Re: [xsl] Streaming terminology: Climbing

[Michael Kay <mike@xxxxxxxxxxxx>]

> Would you elaborate on this please? I thought that the context node and the
current streaming position are one and the same, i.e., context node = current
streaming position?

No. If you do

<xsl:for-each select="ancestor::*>

it changes the context node (to somewhere earlier in the document) but it
doesn't change the current streaming position.
>
> You wrote:
>
> 	For example the construct
>
> 		descendant::x/parent::y
>
> 	is climbing, although it does not select ancestors
> 	of the context node. It's climbing because it reads
> 	information from the "stack" of ancestor data that
> 	is maintained as the streaming processor works its
> 	way through the document ...
>
> Truthfully, my heart sank when I read that, as it reveals that I still do
not grasp what "climbing" means. I thought that climbing means "climbing up
the XML tree."

Intuitively, yes, that's why we chose the name. But intuition doesn't always
give the right answer.

But I prefer to think in terms of permitted navigation paths (rather as we
drew them in the complex diagrams of the 2010 draft, see for example

http://www.w3.org/TR/2010/WD-xslt-21-20100511/#streamability-choice-and-repet
ition

Essentially the permitted navigation paths allow

C* D? A*

where C is a child step, D is a descendant step, and A is an attribute or
ancestor step.

This grammar can be defined by a finite state machine, and the states of this
machine can be labelled "striding", "crawling", and "climbing"; "climbing" is
the state that is reached after an A step, so any path that includes one or
more A steps takes you to a climbing state.

> Clearly descendant::x is not a construct that is climbing up the XML tree.
Are you saying that descendant::x/parent::y is climbing because there is a
portion of it that climbs up the XML tree, namely parent::y climbs up the XML
tree?

Yes, that's essentially what I've been saying in different words.
>
> You finished the above sentence with
>
> 	... and when you're positioned in that stack, your ability
> 	to select downwards is constrained.
>
> Would you elaborate on that please? It sounds like a very important
concept.

I thought your slides expressed that rather well. If the streaming position is
at a given node N, then information about the ancestors of N is available from
the stack, but information about the descendants of the ancestors of N isn't
available: either these precede N, in which case they have been forgotten, or
they follow N, in which case we haven't got to them yet.

Michael Kay
Saxonica
>
> Thanks!
>
> /Roger
>
> -----Original Message-----
> From: Michael Kay [mailto:mike@xxxxxxxxxxxx]
> Sent: Wednesday, January 29, 2014 6:50 AM
> To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx
> Subject: Re: [xsl] Streaming terminology: Climbing
>
> Yes, it's a very useful explanation.
>
> I think your conceptual model that the processor is "positioned just after
the start tag" is probably helpful to many people but it's worth pointing out
that it's not in the spec and we try not to over-constrain implementations.
There's also a glitch in it in that we (deliberately) define has-children() to
be an inspection construct, and to evaluate this function the processor needs
to peek just beyond the end of the start tag. (Saxon doesn't yet implement
this).
>
> There's also a minor error where you say:
>
> A construct is climbing if, when executed, it yields the context node (self)
or ancestors of the context node.
>
> Apart from the fact that attribute selection is also climbing (despite the
name), this confuses the role of the context node and the current streaming
position. For example the construct
>
> descendant::x/parent::y
>
> is climbing, although it does not select ancestors of the context node. It's
climbing because it reads information from the "stack" of ancestor data that
is maintained as the streaming processor works its way through the document,
and when you're positioned in that stack, your ability to select downwards is
constrained.
>
> Incidentally, at one time we defined the spec so that as well as retaining
"the start tags of all ancestors" in this stack, we also retained some
positional information about the position of each ancestor relative to its
siblings, allowing match patterns such as match="para[1]". This has been
dropped from the spec, but it's partially implemented in Saxon, and I have yet
to decide whether to retain it in some form.
>
> Michael Kay
> Saxonica