Examples of com.hp.hpl.jena.tdb.index.bplustree.BPlusTree

com.hp.hpl.jena.tdb.index.bplustree.BPlusTree

B-Tree converted to B+Tree B-Tree taken from: Introduction to Algorithms, Second Edition Chapter 18: B-Trees by Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest and Clifford Stein Includes implementation of removal. Notes: Stores "records", which are a key and value (the value may be null). In this B+Tree implementation, the (key,value) pairs are held in RecordBuffer, which wrap a ByteBuffer that only has records in it. BPTreeRecords provides the B+Tree view of a RecordBuffer. All records are in RecordBuffer - the "tree" part is an index for finding the right page. The tree only holds keys, copies from the (key, value) pairs in the RecordBuffers. Notes: The version above splits nodes on the way down when full, not when needed where a split can bubble up from below. It means it only ever walks down the tree on insert. Similarly, the delete code ensures a node is suitable before decending. Variations: In this impl, splitRoot leaves the root node in place. The root is always the same block.

    int writeCacheSize = 100 ;
    FileSet destination = new FileSet(location, Names.indexNode2Id) ;
    BlockMgr blkMgrNodes = BlockMgrFactory.create(destination, Names.bptExtTree, SystemTDB.BlockSize, readCacheSize, writeCacheSize) ;
    BlockMgr blkMgrRecords = BlockMgrFactory.create(destination, Names.bptExtRecords, SystemTDB.BlockSize, readCacheSize, writeCacheSize) ;
    Iterator<Record> iter2 = Iter.iter(sortedDataBag.iterator()).map(transformPair2Record) ;
      BPlusTree bpt2 = BPlusTreeRewriter.packIntoBPlusTree(iter2, bptParams, recordFactory, blkMgrNodes, blkMgrRecords) ;
      bpt2.sync() ;
      bpt2.close() ;
    
    sortedDataBag.close();
  }

Examples of com.hp.hpl.jena.tdb.index.bplustree.BPlusTree

Related Classes of com.hp.hpl.jena.tdb.index.bplustree.BPlusTree