You can do that if you wish. This is part of the standard. Look at the existing canonical decomposition mappings in the UCD (or just look at them in the charts which display them). Note that this will not make any difference for all conforming Unicode processes.
For example you can freely normalize texts to the NFD form (even if this form is not recommanded in many interchange protocols like HTML). Le 5 mars 2012 18:33, Denis Jacquerye <[email protected]> a écrit : > Hi, > > Could the following be decomposed instead of being encoded as single > characters? > COMBINING LATIN SMALL LETTER A WITH DIAERESIS > COMBINING LATIN SMALL LETTER O WITH DIAERESIS > COMBINING LATIN SMALL LETTER U WITH DIAERESIS
