Package edu.caltech.nanodb.queryeval

This package contains the query-evaluation components of the database. Query evaluation includes the following kinds of statements:

• SELECT statements
• INSERT statements
• UPDATE statements
• DELETE statements

In fact, the only real difference with any of these statements is what is done with the rows that are retrieved.

Commands that drive the evaluation components are represented as the QueryCommand class and its subclasses. The details of the commands being performed are filled in by the SQL parser.

Interface Summary

Interface	Description
Planner	This interface specifies the common entry-point for all query planner/optimizer implementations.
TupleProcessor	This interface allows us to decouple the implementation of the query evaluation loop from the operations performed on each tuple generated by the evaluator.

Class Summary

Class	Description
ColumnStats	This class holds some useful statistics for a specific column.
ColumnStatsCollector	This class facilitates the collection of statistics for a single column of a table being analyzed by the TableManager.analyzeTable(edu.caltech.nanodb.relations.TableInfo) method.
CostBasedJoinPlanner	This planner implementation uses dynamic programming to devise an optimal join strategy for the query.
CostBasedJoinPlanner.JoinComponent	This helper class is used to keep track of one "join component" in the dynamic programming algorithm.
Definitions	This class holds expression aliases and view definitions during query planning.
EvalStats	This class holds statistics generated from a query evaluation operation, such as the number of rows fetched and the total time to perform the evaluation.
PlanCost	This class holds a collection of values that represent the cost of an operation against the database.
PrettyPrintTools	Various methods to use for pretty-printing, such as delimiters, padding, and such.
PrettyTuplePrinter	This implementation of the tuple-processor interface prints out the schema and the tuples produced by the SELECT statement in a cooler, prettier fashion.
QueryEvaluator
SelectivityEstimator	This utility class is used to estimate the selectivity of predicates that appear on Select and Theta-Join plan-nodes.
SimplestPlanner	This class generates execution plans for very simple SQL SELECT * FROM tbl [WHERE P] queries.
StatisticsUpdater	This helper class provides basic functionality for updating column statistics based on a selection predicate.
TableStats	This class is a simple wrapper for table-file statistics.
TuplePrinter	This implementation of the tuple-processor interface simply prints out the schema and tuples produced by the SELECT statement.

Exception Summary

Exception	Description
InvalidSQLException	This exception is used to signal when a SQL query contains a semantic error that prevents its evaluation.