By default, the FormatColumn property is empty. The cells in the column use the provided format only if is valid ( not empty, and syntactically correct ), to display data in the column. The FormatColumn property provides a format to display all cells in the column using a predefined format. The expression may be a combination of variables, constants, strings, dates and operators, and value. The value operator gives the value to be formatted.  A string is delimited by ", ` or ' characters, and inside they can have the starting character preceded by \ character, ie "\"This is a quote\"". A date is delimited by # character, ie #1/31/2001 10:00# means the January 31th, 2001, 10:00 AM. The cell's HTML format is applied only if the CellCaptionFormat or Def(exCellCaptionFormat) is exHTML. If valid, the FormatColumn is applied to all cells for which the CellCaptionFormat property is not exComputedField. This way you can specify which cells use or not the FormatColumn property. The ComputedField property indicates the formula of the computed column.

For instance:

• the "currency(value)" displays the column using the current format for the currency ie, 1000 gets displayed as $1,000.00
• the "longdate(date(value))" converts the value to a date and gets the long format to display the date in the column, ie #1/1/2001# displays instead Monday, January 01, 2001
• the "'<b>' + ((0:=proper(value)) left 1) + '</b>' + (=:0 mid 2)" converts the name to proper, so the first letter is capitalized, bolds the first character, and let unchanged the rest, ie a "mihai filimon" gets displayed "Mihai Filimon".
• the "len(value) ? ((0:=dbl(value)) < 10 ? '<fgcolor=808080><font ;7>' : '<b>') + currency(=:0)" displays the cells that contains not empty daya, the value in currency format, with a different font and color for values less than 10, and bolded for those that are greater than 10, as can see in the following screen shot in the column (A+B+C):

The value keyword in the FormatColumn property indicates the value to be formatted. 

The supported binary arithmetic operators are:

• * ( multiplicity operator ), priority 5
• / ( divide operator ), priority 5
• mod ( reminder operator ), priority 5
• + ( addition operator ), priority 4 ( concatenates two strings, if one of the operands is of string type )
• - ( subtraction operator ), priority 4

The supported unary boolean operators are:

• not ( not operator ), priority 3 ( high priority )

The supported binary boolean operators are:

• or ( or operator ), priority 2
• and ( or operator ), priority 1

The supported binary boolean operators, all these with the same priority 0, are :

• < ( less operator )
• <= ( less or equal operator )
• = ( equal operator )
• != ( not equal operator )
• >= ( greater or equal operator )
• > ( greater operator )

The supported ternary operators, all these with the same priority 0, are :

• ? ( Immediate If operator ), returns and executes one of two expressions, depending on the evaluation of an expression. The syntax for is 

"expression ? true_part : false_part"

, while it executes and returns the true_part if the expression is true, else it executes and returns the false_part. For instance, the "%0 = 1 ? 'One' : (%0 = 2 ? 'Two' : 'not found')" returns 'One' if the value is 1, 'Two' if the value is 2, and 'not found' for any other value. A n-ary equivalent operation is the case() statement, which is available in newer versions of the component.

The supported n-ary operators are (with priority 5):

• in (include operator), specifies whether an element is found in a set of constant elements. The in operator returns -1 ( True ) if the element is found, else 0 (false) is retrieved. The syntax for in operator is 

"expression in (c1,c2,c3,...cn)"

, where the c1, c2, ... are constant elements. The constant elements could be numeric, date or string expressions. For instance the "value in (11,22,33,44,13)" is equivalent with "(expression = 11) or (expression = 22) or (expression = 33) or (expression = 44) or (expression = 13)". The in operator is not a time consuming as the equivalent or version is, so when you have large number of constant elements it is recommended using the in operator. Shortly, if the collection of elements has 1000 elements the in operator could take up to 8 operations in order to find if an element fits the set, else if the or statement is used, it could take up to 1000 operations to check, so by far, the in operator could save time on finding elements within a collection.

• switch (switch operator), returns the value being found in the collection, or a predefined value if the element is not found (default). The syntax for switch operator is 

"expression switch (default,c1,c2,c3,...,cn)"

, where the c1, c2, ... are constant elements, and the default is a constant element being returned when the element is not found in the collection. The constant elements could be numeric, date or string expressions.  The equivalent syntax is "%0 = c 1 ? c 1 : ( %0 = c 2 ? c 2 : ( ... ? . : default) )". The switch operator is very similar with the in operator excepts that the first element in the switch is always returned by the statement if the element is not found,  while the returned value is the value itself instead -1. For instance, the "%0 switch ('not found',1,4,7,9,11)" gets 1, 4, 7, 9 or 11, or 'not found' for any other value. As the in operator the switch operator uses binary searches for fitting the element, so it is quicker that iif (immediate if operator) alterative.

