Overview of the Sybase appendix

Overview

The object of this annex is to present the method used to adapt generation to the Sybase V11 RDBMS. The annex is divided into the following model units :class, association and attribute. For each unit, explanations are related to the generation carried out and the optional annotations. The last section deals with the characteristics of the application part.

The generation deals especially with the mapping of constraints and types.

Annotations specific to Sybase allow RDBMS features to be used in a simple way (optimization, additional constraints).

Environment variables

The compilation of applications for the Sybase RDBMS requires the presence of the following environment variables:

· SYBASE must contain the Sybase root directory path

· PATH must contain $SYBASE/bin

· LANG must not be positioned to us

The execution of applications produced for the Sybase RDBMS requires the presence of the following environment variables:

· O_SQL_USER must contain the name of the Sybase RDBMS known password, which will be used to create relational tables. The format of this variable is as follows: name/passwd.

· LANG must not be positioned to us.

End of the SQL command

Each Transact-SQL statement must end with the "go" command (this can be renamed). This command replaces the traditional ';' in SQL.

Identifiers

The size of Sybase identifiers is limited to 30 characters. The first character should be alphabetical.

Case sensitivity

Sybase standard installation makes the SQL server "case sensitive". This has the effect of requiring that SQL generation conform with the case of the UML model classes or attributes. The "SQL upper case" parameter must not be selected in this case.

However, it is possible to make identifiers case-sensitive, by reconfiguring the "sort" order" of the SQL server (see Sybase SQL Server, Ref Manual Vol 1 page 5‑41).

You will then be free to specify the desired case option when generating the SQL.