cocoon-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Steven Maring <Steven.Mar...@trcinc.com>
Subject RE: Component approach to web development using Cocoon
Date Thu, 06 Jan 2000 10:36:06 GMT
I'm still not sure I follow this.  It sounds like you are saying that you
want to take a static XML and overide specific nodes with a more dynamic XML
node.  If that is true, can you come up with a better example.  I don't yet
see the need.

I also disagree with:
>3) XSL can't control more then 1 document at a time. So XSL is not the
>proper tool to do the merge of the documents.

XSL can load additional XML documents using the document() function.  And
XML can load additional nodes from an external entity into the root node.

Why don't you just stick with a single XML file and take advantage of
Cocoon's ability to have a processing instruction in XML call out to a
Servlet(or whatever) for dynamic processing?

-Steve M.
Tampa, FL

-----Original Message-----
From: Muly Oved [mailto:mulyoved@hotmail.com]
Sent: Thursday, January 06, 2000 5:02 AM
To: cocoon-users@xml.apache.org
Subject: Re: Component approach to web development using Cocoon


Thanks for the details answers. I will try to clear a few things.

>> HTML1+HTML2+XSL -> HTML3

>what does "html1 + html2" mean? what do you use to go from html1 to
>html2?

1) To enable parsing the html need to be converted to XHTML (which is a
valid XML document).
2) The + is a very limited operation in this context, all it does is taking
xhtml1 and replace one of it's section with section from xhtml2, the
sections are identified by the ID attribute.
3) XSL can't control more then 1 document at a time. So XSL is not the
proper tool to do the merge of the documents.


>>Regarding the list of technologies you are listing as alternatives, they
are used for
>>different realms of responsibility.  Here is a breakdown of what they
address:
This is exactly the motivation for that operation.
There are many technologies to create web pages. Many of those can be
represented as a static file on the server ready to be delivered to the
client. HTML, DHTML, JavaScript, Applet, XML+XSL, JSP, (JSP is included and
Servlet is not included)

What I suggest is a way to automate the management of using those different
technologies, and in a way that is orthogonal to those technologies, it is
not conflicting with Cocoon XML framework.

For example (not a good one)
The static page will include
<DIV ID="date">6-Jan-2000</DIV>

The component page can include
<DIV ID="date">
<SCRIPT LANGUAGE="JavaScript">
..
</SCRIPT>
</DIV>

Or
<DIV ID="date">
<%= currentDate%>
</DIV>
(In this case xhtml1+jsp2=jsp3 so the result is jsp which is still just a
valid xml document)

When thinking what will be the best approach to implement this it seem to me
that yet another xml language need to be designed to provide the
instructions, it can use the XPath spec to specify the parts but still need
to include new commands.

Do you think there is any value in such approach for organizing the
development of a web site?
Do you think it has a place in the Cocoon framework?

Regards,

Muly.



Mime
View raw message