• case() (case operator) returns and executes one of n expressions, depending on the evaluation of the expression ( IIF - immediate IF operator is a binary case() operator ). The syntax for case() operator is:

"expression case ([default : default_expression ; ] c1 : expression1 ; c2 : expression2 ; c3 : expression3 ;....)"

If the default part is missing, the case() operator returns the value of the expression if it is not found in the collection of cases ( c1, c2, ...). For instance, if the value of expression is not any of c1, c2, .... the default_expression is executed and returned. If the value of the expression is c1, then the case() operator executes and returns the expression1. The default, c1, c2, c3, ... must be constant elements as numbers, dates or strings. For instance, the "date(shortdate(value)) case (default:0 ; #1/1/2002#:1 ; #2/1/2002#:1; #4/1/2002#:1; #5/1/2002#:1)" indicates that only #1/1/2002#, #2/1/2002#,  #4/1/2002# and  #5/1/2002# dates returns 1, since the others returns 0. For instance the following sample specifies the hour being non-working for specified dates: "date(shortdate(value)) case(default:0;#4/1/2009# : hour(value) >= 6 and hour(value) <= 12 ; #4/5/2009# : hour(value) >= 7 and hour(value) <= 10 or hour(value) in(15,16,18,22); #5/1/2009# : hour(value) <= 8)" statement indicates the working hours for dates as follows:

• • #4/1/2009#, from hours 06:00 AM to 12:00 PM
  • #4/5/2009#, from hours 07:00 AM to 10:00 AM and hours 03:00PM, 04:00PM, 06:00PM and 10:00PM
  • #5/1/2009#, from hours 12:00 AM to 08:00 AM

The in, switch and case() use binary search to look for elements so they are faster then using iif and or expressions.

Obviously, the priority of the operations inside the expression is determined by ( ) parenthesis and the priority for each operator. 

The supported conversion unary operators are:

• type (unary operator) retrieves the type of the object. For instance type(%0) = 8 specifies the cells that contains string values. 

  Here's few predefined types:

  • 0 - empty ( not initialized )
  • 1 - null
  • 2 - short
  • 3 - long
  • 4 - float
  • 5 - double
  • 6 - currency
  • 7 - date
  • 8 - string
  • 9 - object
  • 10 - error
  • 11 - boolean
  • 12 - variant
  • 13 - any
  • 14 - decimal
  • 16 - char
  • 17 - byte
  • 18 - unsigned short 
  • 19 - unsigned long
  • 20 - long on 64 bits 
  • 21 - unsigned long on 64 bites
• str (unary operator) converts the expression to a string
• dbl (unary operator) converts the expression to a number
• date (unary operator) converts the expression to a date

Other known operators for numbers are:

• int (unary operator) retrieves the integer part of the number
• round (unary operator) rounds the number ie 1.2 gets 1, since 1.8 gets 2
• floor (unary operator) returns the largest number with no fraction part that is not greater than the value of its argument
• abs (unary operator) retrieves the absolute part of the number ie -1 gets 1, 2 gets 2
• value format 'flags' (binary operator) formats the value with specified flags. If flags is empty, the number is displayed as shown in the field "Number" in the "Regional and Language Options" from the Control Panel. For instance the 1000 format '' displays 1,000.00 for English format, while 1.000,00 is displayed for German format. 1000 format '2|.|3|,' will always displays 1,000.00 no matter of settings in the control panel. If formatting the number fails for some invalid parameter, the value is displayed with no formatting. 

