cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Torsten Curdt" <tcu...@dff.st>
Subject RE: XObject
Date Tue, 05 Dec 2000 12:38:21 GMT
> >> Object name: +/-0 (I'd prefer XMLFragment, but I'm easy on it)
> > Still not happy with name Ricardo?
> 
> A bit of history: XObject originally meant "an object capable
> of generating an XML representation of itself".
> 
> The idea was to provide a toDOM() method with semantics similar
> to those of Object.toString().
> 
> Its use in XSP (as the result of an <xsp:expr>, in particular)
> was to inline the Node returned by toDOM() in the corresponding
> document position.
> 
> In SAX, the same effect can be achieved by passing the XSP page's
> content handler as argumento the toSAX() method. This is not yet
> implemented.

What has to be implemented? Couldn't I just use
<xsp:logic>
  obj.toSAX(this.contentHandler);
</xsp:logic>

> Methods for handling <xsp:expr> were mistakenly omitted from XSP
> for C2. They must be included, of course, to enable the use of
> XObject's new incarnation.
> 
> The name "XObject" is also used by Xalan, so it's a good idea to
> avoid using it here.

Yepp! +1

> If we agree on the more general semantics of "able to generate an
> XML representation of itself", the name "XSPObject" is probably too
> XSP-centric.
> 
> It's accurate to say that such an object generates an XMLFragment,
> but this does not seem to reflect the above mentioned semantics...
> 
> Not that I suggest an ugly name like "XMLizable", but we should
> probably delve deeper...

I guess you are right: "XMLizable" seems to be the most accurate name.
Do you really think the name is that ugly that we should go for something
else?
--
Torsten

Mime
View raw message