I think Julik brought up a very important issue, and I wish it had gotten more attention. Ruby's Unicode string handling is broken, mostly because it doesn't count multibyte characters correctly.
Thijs Van Der Vossen wrote: >If you _need_ a dynamic language with a true and tested Unicode >String type _right now_ you might want to take a look at Python. ;-) Well, Julik did have a look at Python: "The Python and Perl ways of doing it are to distinguish between a 'bytestring' and a 'unicode string.' This is a way of the apocalypse." More importantly, though, why should we defer to other languages and frameworks? We love Rails, we love Ruby, and by making a small change in the String class we'll have best-in-class Unicode support. Add Globalize into the mix and you open up huge possibilities. Typo with out-of-the-box support for dozens of languages, including localized date display. Instiki with built in multi-language support, so that the rails wiki could be easily translated into dozens of languages. Ecommerce sites that are actually useful outside the US and UK. Because of the power and flexibility of Ruby and Rails, we can add this elusive i18n stuff pretty easily. Why not do it? It's a make-or-break feature for millions of people.
_______________________________________________ Rails-core mailing list Rails-core@lists.rubyonrails.org http://lists.rubyonrails.org/mailman/listinfo/rails-core
