We shouldn't have isDigit and digitValue in our Unicode class, because there's no context where we want to support arbitrary Unicode digits. All the places we support digits they need to be ASCII digits only.
My plan is to remove unused functions from our Unicode headers and also remove used functions that are being used when an ASCII function should be used instead. - digitValue is never used except where it should be ASCII-only - isDigit is never used except where it should be ASCII-only - isFormatChar is unused As far as I can tell, the other functions are used correctly, so I won't touch them for now.
Created attachment 24309 [details] patch
Created attachment 24310 [details] patch Oops, included unrelated changes by accident.
Comment on attachment 24310 [details] patch r=me, but I think you should land the cleanups separately.
http://trac.webkit.org/changeset/37531