[
Lists Home |
Date Index |
Thread Index
]
> From: John Cowan [mailto:jcowan@reutershealth.com]
> Sent: Saturday, March 23, 2002 5:33 PM
> To: Julian Reschke
> Cc: michael.h.kay@ntlworld.com; 'Joshua Allen'; 'Rick Jelliffe';
> xml-dev@lists.xml.org
> Subject: Re: [xml-dev] MSXML DOM Special Chars Less Than 32
>
>
> Julian Reschke scripsit:
>
> > Some of the requirements are:
> >
> > - the format of strings that *can* be represented as XML
> characters doesn't
> > change
> > - non-XML characters must be ignored by implementations not knowing the
> > escaping mechanism
>
> Those requirements are impossible. An escaping mechanism naturally
> requires an escape character (which means that the format of strings
> containing the escape character *must* change), and the only way
> for an escape-blind system to ignore the escape is for escaped
> characters to be represented by the null string.
> If the escapes are elements, that makes the matter even worse,
> since escape-blind systems will see violations of the schema.
Why impossible?
You just mentioned a format that fulfills the requirements (mixed content,
using special elements to represent the offending characters):
here we have a <ctl:bel xmlns:ctl="..." /> character.
If there's an *additional* requirement that the format may not affect
validation, then there'd be still the possibility to use PI-based escapes:
here we have a <?ctl-bel?> character.
However, I'm not sure that this requirement really exists (it certainly
doesn't for SOAP and WebDAV).
(it would be great if this discussion would lead to a common approach that
could be applied to multiple protocols / XML formats).
|
