Introduction

Introduction – Introduction to MDB2_Schema

Introduction to MDB2_Schema

MDB2_Schema builds upon MDB2 to provide tools to manage your database schema using XML which is both platform- and database-independent.

The XML format is inherited from the Metabase package and was improved to be able to represent new entities such as Data Manipulation Instructions and Foreign Keys.

It enables users to maintain RDBMS independent schema files in XML that can be used to create, alter and drop database entities (also called as DDL: Data Definition Language) and insert data (also called as DML: Data Manipulation Language) into a database. Reverse engineering database schemas from existing databases is also supported. It also features the hability to parse database schemas and database data in separated files. However, in this document the term "schema file" will be used to designate any XML supported by MDB2_Schema, no matter its nature.

Reading of schema files is handled by MDB2_Schema_Parser and writing to them by MDB2_Schema_Writer. There is also the MDB2_Schema_Validate class which is called after something is parsed. It is supposed to check the logical integrity of the schema file, for instance whether you are trying to declare a index field that does not exist or even when you are trying to insert a float in a date field.

Currently two parsers are available, one based on the package XML_Parser (default) and another based on XML_Unserializer. As MDB2_Schema is still beta, both parsers have its own limitation. The default one can't parse more than 2 nested expressions. The later doesn't respect DML instructions in the order they are specified in the schema file.

Although MDB2_Schema API supports input from physical databases and schema files, internally it always operates using a database schema array that will be described later in the documentation.

    Поддержать сайт на родительском проекте КГБ