  The ' flags' for format operator is a list of values separated by | character such as 'NumDigits|DecimalSep|Grouping|ThousandSep|NegativeOrder|LeadingZero' with the following meanings: 

  • NumDigits - specifies the number of fractional digits, If the flag is missing, the field "No. of digits after decimal" from "Regional and Language Options" is using.
  • DecimalSep - specifies the decimal separator. If the flag is missing, the field "Decimal symbol" from "Regional and Language Options" is using.
  • Grouping - indicates the number of digits in each group of numbers to the left of the decimal separator. Values in the range 0 through 9 and 32 are valid. The most significant grouping digit indicates the number of digits in the least significant group immediately to the left of the decimal separator. Each subsequent grouping digit indicates the next significant group of digits to the left of the previous group. If the last value supplied is not 0, the remaining groups repeat the last group. Typical examples of settings for this member are: 0 to group digits as in 123456789.00; 3 to group digits as in 123,456,789.00; and 32 to group digits as in 12,34,56,789.00. If the flag is missing, the field "Digit grouping" from "Regional and Language Options" indicates the grouping flag.
  • ThousandSep - specifies the thousand separator. If the flag is missing, the field "Digit grouping symbol" from "Regional and Language Options" is using.
  • NegativeOrder - indicates the negative number mode. If the flag is missing, the field "Negative number format" from "Regional and Language Options" is using. The valid values are 0, 1, 2, 3 and 4 with the following meanings:
    • 0 - Left parenthesis, number, right parenthesis; for example, (1.1) 
    • 1 - Negative sign, number; for example, -1.1
    • 2 - Negative sign, space, number; for example, - 1.1
    • 3 - Number, negative sign; for example, 1.1-
    • 4 - Number, space, negative sign; for example, 1.1 -
  • LeadingZero - indicates if leading zeros should be used in decimal fields.  If the flag is missing, the field "Display leading zeros" from "Regional and Language Options" is using. The valid values are 0, 1

Other known operators for strings are:

• len (unary operator) retrieves the number of characters in the string
• lower (unary operator) returns a string expression in lowercase letters
• upper (unary operator) returns a string expression in uppercase letters
• proper (unary operator) returns from a character expression a string capitalized as appropriate for proper names
• ltrim (unary operator) removes spaces on the left side of a string
• rtrim (unary operator) removes spaces on the right side of a string
• trim (unary operator) removes spaces on both sides of a string
• startwith (binary operator) specifies whether a string starts with specified string
• endwith (binary operator) specifies whether a string ends with specified string
• contains (binary operator) specifies whether a string contains another specified string
• left (binary operator) retrieves the left part of the string
• right (binary operator) retrieves the right part of the string
• a mid b (binary operator) retrieves the middle part of the string a starting from b ( 1 means first position, and so on )
• a count b (binary operator) retrieves the number of occurrences of the b in a
• a replace b with c (double binary operator) replaces in a the b with c, and gets the result.

Other known operators for dates are:

• time (unary operator) retrieves the time of the date in string format, as specified in the control's panel.
• timeF (unary operator) retrieves the time of the date in string format, as "HH:MM:SS". For instance the timeF(1:23 PM) returns "13:23:00"
• shortdate (unary operator) formats a date as a date string using the short date format, as specified in the control's panel.
• shortdateF (unary operator) formats a date as a date string using the "MM/DD/YYYY" format. For instance the shortdateF(December 31, 1971 11:00 AM) returns "12/31/1971".
• dateF (unary operator) converts the date expression to a string expression in "MM/DD/YYYY HH:MM:SS" format.
• longdate (unary operator) formats a date as a date string using the long date format, as specified in the control's panel.
• year (unary operator) retrieves the year of the date (100,...,9999)
• month (unary operator) retrieves the month of the date ( 1, 2,...,12 )
• day (unary operator) retrieves the day of the date ( 1, 2,...,31 )
• yearday (unary operator) retrieves the number of the day in the year, or the days since January 1st ( 0, 1,...,365 )
• weekday (unary operator) retrieves the number of days since Sunday ( 0 - Sunday, 1 - Monday,..., 6 - Saturday )
• hour (unary operator) retrieves the hour of the date ( 0, 1, ..., 23 )
• min (unary operator) retrieves the minute of the date ( 0, 1, ..., 59 )
• sec (unary operator) retrieves the second of the date ( 0, 1, ..., 59 )

The expression supports also immediate if ( similar with iif in visual basic, or ? : in C++ ) ie cond ? value_true : value_false, which means that once that cond is true the value_true is used, else the value_false is used. Also, it supports variables, up to 10 from 0 to 9. For instance, 0:="Abc" means that in the variable 0 is "Abc", and =:0 means retrieves the value of the variable 0. You can use variables to avoid computing several times the same thing.