Addendum Clipper documentation. Original transform docs are too limited :

/* TODO: Make a real document out of this */

NUMBERS

The following functions mean something :

@X      DB after negative numbers
@(      Quotes around negative numbers
@)      Quotes around negative numbers without leading spaces
@B      Left justified
@C      CR after positive numbers
@E      Exchange . and , (Not in US/UK manual, does work !)
@Z      Return spaces if value is 0

New addition :
@0      Make a zero padded string out of the number.
@L      Same of previous.

The following templates mean something to numbers :
9       Digit
#       Digit
.       Dot
$       Pad left with $
*       Pad left with *
,       Comma   (Real pathetic. Just like Clipper)

The following bugs have been fixed : (Just like Xbase++ did !)
@(   Does not give incorrect results when it should overflow
@D   Ignored. So it should be !

The default string is not properly formatted :
Decimals issue currently being debated. Besides it is a STR problem.

LOGICALS

The following functions mean something :

@R      Show remainder

The following templates mean something :

Y       Y/N
#       T/F (Also isn't the NG)
L       T/F

STRINGS

The following functions mean something :
@!      Upper case
@R      Show remainder

The following templates mean something :
!       Upper
# 9     Digit
A N X   Text

DATES

Uses SET DATE FORMAT
Uses SET CENTURY

The following functions mean something :
@E      Use British date format
        /* QUESTION :
                Clipper appears to reverse the MM and DD parts of the
                date, regardless of SET DATE FORMAT, because if the
                date format is British, using @E displays the date in
                American. Harbour always treats @E as British.

                So is Harbour's behaviour a bug fix or a bug?
        */