For the most part, handling non-ASCII text is not a Ramaze concern. It's important that you understand how Ruby handles string encodings and how HTTP and HTML handle them. In Ruby 1.8.6 the String class is quite naive about encodings. It is essentially just an array of bytes internally, so handling anything other than ASCII requires understanding and careful thought. If you simply accept strings entered by the user, store them somewhere (e.g. a database) and then show them back to your users without manipulating their contents then there's no problem. The bytes of your String go back out exactly as they came in and so everything works out fine. However for that to work you need to make sure that the user's web browser is sending and displaying strings in the same encoding. This can be achieved by setting a couple of important things in your HTML, and by setting HTTP headers. To setup your HTML to state that it is being output as UTF-8 and to ask the browser to send data as UTF-8: * Add accept-charset="utf8" in your
tags, e.g. * Add a meta tag at the top of your section for all pages (e.g. by putting it in your common page layout file): ... With Ruby 1.9 the String class is engineered as a sequence of characters and natively supports multiple string encodings. This means that you should be able to manipulate your strings successfully once they're into Ramaze. Still, you need to setup your website to ensure that browsers are told how to send and receive data, in the manner described above. With a little effort and understanding it's possible to successfully manipulate non-ASCII strings in Ruby 1.8.6 as well, but that's beyond the scope of this document. As well as setting indicators for string encodings in your HTML, you can set HTTP headers to reinforce your intent. Ramaze makes this easy - simply add the following in start.rb, before your call to Ramaze.start, to set the default headers that Ramaze sends back: Ramaze::Global.content_type = 'text/html; charset=utf-8' Ramaze::Global.accept_charset = 'utf-8'