See also: IRC log
<trackbot> Date: 04 December 2013
Steven: On the linkedIn discussion, there were two people announcing online XForms composers
with context nodes
http://lists.w3.org/Archives/Public/public-forms/2013Dec/0001.html
Erik: Expressions in XPath2 don't
necessarily need a context node.
... there is a total of 12 refs to the model "associated
with
... the in-scope evaluation context node"
... [describes the content of the mail]
Steven: Is the problem only with events?
Erik: With actions
Steven: Oh yes.
... It would be a pain if you always had to say which model was
involved
Erik: Exactly. I think it should
still be optional.
... I think though that if it is not there, it shouldn't
require a context node.
Steven: So there's always a default model. And you switch with the model attribute.
Nick: It's only when you start using variables, that we might get different behaviour.
Erik: The bind attribute also changes the context.
Nick: Isn't there a restriction that you can't switch model and then use a bind from a different model?
Erik: Not sure
... Let me look.
... Hmm, well our implementation doesn't check for that. It's
not ambiguous.
Nick: At last week's call, Erik gave an example of the problem with variables
<nvdbleek> Example of something contra-intuitive:
<nvdbleek> <xf:var name="my-var" value="foo" model="model-2"/>
<nvdbleek> <xf:repeat ref="$my-var" model="model-1">
<nvdbleek> <nvdbleek> <!-- inscope model is model-2 NOT model-1 -->
<nvdbleek> <nvdbleek> <xf:var name="my-var" value="foo" model="model-2"/>
<nvdbleek> <nvdbleek> <xf:repeat ref="$my-var" model="model-1">
<nvdbleek> <nvdbleek> <!-- inscope model is model-2 NOT model-1 -->
<nvdbleek> <nvdbleek> </xf:repeat>
Steven: Is there a proposal for fixing the variable problem?
Erik: I think that it is a corner
case
... I think a better solution to use the model attribute, and
bind attribute rather than the context
Steven: Are there examples of Forms that wouldn't work any more? Would it affect interoperability?
Erik: Let me think.
... I cannot think of any behaviour that would change.
... It seems to be quite a corner case.
<nvdbleek> From the section 4.1 Evaluation Context:
<nvdbleek> Once the context item of the in-scope evaluation context has been determined according to the rules above, if the element expresses a model attribute that refers to a model other than the one containing the context item (if the context item isn't an instance node, the model is the model specified on the element or the model of its parent if no model attribute is specified on the element), then the context item of the in-scope evaluation context is change[CUT]
<nvdbleek> the top-level document element node of the default instance of the referenced model, and the context position and size are changed to 1.
Erik: I rarely use those actions.
Steven: I agree. In fact their existence at all is a bit odd.
Nick: I think everything will be compatible, except for variables, but they weren't in 1.1 anyway.
Steven: Is everyone happy with this?
Alain: Yes, no problem.
Steven: Ship it!
Nick: Someone has to make the change.
<scribe> ACTION: Erik to implement solution for action-159 [recorded in http://www.w3.org/2013/12/04-forms-minutes.html#action01]
<trackbot> Created ACTION-1962 - Implement solution for action-159 [on Erik Bruchez - due 2013-12-11].
close ACTION-1959
<trackbot> Closed ACTION-1959.
http://lists.w3.org/Archives/Public/public-forms/2013Dec/0002.html
Erik: Pretty small changes.
Steven: Do we define "context item"?
Erik: Yes.
<nvdbleek> http://www.w3.org/TR/xpath20/#dt-context-item
Steven: I'm having trouble
finding a reference to the definition. It ought to be in the
Glossary, I think, at least.
... item is defined there
... I guess these are editorial issues
Nick: I'll fix the alphabetical order.
Erik: There is a confusion, since we have "model item" which is a different thing.
Steven: Should 'node' be in the
glossary?
... How about 'Context item'?
Erik: There are 1200+ refs to
'node' in the spec.
... we need to say that a node is an XML node. Is that in the
XML spec?
<ebruchez> XPath 1.0: "XPath models an XML document as a tree of nodes"
<ebruchez> XPath 2.0: http://www.w3.org/TR/xpath-datamodel/#dt-node
Steven: So 'node' is defined in
XPath.
... so we should include node in the glossary pointing to
XPath?
<ebruchez> DOM: http://www.w3.org/TR/DOM-Level-3-Core/core.html
<ebruchez> DOM: "The DOM presents documents as a hierarchy of Node objects that also implement other, more specialized interfaces"
Xpath1: http://www.w3.org/TR/xpath/#data-model
Erik: We do explicitely state that we use an XPath data model.
<ebruchez> See 3.2.2 The instance Element
Steven: So we can define node as
'the definition as used in the XPath data model'.
... Can you add it Nick?
Nick: We already have a ref to the XPath data model?
Steven: Yes.
Erik: None from me.
[ADJOURNED]
This is scribe.perl Revision: 1.138 of Date: 2013-04-25 13:59:11 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/Forms/XForms/ Succeeded: s/the/... the/ Succeeded: s/model./model?/ Succeeded: s/there/their/ Succeeded: s/bel/ble/ No ScribeNick specified. Guessing ScribeNick: Steven Inferring Scribes: Steven Default Present: Steven, ebruchez, [IPcaller], nvdbleek Present: Steven ebruchez [IPcaller] nvdbleek Regrets: Philip Agenda: http://lists.w3.org/Archives/Public/public-forms/2013Dec/0003.html Found Date: 04 Dec 2013 Guessing minutes URL: http://www.w3.org/2013/12/04-forms-minutes.html People with action items: erik[End of scribe.perl diagnostic output]