TYPE u may be combined with date/time and numeric fields to force the relevant formatting options to be determined by the current Language Definition File (LDF). See the system parameter Language for establishing the LDF, and GTLANG for how to query it.

In the case of date fields, u must precede the D or d code, and will determine the order of the sub-fields (DDMM{CC}YY vs MMDD{CC}YY). Without the u code, D and d fields would always use the MMDD{CC}YY format. And even with u, U fields will always use DDMM{CC}YY.

Note that the display/editing format and the return formats are normally the same, so any application using u must be prepared to deal with the possibility of the date being in either format, depending on the LDF. To avoid that confusion, use of the > code is highly recommended along with u; this allows the application to always deal with dates in the {CC}YYMMDD format, regardless of how the user sees and edits them.

Note that in the case of the Windows date picker control (used for dates when the |G code is specified), u will affect the format of the dates as seen by the application, but it will be the Windows locale (regional settings) that determines how the date picker control will present the date to the user for editing.

For numeric fields (#, H, $), u forces INFLD to use the decimal point and thousands separator characters specified in the LDF, rather than the USA default characters ("." and "," respectively). In this case, the u does not have to precede the numeric type code.

Note that this will only have an effect if the field would otherwise allow entry or display of decimals and thousands separators. See type codes . (period), M, $, H as well as the defpt and maxpt parameters.

When specified prior to either of the date codes (d or D), u forces the editing format (as well as the return format, unless the > code is specified) to match current language definition file specification. As mentioned above, the > code is highly recommended whenever u is specified.