Re: [xml-dev] SAX Topic 2004-0003: Unicode normalization

"Karl Waclawek" <[email protected]> Sun, 22 Feb 2004 16:50:18 -0500
Newsgroups gmane.text.xml.sax.devel,gmane.text.xml.devel
Message-ID <005901c3f98d$dcdd9540$0207a8c0@karl>
----- Original Message ----- 
From: "David Megginson" <[email protected]>
To: "XML Developers List" <[email protected]>
Sent: Sunday, February 22, 2004 9:28 AM
Subject: [xml-dev] SAX Topic 2004-0003: Unicode normalization


> * Topic 2004-0003: Unicode normalization
> 
> - add a feature to request Unicode normalization checking (John Cowan,
>    2004-02-18)

This should not be a "must recognize" feature, as the XML 1.1 spec
does not require it as a "MUST".

> - add a feature to XMLReaderFactory, such as
>    http://xml.org/sax/features/unicode-normalization-checking; most
>    appropriate at the factory level (Norm Walsh, 2004-02-19); Jeff
>    Rafter (2004-02-19) and Elliotte Rusty Harold (2004-02-19) disagree;
>    Norm wants feature in factory so that factory can return different
>    parsers (2004-02-20)

Ideally the factory is the better place for it, as it would
be for requesting a validating parser. But what about
backwards compatibility?

> - unicode-normalization-checking should be false by default (Elliotte
>    Rusty Harold, 2004-02-19)
> 
> - errors should be reported through the error() callback (Elliotte
>    Rusty Harold, 2004-02-19); Norm Walsh agrees (2004-02-20); using
>    error() instead of warning() may cause too much noise (Jeff Rafter,
>    2004-02-20); should error() or warning() be called only on the first
>    problem? (Jeff Rafter, 2004-02-20)

Should be called on every problem. If the app decides that
parsing should continue despite an error, that does not mean
that further errors should be kept hidden from it automatically.

It is conceivable to have user options for ignoring certain
kinds of errors. In the case of Unicode normalization:
What about the ability to turn normalization checking
off in the middle of parsing? 
 
> - may want to standardize an exception subtype for normalization
>    errors (Elliotte Rusty Harold, 2004-02-19); Norm Walsh agrees
>    (2004-02-20)

What about a special exceptionId instead?
(I know exception ids are not well standardized yet).
 
Karl


-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
List: sax-devel, [email protected]
See:  http://www.saxproject.org/
https://lists.sourceforge.net/lists/listinfo/sax-devel