Best Java code snippets using org.apache.lucene.util.fst.BytesStore.finish (Showing top 6 results out of 315)

How to use
finish
method
in
org.apache.lucene.util.fst.BytesStore