Re: Rewrite of feature tag syntax rules

On Wed, 14 May 1997, Larry Masinter wrote:

> It's a common difficulty with URL processing, to the point
> where we wrote a special section in the revised URL draft
> on "when to escape and when not to escape". But the short
> answer is that you cannot 'unescape' a URL except when you are
> parsing it into its component parts. Thus, URL equivalence
> using %XX = <character represented by that byte> is unacceptable. 
> 
> Look at latest draft-fielding-url-syntax-* in the internet
> drafts directory.

That is true of URL processing, but I don't agree that is need be true of
a quasi URL used as a feature tag qualifier. The URL processing problem is
that %xx are used to escape characters which have syntactic significance
to the parsing process. The classic URL was/is expected to carry arbitrary
user input data, etc.  The feature code usage doesn't need that level of
complexity. 

There will be a lot less confusion if the feature code usage specifies
URLs which can be simply decoded and compared than if encoding is allowed
but the compares are made on the encoded result. Successful comparison of
the encoded result requires that methodology for encoding be much more
precisely specified than is now the case

Dave Morris

Received on Thursday, 15 May 1997 09:45:54 UTC