RE: another 1.1 question part II

OK, good I see what is going on now.

The bind element does not technically have a Node Set binding, and the 
submission doesn't technically have a Single Node binding.  This is 
because a Node Set binding is defined to be that which can be expressed 
via nodeset+[model] attribute set *or* a bind attribute.  Similarly, a 
Single Node binding is defined to be that which can be expressed as a 
ref+[model] attribute set *or* a bind attribute.  The bind and submission 
elements do not allow the model and bind attributes.

So, the first of the two paragraphs reflected my attempt to say that the 
rules for in-scope context work for bind and submission too, even though 
they are only capable of expressing a binding to a nodeset or a binding to 
a single node, respectively. 

Clearly, I didn't quite carry it through to completion.  The latter 
paragraph cited by Aaron should use the words "If the bind element does 
not express a binding to a node set..." and "Otherwise, the bind has an 
expressed nodeset attribute, and computed expressions...".

A similar check should be done to make sure we say similar things about 
submission.

I am assuming you guys have no problem with me forwarding this to the 
www-forms-editor list as it is just an editorial mistake that I can then 
take the action item to rectify.

Best regards,
John M. Boyer, Ph.D.
STSM: Workplace Forms Architect and Researcher
Co-Chair, W3C Forms Working Group
Workplace, Portal and Collaboration Software
IBM Victoria Software Lab
E-Mail: boyerj@ca.ibm.com  http://www.ibm.com/software/

Blog: http://www.ibm.com/developerworks/blogs/page/JohnBoyer





"Klotz, Leigh" <Leigh.Klotz@xerox.com> 
02/15/2007 04:48 PM

To
John Boyer/CanWest/IBM@IBMCA
cc
"Aaron Reed" <aaronr@us.ibm.com>, "Jan J Kratky" <kratky@us.ibm.com>
Subject
RE: another 1.1 question part II






I think Aaron should be the authority on this but I'll take a stab.
Both of these paragraphs are in the current 1.0 errata:
 
http://www.w3.org/2006/03/REC-xforms-20060314-errata.html#E17c
"Note: The rules above for determining the in-scope evaluation context are 
applicable to XForms elements that can contain a binding expression 
attribute even if they cannot contain a Single Node Binding or Node Set 
Binding. The elements bind and submission are examples." 
The last sentence is confusing because, as Aaron points out, bind can have 
a Node Set binding and submission can contain a Single Node Binding. 
I've read this a few times and I can't figure out what it means to have a 
"binding expression attribute" but not "contain a Single Node Binding or 
Node Set Binding."  So if this does means something, that's what needs to 
be explained.
If the issue is instead just finding an example of an element that has an 
in-scope evaluation context but no binding attributes, then I would 
suggest action, which has an evaluation context for its children in XForms 
1.0, and for its conditional attributes in XForms 1.1.
I think that the reason Aaron brought up the second paragraph (the one at 
the very bottom of this thread); is that states that bind can have a Node 
Set Binding, which contradicts the statement bove that it can't.
Leigh.

From: John Boyer [mailto:boyerj@ca.ibm.com] 
Sent: Thursday, February 15, 2007 3:14 PM
To: Klotz, Leigh
Cc: Aaron Reed; Jan J Kratky
Subject: RE: another 1.1 question part II


Hi, 
Could you guys explain a little more about what wording is problematic? 

One comment from Leigh agrees with Aaron about some juxtaposition or 
another, Aaron says something is fishy, etc. 

But I don't have enough time to try to guess which pieces of the spec or 
errata need work. 

Could you instead say "Section X.Y.Z says A and it should say B because of 
reason C." 

You could post to the forms editor list too so we can pick it up as a last 
call comment. 

