OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   Re: [xml-dev] version numbers and infosets

[ Lists Home | Date Index | Thread Index ]

> At 1:26 PM +0100 7/23/02, Richard Tobin wrote:
> >But as I said, I've been persuaded that we shouldn't require 1.0
> >parsers to reject documents labelled other than 1.0.

I would like to hear some more of the arguments that persuaded you, as I am
not yet convinced myself.

> This just makes the existing mess of different conformance levels to
> XML worse.


> I think this is a very bad idea. Leave the existing spec
> alone: no erratum, no change. XML 1.0 is defined by the XML 1.0 spec
> as originally published. Make any changes in future versions if you
> really must, but don't touch XML 1.0.

I fully agree. I have the gut feeling that this was done to "save people
work" which translates to implementors not needing to get out new versions
of their parsers. But I think there is a problem-- when working on a parser
for XML 1.0, I always knew that I would have to make changes to support 1.1.
That's what I thought the whole version thing was for. I thought that the
versioning system was designed for versioning-- at this point it is little
more than a hint-- which according to the erratum should be assumed to be
wrong. So that instead of rejecting 1.1 documents out of hand the parser
should try to parse the document, failing who knows when.

In terms of error messages it would make much more sense to receive "This
parser does not support version '1.1'" than to receive a strange character
error somewhere near the end of a 10 meg file. If for some reason, you
wanted the 1.0 parser to attempt to parse a 1.1 document, there was already
a system in place to do this-- leave off the <?xml?> decl. Of course then
you couldn't specify encoding in the document, but externally it would still
be possible.

The de facto usage has been to fail on != 1.0. This represents a change--
not a fix to an error-- the fact that the previous wording could not be
reached through any production was perhaps a problem-- but the solution
would be to make it reachable IMHO.

Either way, thanks for bringing this to the community!

Jeff Rafter
Defined Systems
XML Development and Developer Web Hosting


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS