W3C home > Mailing lists > Public > www-svg@w3.org > June 2005

Re: [SVGMobile12] Comments: Introduction

From: Boris Zbarsky <bzbarsky@MIT.EDU>
Date: Tue, 07 Jun 2005 09:02:27 -0400
Message-ID: <42A59AE3.7040409@mit.edu>
To: Craig Northway <craign@cisra.canon.com.au>
CC: w3c-svg-wg@w3.org, www-svg@w3.org

Craig Northway wrote:
>>> current SVG document fragment
>>>    The XML document sub-tree which starts with the ancestor 'svg'
>>>    element of a given SVG element, with the requirement that all
>>>    container elements between the 'svg'
>>>    and this element are all elements in the SVG language and namespace.
>>
>> So one always talks about the "current SVG document fragment" of some 
>> specific node, then?  Perhaps that should be made clear in the naming.
> 
> Yes, you do talk about the current SVG document fragment of a node. Are 
> you suggesting the name of this term should change?

I guess that would be too drastic... Perhaps the definition should 
clearly say that the term only makes sense when reference is made to a 
particular node?

> Perhaps but, I think this is needed to make it clear that only the 
> elements defined by the W3C SVG language are acceptable.

I see.  So if I have a <foo> in the W3C SVG namespace then that makes 
things not work here?  That makes sense, but is probably worth an 
informative example somewhere.

>> It seems that there can be no "current SVG document fragment" for a 
>> node

> Yes, thats what I believe. How's this:
 >
 >    current SVG document fragment
 >    The current SVG document fragment of a element is the XML document
 >    sub-tree which starts with the ancestor 'svg'
 >    <cid:part1.06080305.01090602@cisra.canon.com.au> element. For this
 >    to be a valid current SVG document fragement all container elements
 >    between the 'svg' <cid:part1.06080305.01090602@cisra.canon.com.au>
 >    and this element must be elements in the SVG language and namespace.
 >

That phrasing doesn't address my concern, which is that we clearly say 
that there may be no current SVG document fragment for a given node (and 
that all places which talk about current SVG document fragments handle 
the case when there isn't one in some way, but I suspect that already 
falls out of the general way rendering is speficied).  This also has the 
same "starts with" problem as "SVG document fragment".

Perhaps something like this:

-------------------------------------
current SVG document fragment
The current SVG document fragment of an element is the XML document
sub-tree such that:

1) The sub-tree has an 'svg' element as its root.
2) The sub-tree contains the element in question
3) All ancestors of the element in question in the sub-tree are elements 
in the SVG language and namespace.
-------------------------------------

I couldn't figure out a way to express the three criteria in sentence 
form without being very long-winded and hard to follow, but I see 
nothing wong with listing them in a list like this...


To modify this definition for 1.2 Full, I believe all you'd have to do 
is change condition (3) to:

3') All ancestors of the element in question in the sub-tree are 
elements in the SVG language and namespace and are either the root of 
the sub-tree or not an 'svg' element.

-Boris
Received on Tuesday, 7 June 2005 13:06:26 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:30 GMT