如何从字符串中过滤状态缩写

I have XML where the /forms/form/@number can be in almost any format (any combination of letters, numbers and single spaces). Occasionally there will be @number values which are the same form...but are state specific. When this happens the last 3 characters will be a single space followed by the state abbreviation. So the last two characters are the state abbreviation...if it is present at all. I need to be able to select these 'identical' forms for processing.I have XML where the /forms/form/@number can be