<div>On 5/29/07, Jay McCarthy <<a href="mailto:jay.mccarthy@gmail.com">jay.mccarthy@gmail.com</a>> wrote:<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> > It's not that difficult: bytes->string/utf-8 and family. > The main problem is that if this is going to fail, I don't want to be > the one blaimed for it. The server has no reason to look at these > values, so it irks me that it crashes by them. Furthermore, by moving > the error in the servlet code, it can be caught more effectively and > displayed to the user in an appropriate context.</blockquote> <div> Perhaps an on-demand conversion can be the best of both worlds?  i.e. the bindings are parsed as bytes, but bytes->string* is called with extract-binding, so the error can be trapped by servlet, but we don't need to write explicit conversions in servlet unless necessary (and might be possible to pass in encoding too to switch between utf-8/current-locale/latin-1). Just some more ideas... yinso </div></div>