Processing Embedded SQL Statements

Embedded SQL fully supports all Transact-SQL extensions, including stored procedures, local variables, and control-of-flow language. Due to syntax conflicts with Embedded SQL reserved keywords and because static Embedded SQL statements are compiled into stored procedures by the precompiler, minor restrictions apply to how Transact-SQL extensions are implemented. The restrictions are as follows:

• Abbreviate the Transact-SQL EXECUTE statement to EXEC to avoid conflict with the Embedded SQL EXECUTE statement.
• Avoid using Transact-SQL statement labels in static SQL statements because they conflict with the syntax for host variables. However, you can use Transact-SQL labels in dynamic SQL statements.

Note

If Microsoft SQL Server returns an error during compilation, the SQL Server error code appears in the precompiler error message as a negative number. For example:

SQL Syntax Error -SQLCODE = -207

For a list of messages that are returned by Embedded SQL, see Appendix E, "Messages."

• Because sqlprep converts all static SQL statements into stored procedures, all limitations for Transact-SQL stored procedures apply to static SQL statements. A static SQL statement that contains a single transaction-management statement (such as COMMIT or SAVEPOINT) is not compiled into a stored procedure. Instead, the statement is issued dynamically at run time.

For information about restrictions that relate to other Embedded SQL statements, see Chapter 3, "Embedded SQL Statements."