DataBlade Module Development Overview DataBlade Module Concepts

A DataBlade module is a software package that extends the functionality of Informix Dynamic Server 2000. The package includes SQL statements and supporting code written in an external language or Informix SPL. DataBlade modules can also contain client components.

A DataBlade module adds user-defined database objects that extend the SQL syntax and commands you can use with your Informix database server. A database object is an SQL entity, such as a data type, routine, or database table. Your Informix database server handles DataBlade module objects as built-in objects. When it handles a user-defined database object, it executes the associated source code provided with the DataBlade module.

Extensions to your database server belong to two main categories:

• Types of data. This category includes extended data types that are not built into the database server. Extended data types can contain multiple elements (row, collection, and opaque data types) and data types that support inheritance (distinct and row data types). The internal structure of opaque data types is not accessible through built-in SQL commands, but it can be accessed through user-defined routines and opaque data type support routines.
• Routines. This category includes user-defined routines, aggregates, data type support routines, cast support routines, and routines that support user-defined access methods.

If you are unfamiliar with DataBlade module technology and the DataBlade Developers Kit, you might have the following questions:

• Why should I extend my database server?
• Why should I use a DataBlade module to extend my database server?
• Why should I use the DataBlade Developers Kit to create a DataBlade module?

Each of these questions is addressed in the following subsections.

The primary advantages of using the extensibility of Informix Dynamic Server 2000 over using traditional relational databases and applications are:

• better performance.
• simpler applications.
• transaction control.
• scalability.

Your Informix database server improves the performance of your applications in the following ways:

• User-defined routines are optimized.

When you put your custom routines in the database server, the query optimizer can calculate when to run them during queries.

• Indexes increase query speed.

Indexes, created with secondary access methods, can efficiently find and compare values. Secondary access methods build and manipulate index structures on data. With your Informix database server, you can create indexes on data that cannot be sorted in a standard relational database. You can implement your data as extended data types and create functional indexes to speed sorting. A functional index sorts information about the data, instead of the data itself.

For example, if your data is images, you can index features of the images. Then, when you run a query to match an image, the index runs much faster than comparing the binary files of each image.

• Network traffic is reduced.

When you use user-defined routines and other extensibility features, you perform more processing on the data within the database server. Therefore, you send less data to the client application.

Using DataBlade modules simplifies applications in the following ways:

• DataBlade modules handle code for manipulating and storing data so the application does not have to.
• DataBlade module routines and data types can be accessed using SQL.

SQL is a standard language and does not require complex application code or programming languages.

• DataBlade modules are easy to upgrade.

When you change a DataBlade module, you do not need to relink existing applications; all changes are handled within the database server.

• All data is stored and processed in the same database server.

For example, with a geospatial DataBlade module, geographic coordinates are analyzed and processed by the database server instead of in a complex application. In addition, the geospatial data is easily integrated with other types of data in a relational database.

• DataBlade modules are easy to combine.

You can combine DataBlade modules that handle different kinds of data in the same database. You can then create one application to integrate all the data.

For example, if a broadcast news company wanted to integrate video, images, audio, and text data for its programs, it could store all the data in one database and use a DataBlade module for each type of data. Then the company could use an application that employs the Informix Web DataBlade module to access information and display it in a Web browser.

DataBlade modules become part of the database. Therefore, all operations carried out by DataBlade module routines are supported by database services, such as backup, rollback, and recovery. You can safely store your data, which you formerly stored in files, in the database by using smart large objects.

DataBlade extensions to Dynamic Server scale to a large number of users just as well as the database server itself.

You can extend your Informix database server without creating a DataBlade module by executing the SQL statements to create each object individually. However, the advantages of packaging extended database objects in a DataBlade module include:

• control.
• code reuse.

DataBlade modules contain all related extended objects, enabling you to easily install, upgrade, and remove a whole module at once. If you need to fix a problem or add a feature to a program, you only have to do it in one place-the DataBlade module. Because a DataBlade module is a package ready to be distributed commercially or internally, these changes can be easily extended to any application that uses the DataBlade module. In addition, DataBlade modules make it easy for you to maintain version information about the software.

DataBlade modules can use the functionality of other DataBlade modules through interfaces. Interfaces are references to other DataBlade modules. When you include an interface in a DataBlade module, you create a dependency so that your DataBlade module can be used only if the DataBlade module that provides the interface is installed in the database server.

Some DataBlade modules are specifically designed as foundation DataBlade modules. Foundation DataBlade modules are not usually intended to be used alone. For example, the Informix Large Object Locator DataBlade module handles the location of the large objects that other DataBlade modules use to store their data.

Although you can create a DataBlade module manually, you can reduce development time considerably if you use the DataBlade Developers Kit.

Informix provides three graphical user interfaces for DataBlade module development:

• BladeSmith. To create your DataBlade module.
• BladePack. To package your DataBlade module.
• BladeManager. To make your DataBlade module available in a database.

In addition, the DataBlade Developers Kit provides the following tools for debugging your DataBlade module on Windows NT:

• DBDK Visual C++ Add-In. To debug your DataBlade module within Microsoft Visual C++.
• IfxQuery. To execute SQL debugging tests from within Microsoft Visual C++.

The DataBlade Developers Kit reduces development time because it:

• uses wizards to guide you through complex SQL object creation options.
• generates the following types of files:
• Complete SQL definitions for your database objects
• Complete code, or code entry points for C, C++, and Java source code
• Unit tests for debugging user-defined routines, opaque data type support routines, and cast support functions.
• Functional tests for validating user-defined routines, opaque data type support routines, and cast support functions
• automates creating an interactive installation program for UNIX and Windows NT operating systems.

The source code generated by the DataBlade Developers Kit follows good coding practices for your Informix database server and ensures consistency among your user-defined routines.