Programming the Perl DBI

The Significance of Case

You may have noticed that some attribute names use all uppercase letters, like NUM_OF_FIELDS, while others use mixed case letters, like RaiseError. If you’ve seen any descriptions of individual database drivers you may have also noticed some attribute names that use all lowercase letters, like ado_conn and ora_type.

There is a serious method behind the apparently inconsistent madness. The letter case used for attribute names is significant and plays an important part in the portability of DBI scripts and the extensibility of the DBI itself. The letter case of the attribute name is used to signify who defined the meaning of that name and its values, as follows:

UPPER_CASE

Attribute names that use only uppercase letters and underscores are defined by external standards, such as ISO SQL or ODBC.

The statement handle TYPE attribute is a good example here. It’s an uppercase attribute because the values it returns are the standard portable datatype numbers defined by ISO SQL and ODBC, and not the nonportable native database datatype numbers.

MixedCase

Attribute names that start with an uppercase letter but include lowercase letters are defined by the DBI specification.

lower_case

Attribute names that start with a lowercase letters are defined by individual database drivers. These are known as driver-specific attributes.

Because the meanings are assigned by driver authors without any central control, it’s important that two driver authors don’t pick the same name ...