[tbnl-devel] Unicode Parameters in OpenMCL

[Edi Weitz]

Hi!

On Sat, 8 Oct 2005 00:46:30 -0700, Timothy Bennett <timothy.s.bennett at gmail.com> wrote:

> I have come up with a small way to read and correctly display form
> data with non-ascii characters in OpenMCL, which does not support
> Unicode.

It doesn't?  Oh, I didn't know that.

> Or at least, I think I have.  There may be good reason not to do
> things this way - someone let me know if that is the case.
>
> First, I redefined escape-for-html so that it would return NIL if
> its parameter was NIL instead of generating an error.  Then I just
> used escape-for-html instead of escape-string-all when displaying
> the parameter (I was just playing around with the tests that came
> with TBNL here).
>
> My question is this: are there problems that can arise with the use
> of escape-for-html that escape-string-all avoids?  If not, then I'll
> just be using this.

This is assuming that OpenMCL uses 8-bit characters similar to CMUCL:

ESCAPE-STRING-ALL will escape /all/ characters that aren't part of the
7-bit ASCII repertoire while ESCAPE-FOR-HTML will only escape a couple
of "critical" characters like < and &.  This is relevant if your HTML
page contains non-ASCII characters (like, say, umlauts).  If you don't
escape them you better send corresponding information to the browser
(which is a good idea anyway) - either a header like

  Content-Type: text/html; charset=iso-8859-1

(which is the default header used by TBNL) or a meta tag like

  <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">.

Cheers,
Edi.