Count (database)

In the relational databases, a table is a whole of data (the values ) organized according to a model of lines (horizontal) and columns (vertical). Each table is the physical implementation of a relation between the various fields. Each correspondence is defined by a line of the table, and the values of a field are in the same column.

The concept of table appeared in the years 1970 at IBM with the relational Algèbre which is a mathematical theory in relation to the Set theory. The purpose of this theory is to clarify and facilitate the use of a Database.

Implementation

Design of a Table

When designing relational Database, it is important clearly to define all the tables which will compose it and the various relations which bind them, so as to be able to draw up the conceptual Schéma which will make it possible to describe the operation of the base given before setting up it by means of computer.

One will also distinguish the current tables which will contain various fields containing of information, and the tables of connection will ensure the connection between two current tables.

It is also essential to name the various tables rigouresement, by using technical standards. For example, that makes it possible to avoid the inconsistencies in the base and allows its evolutionarity.

One can for example fix oneself like normalizes:

Name of table always in the singular
For the tables connection to start with " l_" follow-up of the 2 dependant tables by classifying them alphabetically.

That will give for example for a table anybody and a table car , one defines the table of connection I_personne_voiture .

Contents of a Table

By nature, each column of a table, also named fields, must contain data of the same type and this fields must be also named. There are certain rules to respect the fact in particular that it is necessary to put an identifier for each recording in the table.

For the various fields contained in the table, it is also necessary to respect standards for the names, for example, to impose the name of the fields in the singular.

There are two possibilities:

To put an identifier which car increments entered data progressively

To choose an identifier which is single (for example: number plate for a database of cars)

That thus makes it possible to have a well indexed table, and that then facilitates perfectly research.

Work on a Table

There are two levels of work on a table:

a level of definition of the data of a table: who allows to define, bind, and force the data via a Data definition language
a level of handling of the data of a table: who allows to add, remove, seek data via a Data manipulation language

Currently, the language standardized to work on the table S is SQL, it is used with certain alternatives on the majority of the management systems of databases.

Example of management systems of databases using of the table S

Access of Microsoft
FileMaker Pro of FileMaker
HSQLDB bases relational written in Java
Informix
Ingres
Interbase of Borland
Oracle of Oracle Corporation
SQL Server of Microsoft
SQLite
MySql
Paradox of Corel

See too

Table

External bonds

databases, DBMS, waiters, introduction, a course on the relational Database S and Table S.

Random links: