Below, you will find three date format functions, all of which make some use of Unix timestamps, a concept most ASP developers will not be familiar with. Unix timestamps are defined as the number of seconds since the epoch, 00:00:00 01/01/1970, and are a very sensible way to store dates and times in a database, in my opinion. They are very easy to work with, especially when selecting data from a database, and once you are used to them, they can make life much much simpler.

NOTE: These functions were updated on the 15th March 2004 to fix a couple of known bugs. In addition, any scripts using these functions may need to be tweaked to take account of your locale. Each locale will use a slightly different date and time format, so please check yours before using these functions, or simply set your locale to UK. To set your locale to UK, you can use the code below:

<%@LANGUAGE=VBSCRIPT" LCID=2057%>

There is also a [url=http://www.microsoft.com/globaldev/reference/lcid-all.mspx]complete list of valid locales[/url] on the Microsoft website, with the values in the right hand column being the ones you will want.

Finally, the "monthname" function is VBScript is not present in every version, so may need to be defined manually. Please see comments on this article for an example.

The first function, UDate(), will convert a normal date into a Unix timestamp. The second function, unUDate(), simply does the reverse, converting any timestamp into a readable date. The last function takes a date and a format string, and returns a formatted date according to what you put into the formatting string.

UDate()

function UDate(oldDate)
    UDate = DateDiff("s", "01/01/1970 00:00:00", oldDate)
end function

In practice, UDate is very easy to use. For example, if you want to know the current Unix time (the number of seconds since the epoch), you would use something like the following, which would give intCurrent_Unix_Time a value of 1069916571 at the time of writing.

intCurrent_Unix_Time = UDate(Now())

unUDate()

function unUDate(intTimeStamp)
    unUDate = DateAdd("s", intTimeStamp, "01/01/1970 00:00:00")
end function

unUDate is as easy to use as UDate. Simply feed in an integer, and it will return a matching date. It will treat the integer as the number of seconds between the epoch and the desired date, and will return a properly formatted date for normal use in VBScript accordingly.

formatDate()

Now here's where it starts to get tricky. formatDate is also quite easy to use, but only once you get the hang of how it works. As you can see from the below, formatDate requires two values to be passed to it. The first, format, is a string telling the function how you would like the date formatted, and the second, intTimeStamp, can be either a Unix timestamp (preferred) or a normal VBScript date.

The format of the string is what we want to make use of to lay out our dates exactly as we would like, and perhaps this is best shown with an example. I would like a date displayed on my site in the following format: "7:11am, Thursday 27th November, 2003". In normal VBScript that would be a pain to format - I'd probably just end up giving up and using the normal VBLongTime and VBLongDate to display something similar. I would prefer complete control though...

The format string for formatDate can include a whole selection of characters to represent certain date or time entities. There's a table of the ones included in this function just below here. The format string needs to include these entities where it wants them to be replaced by the correct time or date value. To get the above formatted date and time, then, I could use the following:

strDateTime = formatDate("%g:%i%a, %l %j%O %F, %Y", UDate(Now()))

That looks a lot harder than it actually is. To tell the function we are aiming to replace a specific entity in the string with a time or date element, we use a symbol made up of a percent sign (%) and a letter. It is a case sensitive function, so be careful you form your format strings correctly. Without further ado, here's a list of the entities you can include in the format string (anything in the string that is not an entity from the list below will still be in the string when it has been processed).

• %A - AM or PM
• %a - am or pm
• %m - Month with leading zeroes (01 - 12)
• %n - Month without leading zeroes (1 - 12)
• %F - Month name (January - December)
• %M - Three letter month name (Jan - Dec)
• $d - Day with leading zeroes (01 - 31)
• %j - Day without leading zeroes (1 - 31)
• %H - Hour with leading zeroes (12 hour)
• %h - Hour with leading zeroes (24 hour)
• %G - Hour without leading zeroes (12 hour)
• %g - Hour without leading zeroes (24 hour)
• %i - Minute with leading zeroes (01 to 60)
• %I - Minute without leading zeroes (1 to 60)
• %s - Second with leading zeroes (01 to 60)
• %S - Second without leading zeroes (1 to 60)
• %L - Number of day of week (1 to 7)
• %l - Name of day of week (Sunday to Saturday)
• %D - Three letter name of day of week (Sun to Sat)
• %O - Ordinal suffix (st, nd rd, th)
• %U - UNIX Timestamp
• %Y - Four digit year (2003)
• %y - Two digit year (03)

You can include any of the above in the format string at any point. I would recommend not using a percentage sign if you can help it (except obviously as part of one of the elements above), and in the same vein, using ordinals anywhere except after a number is slightly foolish.

