This does point out the strange fact that attributes are not children of
their parents - I think? Is it true that an attribute node's parent is
the element it's attached to? I'd assume so, but didn't check the spec...
-Mike
On 02/28/2012 01:44 PM, Andrew Welch wrote:
> On 28 February 2012 18:15, David Lee<> wrote:
>
>> Could you reference where in the specs that the child axis doesn't contain attributes ?
>> The best I can find is :
>>
>> * the child axis contains the children of the context node
>>
>>
>> And then:
>> * the attribute axis contains the attributes of the context node; the axis will be empty unless the context node is an element
>>
>>
>> But nothing I can spot that says attributes are not children.
>> I 'know' the answer but I can't find where in the specs its explicit.
>>
> I know what you mean, it's intuitive to think attributes are
> children... but what you are asking there you could also say about
> parent nodes - where is it explicit saying that parent nodes are not
> children.
>
> The other thing I used to get caught out by with attributes is that
> the attribute value isn't a separate node from the attribute name,
> both are in the same single node.
>
>
>
