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

 


Help: OASIS Mailing Lists Help | MarkMail Help

 


 

   Re: SAX Comments

[ Lists Home | Date Index | Thread Index ]
  • From: Matt Sergeant <matt@sergeant.org>
  • To: Rob Lugt <roblugt@elcel.com>
  • Date: Tue, 05 Dec 2000 09:49:10 +0000 (GMT)

On Tue, 5 Dec 2000, Rob Lugt wrote:

> At the risk of this thread becoming "which SAX is real SAX?", I was
> referring to the SAX 2.0 specification that can be found at David
> Megginson's site: http://www.megginson.com/SAX/Java/javadoc/index.html
>
> This is specifically a java binding of SAX 2.0, but I would expect other
> "compliant" SAX 2.0 implementations to follow the same handler/callback
> model and to support the same set of callbacks.  The perl/SAX implementation
> is significantly different to this in that it has many additional callbacks
> not present in the "official" specification.

I still see absolutely no reason to be forced to make the API's the same
across languages. The old argument of being easy to port from one language
to another doesn't seem to have any weight behind it (has *anyone* had to
do this?) when faced with the benefits of supporting a particular
language's paradigm.

> Is the java binding considered to be the "official" specification?  Or is it
> simply the one that is best documented?

Its just the one you know about best, probably. I've not noticed any flaws
in the current SAX documentation for other languages.

> Am I wrong to get too hung up about
> SAX compliance?  I can see that for java developers strict compliance is a
> way for multiple implementations to support a given interface, and therefore
> enable plug-compatible parsers to be selected at runtime.  Microsoft's COM
> interfaces can be used in a similar fashion.  But "standardized" bindings
> for other languages (eg C++) do not exist - so it is impossible for non-java
> implementations to claim true SAX 2.0 compliance.  Perhaps this is the
> reason that perl/SAX evolved to be so different to the java binding?

Perl supports a different programming paradigm to java. Hashes are first
class objects. It only makes sense to use them there, rather than force
them into the Java way. That happened with Perl's XML::DOM module, and
many people don't like having to use Java semantics to program their Perl
code.

It comes down to a simple question: is cross language portability really
that important for SAX?

-- 
<Matt/>

    /||    ** Director and CTO **
   //||    **  AxKit.com Ltd   **  ** XML Application Serving **
  // ||    ** http://axkit.org **  ** XSLT, XPathScript, XSP  **
 // \\| // **     Personal Web Site: http://sergeant.org/     **
     \\//
     //\\
    //  \\





 

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

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