StringUtils.toUpperEnglish now has a small cache. This should speed up reading from a ResultSet when using the column name.
