Hi Desmond,
I'm aware of the static Character methods isSpaceChar() and isWhiteSpace(), but I don't know of an escape sequence that would filter out non-text characters.
I wonder if you might need to use StreamTokenizer. It allows you to identify number,
word (including single letter), end of line and end of file. You could for example use a switch statement based on the token, to identify numbers, words, end of line, end of file, and use a default to label everything else as spaces, tabs etc.
This way, you'd get all of your strings, characters, numbers and the rest wouldn't matter. It would however mean that single characters such as "$", "%" etc would end up in the default case.
If this sounds related to what you're after, let me know, as I have an example in code.
cheerio
rowan