If you would like to see anything else added to the list, above, please email me and I will see what I can do.

The below is the formatDate function itself. To use, simply copy and paste it into a script or include file and call it as above, with whichever format string you require.

function formatDate(format, intTimeStamp)
    dim unUDate, A

    ' Test to see if intTimeStamp looks valid. If not, they have passed a normal date
    if not (isnumeric(intTimeStamp)) then
        if isdate(intTimeStamp) then
            intTimeStamp = DateDiff("S", "01/01/1970 00:00:00", intTimeStamp)
        else
            response.write "Date Invalid"
            exit function
        end if
    end if
    
    if (intTimeStamp=0) then 
        unUDate = now()
    else
        unUDate = DateAdd("s", intTimeStamp, "01/01/1970 00:00:00")
    end if

    unUDate = trim(unUDate)

    dim startM : startM = InStr(1, unUDate, "/", vbTextCompare) + 1
    dim startY : startY = InStr(startM, unUDate, "/", vbTextCompare) + 1
    dim startHour : startHour = InStr(startY, unUDate, " ", vbTextCompare) + 1
    dim startMin : startMin = InStr(startHour, unUDate, ":", vbTextCompare) + 1

    dim dateDay : dateDay = mid(unUDate, 1, 2)
    dim dateMonth : dateMonth = mid(unUDate, startM, 2)
    dim dateYear : dateYear = mid(unUDate, startY, 4)
    dim dateHour : dateHour = mid(unUDate, startHour, 2)
    dim dateMinute : dateMinute = mid(unUDate, startMin, 2)
    dim dateSecond : dateSecond = mid(unUDate, InStr(startMin, unUDate, ":", vbTextCompare) + 1, 2)

    format = replace(format, "%Y", right(dateYear, 4))
    format = replace(format, "%y", right(dateYear, 2))
    format = replace(format, "%m", dateMonth)
    format = replace(format, "%n", cint(dateMonth))
    format = replace(format, "%F", monthname(cint(dateMonth)))
    format = replace(format, "%M", left(monthname(cint(dateMonth)), 3))
    format = replace(format, "%d", dateDay)
    format = replace(format, "%j", cint(dateDay))
    format = replace(format, "%h", mid(unUDate, startHour, 2))
    format = replace(format, "%g", cint(mid(unUDate, startHour, 2)))

    if (cint(dateHour) > 12) then
        A = "PM"
    else
        A = "AM"
    end if
    format = replace(format, "%A", A)
    format = replace(format, "%a", lcase(A))

    if (A = "PM") then format = replace(format, "%H", left("0" & dateHour - 12, 2))
    format = replace(format, "%H", dateHour)
    if (A = "PM") then format = replace(format, "%G", left("0" & cint(dateHour) - 12, 2))
    format = replace(format, "%G", cint(dateHour))

    format = replace(format, "%i", dateMinute)
    format = replace(format, "%I", cint(dateMinute))
    format = replace(format, "%s", dateSecond)
    format = replace(format, "%S", cint(dateSecond))
    format = replace(format, "%L", WeekDay(unUDate))
    format = replace(format, "%D", left(WeekDayName(WeekDay(unUDate)), 3))
    format = replace(format, "%l", WeekDayName(WeekDay(unUDate)))
    format = replace(format, "%U", intTimeStamp)
    format = replace(format, "11%O", "11th")
    format = replace(format, "1%O", "1st")
    format = replace(format, "12%O", "12th")
    format = replace(format, "2%O", "2nd")
    format = replace(format, "13%O", "13th")
    format = replace(format, "3%O", "3rd")
    format = replace(format, "%O", "th")

    formatDate = format

end function

Below, you will find three sections. The first section, Reference, is a simple reference listing the most-used of the various symbols and characters used in regular expressions. The second section, Functions, has two functions in it that may make life easier for you. The third section, Examples, is where the fun begins - examples of regular expressions in action.

