eXtremeDB Case Sensitivity

When sorting ASCII strings, the alphabetic order of strings is by default case-sensitive. If it is important to order rows by case-insensitive character comparison, i.e. where the lowercase and uppercase letters in the same position are sorted equally, the case_insensitive or ignore_case keywords can be used on index definitions in the database schema. For example, to specify that index by_name on string filed name will use case-insensitive comparison, use a declaration like the following in the schema file:

 
    tree<name case_insensitive asc> by_name;
     

or

 
    tree<name ignore_case asc> by_name;
     

Also, for eXtremeDB source code licensees, the MCO_CFG_SUPPORT_CASE_INSENSITIVE_CHAR_CMP configuration option is available as a C preprocessor define in target/mcolib/mcocfg.h . Its default value is 1. If its value is changed to 0 in an eXtremeDB build, then all indexes will use case sensitive comparison no matter whether they are declared with the case_insensitive keyword in the schema file or not; i.e. this redefinition of the MCO_CFG_SUPPORT_CASE_INSENSITIVE_CHAR_CMP constant will override the effects of case_insensitive keyword.

To be clear, the following settings will have the stated results:

Value of MCO_CFG_SUPPORT_CASE _INSENSITIVE_CHAR_CMP	Schema declaration	Resulting index comparison
defined as 1 (default)	without case_insensitive	case sensitive
defined as 1 (default)	with case_insensitive	case insensitive
redefined as 0 (special build)	without case_insensitive	case sensitive
redefined as 0 (special build)	with case_insensitive	case sensitive

(Object code licensees can request a special build of eXtremeDB libraries from McObject Support.)