Re: More XSL Discussion

Subject: Re: More XSL Discussion
From: "Martin Bryan" <mtbryan@xxxxxxxxxxxxxx>
Date: Thu, 26 Feb 1998 18:42:45 -0000
Paul Grosso wrote:
>We never talk about creating tags and markup in XSL, we talk about
specifying
>the creation of objects/elements, the attachment of
characteristics/attributes,
>and the copying, creating, suppressing of character data.  XSL is not meant
to
>specify a character-based transformation process; it does not create an
"output
>format," it builds a tree; it is inherently object-based.  It makes no
sense in
>this light to talk of half elements.

While I understand this theoretically it does not address the real life
problem of having to specify an end-tag for every element you start in the
tree within a specific rule, even before you process the children.

Consider an XML file that contains in it a set of task elements between two
paragraphs. Suppose I want to create an HTML ordered list from this set of
task elements. I need to first identify when the first task occurs, and
ensure I start my ordered list before opening the elemetn concerned. I then
need to determine which is the last task in the sequence, and ensure that I
close my ordered list at the end of that element. If neither of these
conditions apply I need to simply convert the task into a list item. The
problem is that unless the tasks are already in some kind of container it is
difficult to create an ordered list container with the correct scope.

Similar problems occur if you need to place more than one consecutive
element within a single cell in a table and the elements are not either
uniquely named or within their own container. Processing a set of elements
to form a single row in a table also tends to require (to my mind
unnecessary) use of the select-elements option.

I feel all these problems could be easily overcome by not requiring that an
end-tag be entered for every element in a rule.

Martin Bryan


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list


Current Thread