This appendix lists the settings required to operate in compliance with ANSI/ISO Entry SQL-92 and the corresponding Ingres defaults.
To determine the setting in effect, use dbmsinfo. For details about dbmsinfo, see Status Information.
Case sensitivity for delimited identifiers is specified when the database is created. For details about createdb, see the Command Reference Guide and the System Administrator Guide.
Query flattening and default cursor mode options are specified when the DBMS Server is configured and started.
If Configuration-By-Forms (cbf) is used to configure the installation, select the Preferences menu item and set the ISO_ENTRY_SQL-92 parameter to ON. This setting ensures ANSI/ISO-compliant behavior for the following areas:
The ANSI Entry SQL-92 standard specifies how case is handled for identifiers (names of tables, for example). Identifiers can be quoted or unquoted (regular or delimited).
Regular identifiers are treated as upper case:
dbsminfo('db_name_case') returns UPPER.
Regular identifiers are treated as lower case:
dbsminfo('db_name_case') returns LOWER.
Delimited identifiers are case-sensitive:
dbsminfo('db_delimited_case') returns MIXED.
Delimited identifiers are not case-sensitive:
dbsminfo('db_delimited_case') returns LOWER.
User names are treated as upper case:
dbsminfo('db_real_user_case') returns UPPER.
User names are treated as lower case:
dbsminfo('db_real_user_case') returns LOWER.
The ANSI Entry SQL-92 standard specifies the default mode for cursors. For details, see Data Manipulation with Cursors.
Direct mode:
dbsminfo('db_direct_update') returns Y
dbsminfo('db_deferred_update') returns N
Deferred mode:
dbsminfo('db_direct_update') returns N
dbsminfo('db_deferred_update') returns Y
The ANSI Entry SQL-92 standard specifies that query flattening is not performed. By default, Ingres uses query flattening to improve the performance of queries.
No Flattening:
dbsminfo('flatten_singleton') returns N
dbsminfo('flatten_aggregate') returns N
Flattening used to optimize queries, including queries involving aggregate subselects or singleton subselects:
dbsminfo('flatten_singleton') returns Y
dbsminfo('flatten_aggregate') returns Y