On Wed, Feb 21, 2001 at 11:46:42PM +0000, Simon Cozens wrote: > OK: just a thought - this has gone pretty much far enough now. Nick and I have > agreed on a model which settles our differences, and which is also extremely > close to your model You cannot get "extremely close to transparency". It is either transparent, or not. And I warned you many times already: if your Perl is not transparent, your Perl is busted. AFAICU, you *cannot* translate 0..256 range from 'locale' to Unicode when upgrading byte->utf8. This would break too many things. But you do not *need* to translate: instead, all you need to do is to back-translate the cultural info tables used for utf8-encoded chars in the 0..255 range. IlyaThread Previous | Thread Next