RE: escape-uri-attributes

> According to section 2.4.2 of RFC 2396 an URI is always in 
> escaped form.
> This means that reserved characters, which are not used in 
> their normal
> URI function, as well as disallowed characters must be 
> encoded in order
> to have a valid URI.
> When the path segements are presented or interpreted in an environment
> they may be decoded.
> As a consequence, I think URIs should always be serialised in 
> escaped form, otherwise you're not serialising URIs.
>
I've found the discussion that followed this mail item very helpful.

But turning back to this message, it seems to me one could equally well
argue that if "a URI is always in escaped form", then there is no need to
escape it while serializing, because by definition it is escaped already.

I think there are a number of cases where escaping URIs turns out to cause
problems, but perhaps the best "excuse" for this facility is that it allows
users to indicate cases where escaping is unnecessary because it has already
been done in the source document.

I think it would also be useful to provide escape-uri() as a function to
provide users with precise control over where and when it is done.

Mike Kay

Received on Tuesday, 29 January 2002 08:21:03 UTC