Package uk.ac.starlink.table.join

Classes for performing table joins by matching rows to each other.

See:
          Description

Interface Summary
MatchEngine Defines the details of object matching criteria.
ProgressIndicator Callback interface for indicating how an algorithm is progressing.
 

Class Summary
AbstractCartesianMatchEngine A matching engine which can match points in an ndim-dimensional space.
AnisotropicCartesianMatchEngine Matcher which matches in an anisotropic N-dimensional Cartesian space.
BinContents Describes the contents of all the bins used in a matching operation.
CombinedMatchEngine A matching engine which provides matching facilities by combining the characteristics of a number of other matching engines.
EqualsMatchEngine Match engine which considers two rows matched if they contain objects which are non-null and equal (in the sense of Object.equals(java.lang.Object)).
HTMMatchEngine Implements the object matching interface for sky coordinates using the HTM (Hierarchical Triangular Mesh) pixel-indexing scheme.
IsotropicCartesianMatchEngine Matcher which matches in an isotropic N-dimensional Cartesian space.
MatchStarTables Provides factory methods for producing tables which represent the result of row matching.
NullProgressIndicator Dummy progress indicator.
ProgressRowSequence RowSequence which logs progress to a ProgressIndicator.
RangeModelProgressIndicator ProgressIndicator implementation that can serve as the model for a JProgressBar.
RowLink Represents an ordered set of RowRefs which are considered in some way linked to each other.
RowMatcher Performs matching on the rows of one or more tables.
RowRef Represents a reference to a table row.
SkyMatchEngine Abstract superclass for match engines which match positions on the sky.
SphericalPolarMatchEngine Match engine which works with tuples representing RA, Dec and range.
TextProgressIndicator ProgressIndicator which logs progress to an output stream.
 

Package uk.ac.starlink.table.join Description

Classes for performing table joins by matching rows to each other. The matching is done in an entirely configurable and pluggable way by defining a MatchEngine which defines what counts as a match between two tuples (rows) and can be used to work out what counts as a near miss too. Matching is (except in pathological cases) an O(N) process, where N is the total number of rows in all the tables participating in a match. MatchEngine implementations are provided for a number of (astronomically) common cases, including N-dimensional Cartesian, spherical polar, exact value and Sky matching (the Sky one uses the Hierarchical Triangular Mesh scheme). You can also combine matches to an arbitrary level of complexity.


Copyright © 2004 CLRC: Central Laboratory of the Research Councils. All rights reserved.