W3C home > Mailing lists > Public > w3c-ietf-xmldsig@w3.org > April to June 2002

Re: Provokant proposal on Exclusive C14n

From: Christian Geuer-Pollmann <geuer-pollmann@nue.et-inf.uni-siegen.de>
Date: Wed, 05 Jun 2002 10:41:42 +0200
To: Aleksey Sanin <aleksey@aleksey.com>
Cc: Joseph Reagle <reagle@w3.org>, John Boyer <jboyer@PureEdge.com>, "Donald E. Eastlake 3rd" <Donald.Eastlake@Motorola.com>, w3c-ietf-xmldsig@w3.org
Message-ID: <6783774.1023273702@clouseau>

>> Christian Geuer-Pollmann wrote:
>>
>> Right. What I suggest is to include ALL namespace nodes (not the ones
>> from the DOM model but the ones from the XPath model). Every namespace
>> node in an ancestor should effect the current node, not only the
>> namespaces nodes whose owner element is in the document subset.
>>
>
> On Mittwoch, 5. Juni 2002 01:42 -0700 Aleksey Sanin <aleksey@aleksey.com> 
wrote:
> Yes, I agree with this. But I do not like the case
> when we have following document
>
>     <foo:Something xmlns:foo="http://example.org/foo">
>         <bar:Something xmlns:bar="http://example.org/bar" />
>     </foo:Something>
>
> the XPath expression looks like
>
>     self::bar:Something
>
> and the canonicalized output looks like
>
>     xmlns:foo="http://example.org/foo"
>         <bar:Something xmlns:bar="http://example.org/bar"
> xmlns:foo="http://example.org/foo" />     From my point of view it should
> look have *only* <bar:Somehting />
>
>         <bar:Something xmlns:bar="http://example.org/bar" />
>
> So the suggestion is to include all visibly utilized namespace nodes and
> simply do not check namespace nodes for presence in the input nodes set.
>
> Aleksey.

Uuh, it becomes complicated ;-)) I agree on what you say:


If we do 'enforce' that all (relevant) changes in namespace-space become 
visible in the elements where (1) the namespace is utilized or (2) the 
namespace is in the IncusiveNamespaces.PrefixList and 'changes' it's value, 
there's no need to output a namespace attribute if the owner element is 
omitted from the document subset.

I would say: If the owner element (parent?) of a namespace node is not in 
the document subset, the namespace node is not output.

Regards,
Christian

PS: Yesterday, I asked Merlin (not via list) the following:

--On Dienstag, 4. Juni 2002 15:52 +0100 merlin <merlin@baltimore.ie> wrote:

>
> Hi Christian,
>
> Namespace nodes are not output if their parent element
> is not in the node set, unless their prefix is in
> IncusiveNamespaces.PrefixList.
>
> http://www.w3.org/Signature/Drafts/xml-exc-c14n.html
> Section 3, item 3
>   [A namespace node not in IncusiveNamespaces.PrefixList is ignored if]
>   1. Its parent element is not in the node set
>
> Merlin
>
> r/geuer-pollmann@nue.et-inf.uni-siegen.de/2002.06.04/16:19:56
>> Hi Merlin, just a short question about exclusive c14n:
>>
>> In your (inclusive) c14n samples, you outputted attribute nodes and
>> namespaces if they are in the document subset while their owner element
>> is  not.
>>
>> Q: Does the same apply for exclusive c14n? For example, the 'A' element
>> is  not in the subset but it's namespace and attribute axis: ([] denotes
>> "in  the document subset):
>>
>> input:
>>
>> <A [d:a="foo"] [xmlns:a="http://a"] />
>>
>> output:
>>
>> xmlns:a="http://a" d:a="foo"
>>
>>
>> Is this correct?
>>
>> Regards,
>> Christian
Received on Wednesday, 5 June 2002 04:48:56 GMT

This archive was generated by hypermail 2.2.0 + w3c-0.29 : Thursday, 13 January 2005 12:10:16 GMT