POSC Specifications
Relational Implementation

How to Read the Epicentre Relational Implementation Dictionary

This section describes the conventions used in the documentation of the Sample Relational Implementation. It includes an explanation of the structure and content of the table descriptions. The table documentation is derived from the projected DDL and is intended to assist in the understanding of the relational implementation. In addition to this documentation, a complete set of DDL is delivered as text files.

An example of a table in the relational implementation of Epicentre is:


TRANSFORMATION_PARAMETER (TRANSFORMATION_PARAMETER)

(DDL)

This specified the kinds of parameters used by coordinate transformation methods. Examples of these parameters are: scale factor, false easting, central meridian, rules, etc.

Table is related to: data_collection_content, derived_interest_makeup, document_specification_part, @graphical_element, object_activity_involvement, technical_reference_alias, transformation_parameter_usage

Primary Key Constraint on: INSTANCE_S
Fully implemented Unique Constraint on: IDENTIFIER


As each table is presented, the table name is given in uppercase, boldface type. The long name of the corresponding Epicentre entity follows immediately in parenthesis, followed by a description of the table. Next, the columns that compose the table are described. The tables that have a foreign key from this table follows. The intersections that connect this table to other tables are then given. The definition of table constraints completes the description of each table.

Columns

Following the description of the table, the columns are described by a list of entries, one for each column in the table. Each column is described by column name in bold, lowercase. Following the column name, in parentheses, is a list of column qualifiers that are separated by commas and terminated by a colon. The column qualifiers have the following meaning: The data type will follow the colon. This can be any one of the SQL 92 specified data types, e.g. CHAR VARYING, DATE, TIMESTAMP, FLOAT, DECIMAL, etc. The declared data type is followed by a length qualifier in parentheses where appropriate.

The first indented line following the column name will contain the logical attribute name, preceded by its parent entity, that this column was projected from.

The next indented line is the description of the column.

When the column is a foreign key, then the next indented line will contain the phrase "Defines a foreign key to table" followed by the name of the table that is referenced. Where the name is prefixed by an @, then this is the name of a list of possible tables this is related to - this is a hypertext link that will expand the contents of the list.

Table is related to

This section contains a list of tables that are related to this table, i.e., other tables that contain a foreign key from this table. Each item in this list corresponds to a relationship to the table that does not appear as a foreign key in the table. Where a name is prefixed by an @, then this is the name of a list of possible tables this is related to - this is a hypertext link that will expand the contents of the list.

Intersection tables

This is a list of the tables that are related to this table and were created during the projection process to implement a many-to-many relationship.

Table Constraints

This section specifies one or two SQL constraints. The first places a uniqueness constraint on the surrogate key (The surrogate key is considered to be the primary key of the table.). The second statement (if present) defines the combined values of the columns forming the secondary key to be unique for each row entry.

© Copyright 1997- POSC. All rights reserved.