Chapter 6. Building GNADE on different platforms

Table of Contents
Installation of the development snapshot on a Unix like systems
Building the development snapshot on Windows NT/XP

Installation of the development snapshot on a Unix like systems

After you obtained the source code from the net you need to install and compile it. This chapter describes this first steps of installing the environment onto your system.

Unpacking the distribution

The source code is normally distributed as compressed tar file. To unpack the distribution execute the command:

      gunzip -c gnade-src-x.y.z.tar.gz | tar xvf -
   
This will unpack the directory tree of the development environment.

Configuring the GNADE installation

The GNADE environment may be configured to a certain extend. The file make.conf.in contains some parameters which might be adopted to the needs of your system.

After unpacking the distribution change into the top level directory of the GNADE release. Before you run the configure script examine the contents of the file etc/config.site.

site_bindir="/usr/bin"
site_libdir="/usr/lib"
site_adadir="/usr/lib/ada"
   

Because system wide installation of GNADE depends on the type of your system please adopt the following parameters in make.conf.in before executing the configure script as shown below.


       cd gnade-src-....
       ./configure [ --sampledb=<database> ] [ <native> ]      

   
The configure script allowes you to by means of the --sampledb to create a test data base. Additionaly you may give a list of RDBMS producs for which you like to build native bindings. If nothing is given, only the ODBC interfaces relevant part of GNADE will be build. For the supported data base products please check the README file in the gnade directory. If you don't have one of the supported data bases on your system then omit the database. As a result the samples code will be compiled except the code for native bindings, but the sample data base will not be available.

Preparation of the test data base

In order to allow the installation of the test database, most of the commonly known dbcs's require a data base user to be installed. This normally required certain DBA privileges. There for this step is expected to be done manual as shown below (The name of the user, the name of the data base is specified in make.conf.in).


      su <dbcs root> 
      gmake createuser

   
The user may be deleted by the command make removeuser.

In order to test the functionality of the data base you may create the test data base already at this point by the following commands:

      gmake removedb 
      gmake createdb
   
This will create a database gnade which contains at least the table EMPLOYEES which may be checked manually.

Mimer SQL: In case of Mimer SQL the user is created as root, but the make createdb command has to be executed as the same user which is used to run the test examples. If thus is not done, the examples will fail!

Compiling

To build the GNADE executable enter the command below:

   gmake all
   
This will build all components of the GNADE project and the test data base is this has not been done previously.

Installing GNADE globally on the system

The development environment is self containing, which means as long as applications are developed in the directory where GNADE is installed and the make files are used, all components are taken from the GNADE lib directory. This method limits the use to one user. In order to make GNADE available to all users on your system you need to install the GNADE libraries. Installation is done as root by executing the directory the following command:

      gmake install
   

This should install the libraries of the GNADE project in your system.

Integration with IDE Products

The GNADE environment provides limited support for the GPS product of ACT by integrating this document into the GPS help system. In order to integrate with GPS, the environment variable GPS_DOC_PATH has to include the installation directory of GNADE (e.g. /usr/local/gnade/doc).