org.geotools.data.jdbc.fidmapper
Interface FIDMapper

All Superinterfaces:

java.io.Serializable

All Known Implementing Classes:

AbstractFIDMapper

public interface FIDMapper

extends java.io.Serializable

The FIDMapper interface manages the mapping of feature id to the identifiers provided in a database.

Basically a FIDMapper must:

• generate the FID (a String) given the set of values that compose the primary key in the database
• turn the FID into the primary key values, or generate them should the FID be null
• provide notice wheter the identifier values should be included as attributes in the feature or not (this is necessary when reverse engineering the feature type from the database metadata)
• describe the primary key columns, if any (this is necessary when creating the table that describes the feature type in a table)

Concrete instances of this class should provide support for the most common primary key mapping and generation strategis, such as pk with business meaning, serials, sequences, and so on

Classes that implement this interface should ovveride equals to provide a state based comparison.

Author:

Dani Daniele Franzoni, aaime Andrea Aime

Method Summary

java.lang.String	createID(java.sql.Connection conn, Feature feature, java.sql.Statement statement) Creates a new ID for a feature.
int	getColumnCount() Returns the number of columns in the primary keys handled by this mapper
int	getColumnDecimalDigits(int colIndex) Provides the number of decimal digits for this column.
java.lang.String	getColumnName(int colIndex) Returns the name of the specified column in the primary key
int	getColumnSize(int colIndex) Returns the size of a primary key column as it would be provided by the database metadata.
int	getColumnType(int colIndex) Returns the column type by using a constant available in the java.sql.Types interface
java.lang.String	getID(java.lang.Object[] attributes) Returns the FID given the values of the prymary key attributes
java.lang.Object[]	getPKAttributes(java.lang.String FID) Creates the value for the PK attributes given the feature.
boolean	hasAutoIncrementColumns() Returns true if at least one column is of auto-increment type
void	initSupportStructures() This method will be called by JDBCDataStore when creating new tables to give the FID mapper an opportunity to initialize needed data structures, such as support tables, sequences, and so on.
boolean	isAutoIncrement(int colIndex) Returns true if the column is of serial type, that is, its value is automatically generated by the database if the user does not provide one
boolean	isVolatile() Returns true it the FID generated by this mapper are volatile, that is, if asking twice for the same Feature will not provide the same FID.
boolean	returnFIDColumnsAsAttributes() If true the primary key columns will be returned as attributes.

Method Detail

initSupportStructures

public void initSupportStructures()

This method will be called by JDBCDataStore when creating new tables to give the FID mapper an opportunity to initialize needed data structures, such as support tables, sequences, and so on.

getID

public java.lang.String getID(java.lang.Object[] attributes)

Returns the FID given the values of the prymary key attributes

Parameters:

attributes - DOCUMENT ME!

Returns:

getPKAttributes

public java.lang.Object[] getPKAttributes(java.lang.String FID)
                                   throws java.io.IOException

Creates the value for the PK attributes given the feature. If the FID is null, will throw an IOException if not possible. If null is returned, no primary key value needs to be specified, which is what we want for auto-increment fields.

Parameters:

FID - The feature ID is going to be parsed

Returns:

Throws:

java.io.IOException

createID

public java.lang.String createID(java.sql.Connection conn,
                                 Feature feature,
                                 java.sql.Statement statement)
                          throws java.io.IOException

Creates a new ID for a feature.
This is done either by querying the database (for auto-increment like types, for example sequences) or by inspecting the Feature (for example, for primary keys with business meaning that whose attributes are included in the Feature ones).

Parameters:

conn - - the database connection

feature - - the feature that needs the new FID

statement - - the statement used to insert the feature into the database

Returns:

Throws:

java.io.IOException

returnFIDColumnsAsAttributes

public boolean returnFIDColumnsAsAttributes()

If true the primary key columns will be returned as attributes. This is fundamental for primary key with businnes meaning.

Returns:

getColumnCount

public int getColumnCount()

Returns the number of columns in the primary keys handled by this mapper

Returns:

getColumnName

public java.lang.String getColumnName(int colIndex)

Returns the name of the specified column in the primary key

Parameters:

colIndex -

Returns:

getColumnType

public int getColumnType(int colIndex)

Returns the column type by using a constant available in the java.sql.Types interface

Parameters:

colIndex -

Returns:

getColumnSize

public int getColumnSize(int colIndex)

Returns the size of a primary key column as it would be provided by the database metadata. Some fields requires a size specification, such as VARCHAR or NUMBER, whilst other don't have or don't need it (for example, an INTEGER or a TEXT field).

Parameters:

colIndex -

Returns:

getColumnDecimalDigits

public int getColumnDecimalDigits(int colIndex)

Provides the number of decimal digits for this column. This is relevant in particular when the column is a scaled integer such as a NUMBER column

Parameters:

colIndex -

Returns:

isAutoIncrement

public boolean isAutoIncrement(int colIndex)

Returns true if the column is of serial type, that is, its value is automatically generated by the database if the user does not provide one

Parameters:

colIndex -

Returns:

hasAutoIncrementColumns

public boolean hasAutoIncrementColumns()

Returns true if at least one column is of auto-increment type

Returns:

isVolatile

public boolean isVolatile()

Returns true it the FID generated by this mapper are volatile, that is, if asking twice for the same Feature will not provide the same FID.

This is usually true for mappers that try to generate a FID for tables without primary keys.

When this method returns true, it's up to the datastore to decide what to do, but a sane policy may be to prevent Feature writing

Returns: