Java Doc for Table.java in  » Database-Client » prefuse » prefuse » data » Java Source Code / Java DocumentationJava Source Code and Java Documentation

A Table organizes a collection of data into rows and columns, each row containing a data record, and each column containing data values for a named data field with a specific data type. Table data can be accessed directly using the row number and column name, or rows can be treated in an object-oriented fashion using prefuse.data.Tuple instances that represent a single row of data in the table. As such, tables implement the prefuse.data.tuple.TupleSet interface.

Table rows can be inserted or deleted. In any case, none of the other existing table rows are effected by an insertion or deletion. A deleted row simply becomes invalid--any subsequent attempts to access the row either directly or through a pre-existing Tuple instance will result in an exception. However, if news rows are later added to the table, the row number for previously deleted nodes can be reused. In fact, the lower row number currently unused is assigned to the new row. This results in an efficient reuse of the table rows, but carries an important side effect -- rows do not necesarily maintain the order in which they were added once deletions have occurred on the table. If not deletions occur, the ordering of table rows will reflect the order in which rows were added to the table.

Collections of table rows can be accessed using both iterators over the actual row numbers and iterators over the Tuple instances that encapsulate access to that row. Both types of iteration can also be filtered by providing a prefuse.data.expression.Predicate , allowing tables to be queried for specific values.

Columns (alternativele referred to as data fields) can be added to the Table using Table.addColumn(String,Class) and a host of similar methods. This method will automatically determine the right kind of backing column instance to use. Furthermore, Table columns can be specified using a Schema instance, which describes the column names, data types, and default values. The Table class also maintains its own internal Schema, which be accessed (in a read-only way) using the Table.getSchema() method.

Tables also support additional structures. The ColumnMetadata class returned by the Table.getMetadata(String) method supports calculation of different statistics for a column, including minimum and maximum values, and the number of unique data values in the column. prefuse.data.util.Index instances can be created and retrieved using the Table.index(String) method and retrieved without triggering creation using Table.getIndex(String) method. An index keeps a sorted collection of all data values in a column, accelerating the creation of filtered iterators by optimizing query calculations and also providing faster computation of many of the ColumnMetadata methods. If you will be issuing a number of queries (i.e., requesting filtered iterators) dependent on the values of a given column, indexing that column may result in a significant performance increase, though at the cost of storing and maintaining the backing index structure.