Cheers, 
John M. Boyer, Ph.D.
STSM: Workplace Forms Architect and Researcher
Co-Chair, W3C Forms Working Group
Workplace, Portal and Collaboration Software
IBM Victoria Software Lab
E-Mail: boyerj@ca.ibm.com  http://www.ibm.com/software/

Blog: http://www.ibm.com/developerworks/blogs/page/JohnBoyer




"Klotz, Leigh" <Leigh.Klotz@xerox.com> 
01/30/2007 03:11 PM 


To
"Aaron Reed" <aaronr@us.ibm.com> 
cc
John Boyer/CanWest/IBM@IBMCA, "Jan J Kratky" <kratky@us.ibm.com> 
Subject
RE: another 1.1 question part II








I think maybe the various cases got mixed together in the explanation. The 
submission and bind not having nodeset is wrong though, as you correctly 
point out. 
I think maybe the two paragraphs could actually be clearer ;-) 

From: Aaron Reed [mailto:aaronr@us.ibm.com] 
Sent: Tuesday, January 30, 2007 2:12 PM
To: Klotz, Leigh
Cc: John Boyer; Jan J Kratky
Subject: RE: another 1.1 question part II
Hey Leigh,

I have no problem with the paragraph being there. I just wanted to point 
out the inconsistencies in the wording.

When I was reading the evaluation context stuff my mind kept thinking 
about nested binds but I didn't see too much that was directly on point 
other than what you pointed out. I assume that is covered under the 
section on the bind element. I just haven't gotten to that part, yet.

--Aaron
IBM Corporation
Internal Zip: 9022D016 
11501 Burnet Road
Austin, TX 78758
(512)838-9948
inet: aaronr@us.ibm.com
_
(} @
|= Volleyball Rules!!!
/\


"Klotz, Leigh" <Leigh.Klotz@xerox.com>

"Klotz, Leigh" <Leigh.Klotz@xerox.com> 
01/30/2007 11:05 AM 




To

Aaron Reed/Austin/IBM@IBMUS, "John Boyer" <boyerj@ca.ibm.com>, Jan J 
Kratky/Raleigh/IBM@IBMUS 


cc



Subject

RE: another 1.1 question part II







Aaron,
I'm pleased that you're asking these question. I'm just going to comment 
on one item though:

I agree with you about the mistake in the juxtaposition of "cannot 
contain" and "bind" in the first paragraph.
But the second paragraph is (I believe) required, because there are two 
cases of bind elements that do not express a nodeset binding, both of them 
involving nesting:

1. bind as a container to give a subtree (use case is transclusion of 
documents for template-built forms):
<bind>
<bind nodeset="a" type="xsd:boolean" />
<bind nodset="b" type="xsd:integer" />
</bind>

According to section 7.4, as the outer bind is an element contained in a 
model, the context node is the first instance's toplevel element, and that 
sets the context node for the two inner binds.

2. bind inside another bind which sets the context itself (use case is not 
as strong, but it's legal and we have to define it):
<bind nodeset="instance('books')">
<bind type="title">
</bind>

Again following 7.4, this sets the context node to the toplevel element of 
the instance named "books".

There were some questions about what nested bind means when the outer bind 
matches a nodeset with more than one node, especially with regard to id 
attributes on the inner binds.
I'm sure John or Jan will remember what the resolution was.

Leigh.

From: Aaron Reed [mailto:aaronr@us.ibm.com] 
Sent: Tuesday, January 30, 2007 1:50 AM
To: John Boyer; Jan J Kratky; Klotz, Leigh
Subject: another 1.1 question part II 
... 

Note: 
"The rules above for determining the in-scope evaluation context are 
applicable to XForms elements that can contain a binding expression 
attribute even if they cannot contain a Single Node Binding or Node Set 
Binding. The elements bind and submission are examples." 
[snip...] 
"XPath expressions also appear in model item property attributes of the 
bind element to define computed expressions. If the bind element does not 
express a Node Set binding, then the in-scope evaluation context for model 
item property attributes of the bind is equal to the in-scope evaluation 
context for the bind. Otherwise, the bind has a Node Set binding, and 
computed expressions..."

The 'Note:' says that bind elements don't have Node Set binding, but then 
you can see that two paragraphs later the spec says that bind elements 
have Node Set binding. Sounds fishy. :-)
,,,