On Sat, Aug 25, 2012 at 7:46 PM, Frank Millman <fr...@chagford.com> wrote: > Therefore, I think he is saying that he would have preferred that python > standardise on 4-byte characters, on the grounds that the saving in memory > does not justify the performance overhead.
If that's indeed the argument, then at least it's something to argue. What gets difficult is when people complain about the expansion from a 2-byte narrow build to the current 1/2/4-byte representation, which will indeed use more memory if there are a small number of >0xFFFF codepoints. But there's a correctness difference there. ChrisA -- http://mail.python.org/mailman/listinfo/python-list