[Seaside] UTF-8 in form

[Avi Bryant]

On Oct 9, 2004, at 11:06 AM, radoslav hodnicak wrote:
> Well this is a tricky problem to solve. You need to encode reserved 
> html
> characters like <>& etc. I guess the only real solution is to wait for
> squeak to have support for multibyte strings (you could try the 3.8 
> alpha
> version with m17n) where the encoder can loop over characters, not 
> bytes

No, it shouldn't be tricky.  We're only encoding those few reserved 
chars, and the whole point of UTF-8 is that if any of the bytes is an 
ASCII byte, it represents that ASCII char, so the kind of mistake 
you're talking about shouldn't happen.

It looks like this is an older version of Seaside, where we tried to 
encode more of the charset as entities (rather than just assuming 
people's data will be in UTF-8, as we do now).  Hillaire, what does 
WAAbstractHtmlBuilder class>>initialize look like for you?  Does it 
have a large table of entities or just  &quot;, &lt;, &amp;, and &gt;?

Avi