Reference

Character Sets and Grouping

• . - Any single character (except new line character, "\n")
• [] - Encloses any set of characters
• ^ - Matches any characters not within following set
• [A-Z] - Any upper case letter between A and Z
• [a-z] - Any lower case letter between a and z
• [0-9] - Any digit from 0 to 9
• () - Group section. Also can then be back-referenced with $1 to $n, where n is the number of groups
• | - Or. (ab)|(bc) will match "ab" or "bc"

Repetition

• + - One or more
• * - Zero or more
• ? - Zero or one
• {5} - Five
• {1,3} - One to three
• {2,} - Two or more

Positioning

• ^ - Start of string
• $ - End of string
• \b - End of word
• \n - New line
• \r - Carriage return

Miscellaneous

• \ - Escape character
• \t - Tab
• \s - White space
• \w - Matches word (equivalent of [A-Za-z0-9_])

Please note that the escape character mentioned above is not usable in normal VBScript. Regular expression syntax is based upon Perl regular expression syntax. To escape a character in VBScript, you usually double it. For example, the following will print out 'This is a "quoted" piece of text'.

Functions

The first of the functions below, ereg (named after the PHP function to keep me from going quite quite mad), is the one you will probably use most. Simply put, if you feed in a string, pattern, and choose whether or not you would like to ignore the case of letters in either, the function will return TRUE if the string contains the pattern, or FALSE if not.

Next up we have ereg_replace. Like it's shorter cousin, you need to feed it a string, a pattern and choose your case sensitivity. This time, you must also add a replacement. This function will replace all instances of the pattern with the replacement in the string (if you change ".Global = True" to ".Global = False" then the function will only replace the first instance of the pattern with the replacement).

Examples

Example 1: Checking hexadecimal string

A hexadecimal number can be made up of any digit, and any letter, upper or lower case, between a and f, inclusive. So to check if a string is actually hexadecimal, the following will do quite nicely (strOriginalString is the original string to be tested):

The pattern, "[^a-f0-9\s]" matches anything that is not in the set of characters specified (so if there is anything in the string that is not in that set, the function will return True). The characters specified are all letters between a and f inclusive, and we've specified a case insensitive match, so upper case letters will be treated the same way. We are also allowing whitespace (new lines, spaces, carriage returns and tabs), which is what the "\s" represents in regular expressions.

Example string that returns False (and is therefore hexadecimal):

Example 2: Masking the last section of an IP address

An IP address is made up of four sets of numbers seperated by periods. It's common practice, if you are going to display visitor (or any) IP address on your site, to mask the last (fourth) set of numbers. Here's a way to use ereg_replace to do just this:

This is a little more tricky, as you'd hopefully expect from a second example. It looks harder than it is though, so one step at a time. There are actually only a few entities in the pattern - they are just repeated. The most important is this: "([0-9]{1,3})". It matches a section of an IP adress, and is enclosed in brackets so that this section can be used in the replacement of the pattern as well (otherwise we would not be able to keep the first three parts of the IP address to display). You can see these sections in use, referenced with "$2", "$3" and "$4" in the replacement. The pattern within the brackets simply says "between one and three digits between 0 and 9".

The second repeated section is "\.". We use a backslash before the period to indicate that this period (the character following the backslash) is to be treated as a normal period. We call this an escaped character, and this is a fairly common practice. The period, unescaped (without the backslash), is used as a symbol representing "any character except the new line character".

Example input text:

Example output text:

Example 3: Making the second word of every sentence in a string bold, as long as the word before only contains upper case letters and the second word does not contain an even digit

Getting more interesting now, this example is not in the least bit useful in practice, but should prove to be a useful demonstration of the power of regular expressions. It sounds tough - but with regular expressions, it's a walk in the park.

We start by adding an artificial period and space to the beginning of the string, just to make sure we catch the first sentence, and add a line to strip our extra characters out afterwards. We only want those sentences split with punctuation and a space, or we'll end up with bold decimals and it will be very messy indeed. So, we check for puncuation, followed by a space, followed by a word made entirely of capitals, followed by another space, followed by a second word that doesn't contain even numbers, or whitespace, followed by a space. If we find that, we replace it with the same items we picked up in brackets, only with a <strong></strong> tag pair around the second word.

Example input text:

Example output text: