module StringUtil where
	import Data.Char

	{-
	 - doTokenise:
	 -
	 - Performs the operation of the tokenise function, using a list which increases in size as it
	 - reads in the current token (cTok) and a sliding window buffer which checks to see whether or
	 - not the last (separator size) characters match the separator, and if they do the end of the
	 - current token has been reached.  Called by tokenise.
	 -}

	doTokenise sep [] cTok cBuffer = 
		if ((length cTok) > 0 && cTok /= sep) then
			if (cBuffer /= sep) then
				[cTok++cBuffer]
			else
				[cTok]
		else
			if (cBuffer /= sep) then
				[cBuffer]
			else
				[""]
	doTokenise sep (x:xs) cTok cBuffer =
		if (cBuffer == sep) then
			cTok:(doTokenise sep (x:xs) "" [])
		else
			if ((length cBuffer) == (length sep)) then
				doTokenise sep xs (cTok++[head cBuffer]) ((tail cBuffer)++[x])
			else
				doTokenise sep xs "" (cBuffer++[x])

	{-
	 - tokenise:
	 -
	 - Returns a list of the tokens in oString separated by sep.  For example, with the arguments:
	 -
	 - tokenise "--" "One--Two--Three--Four"
	 -
	 - the return value would be
	 -
	 - ["One", "Two", "Three", "Four"]
	 -}

	tokenise sep oString = doTokenise sep oString [] []

	{-
	 - doGetFieldType:
	 - 
	 - Performs the field extraction, called by getFieldType.
	 -}

	doGetFieldType [] current = current
	doGetFieldType (x:xs) current = 
		if (not (isDigit x)) then
			if (x /= '.') then
				"String"
			else
				doGetFieldType xs "Float"
		else
			if (current /= "Float") then
				doGetFieldType xs "Int"
			else
				doGetFieldType xs "Float"

	{-
	 - getFieldType:
	 -
	 - "Brute force" extraction of basic field type.  Starts off doGetFieldType with the type "Int"
	 - as it is the type with the fewest possible number of valid symbols.
	 -}

	getFieldType x = doGetFieldType x "Int"

	{-
	 - nonAlphaNumericSub:
	 -
	 - Takes a character and if it is an invalid non-alphanumeric character (anything but a letter,
	 - a digit or an underscore) replaces it with a string representation.
	 -}

	nonAlphaNumericSub x =
		case x of
			'`' -> "Grave"
			'~' -> "Tilde"
			'!' -> "Exclamation"
			'@' -> "At"
			'#' -> "Hash"
			'$' -> "Dollar"
			'%' -> "Percentage"
			'^' -> "Carat"
			'&' -> "Ampersand"
			'*' -> "Asterisk"
			'(' -> "OpenBracket"
			')' -> "CloseBracket"
			'-' -> "Minus"
			'+' -> "Plus"
			'=' -> "Equals"
			'{' -> "OpenBrace"
			'}' -> "CloseBrace"
			'[' -> "OpenSquare"
			']' -> "CloseSquare"
			'|' -> "Pipe"
			'\\' -> "Backslash"
			':' -> "Colon"
			';' -> "Semicolon"
			'"' -> "Quotation"
			'\'' -> "Apostrophe"
			',' -> "Comma"
			'<' -> "Lessthan"
			'.' -> "Fullstop"
			'>' -> "Greaterthan"
			'?' -> "Question"
			'/' -> "Slash"
			_ -> [x]