Tim, thanks for clarifying where you are coming from on this.

I interpreted "some component of the static context" differently to you. The components are the 'variables' in the static context, not schema objects. The XPST0001 error says "It is a static error if analysis of an expression relies on some component of the static context that has not been assigned a value. " , so I think this message would be issued if, for example, an XPath processor tried to look up an element in the In-scope schema definitions component but found that the component itself had not been built. I think the error that applies to the case we are discussing is XPST0008 as quoted below. Otherwise I don't see when XPST0008 would be issued.

Consequently I am not sure that XPath rules prohibit the genuinely-impossible situation where an expression on choiceElement[1]/A refers to choiceElement[1]/B, because I can't see a crisp definition of in-scope. Maybe I'm missing it? Common sense says that it should be prohibited. But I agree that XPath does not prohibit choiceElement/A referring to choiceElement/B generally.

My conclusions about the rules in 23.1 was perhaps too general, and they should be looked at on an individual basis. But the reason they exist is because the parser is not acting on the whole message and in general is prohibited from looking ahead. Clearly the first two rules are needed (assert/discriminator can look down but not ahead, and outputValueCalc can look ahead). The other three are the ones in question. The current wording came from this action which was spun off from the unordered action 199:

214

Expression Language Data Model (All)
16/7: Augment section 23.1 so that it covers the cases where an XPath references an element that is in a different choice branch or that is in the same unordered sequence or that is floating. These cases could be detected statically (though to do this 100% reliably is not easy) or they could be left until runtime and fail if the element does not exist. Both are schema definition errors as explained by errata 2.120.
23/7: Closed. Runtime schema definition error if the element does not exist. Errata taken.

(Erratum 2.120 is just the one that categorised all the different errors.)

For the choice branch rule: If XPath disallows choice/x[1] referring to choice/y[1] but allows general refs, then I am ok with changing this to match XPath (ie, we drop our rule as it is implied). If it doesn't disallow choice/x[1] referring to choice/y[1] then perhaps we should keep our rule but tighten it up.

I think the unordered rule and floating rule both stem from the choice branch rule - because of the rewrite semantic that turns unordered into a choice.

Regards

Steve Hanson
Architect, IBM DFDL
Co-Chair, OGF DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848

From: Tim Kimber/UK/IBM
To: Steve Hanson/UK/IBM@IBMGB,
Cc: Mark Frost/UK/IBM@IBMGB
Date: 28/04/2014 22:03
Subject: Re: [DFDL-WG] validating expressions on elements in a choice or unordered sequence

I don't think I am missing the point. DFDL's usage of a partial XDM is not in play here.

The static context is defined thus: [Definition: The static context of an expression is the information that is available during static analysis of the expression, prior to its evaluation.] This information can be used to decide whether the expression contains a static error. If analysis of an expression relies on some component of the static context that has not been assigned a value, a static error is raised [err:XPST0001].

This is the definition that is relevant, because the errors are issued by our static analysis ( performed by the DFDL validator ). I believe the rule already prohibits the genuinely-impossible situation where an expression on choiceElement[1]/A refers to choiceElement[1]/B. But it does not prohibit choiceElement/A referring to choiceElement/B - and I don't think DFDL should either.

regards,

Tim Kimber,
IBM Integration Bus Development (Industry Packs)
Hursley, UK
Internet: kimbert@uk.ibm.com
Tel. 01962-816742
Internal tel. 37246742

From: Steve Hanson/UK/IBM
To: Tim Kimber/UK/IBM@IBMGB,
Cc: dfdl-wg@ogf.org, dfdl-wg-bounces@ogf.org
Date: 28/04/2014 13:51
Subject: Re: [DFDL-WG] validating expressions on elements in a choice or unordered sequence

It is certainly easier if we can just do the same as XPath 2.0 stipulates. But I think that this misses the point here.

The XPath error for statically detecting that an expression refers to something that can never exist is XPST0008, which says:

It is a static error if an expression refers to an element name, attribute name, schema type name, namespace prefix, or variable name that is not defined in the static context, except for an ElementName in an ElementTest or an AttributeName in an AttributeTest.

The static context has the notion of "In-scope schema definitions" being "a generic term for all the element declarations, attribute declarations, and schema type definitions that are in scope during processing of an expression.". It doesn't define exactly what is meant by "in-scope" but XPath assumes that it acts on a complete instance of an XDM.

In DFDL we are different to typical XPath usage as we are applying expressions during parsing when the document is incomplete. We can use that as the justification for applying extra constraints, which is exactly why there are additional rules in section 23.1.

So, if there are scenarios where a rule is going to be restrictive then we should consider dropping it. If there are not, but it makes the life of an implementer harder because it is hard to code the rule, then we should consider dropping it. Otherwise keep it.

Regards

Steve Hanson
Architect, IBM DFDL
Co-Chair, OGF DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848

From: Tim Kimber/UK/IBM@IBMGB
To: dfdl-wg@ogf.org,
Date: 11/04/2014 14:03
Subject: Re: [DFDL-WG] validating expressions on elements in a choice or unordered sequence
Sent by: dfdl-wg-bounces@ogf.org

I would be quite uncomfortable with DFDL not being a 'proper subset' of XPath 2.0. I understand the motivation ( having personally been involved in coding a query engine for DFDL ) but I think the cure would be worse than the complaint. Consistent with that, I think I agree with Mark's suggestion - a DFDL processor should just 'do what an XPath processor would do'.

regards,

Tim Kimber,
IBM Integration Bus Development (Industry Packs)
Hursley, UK
Internet: kimbert@uk.ibm.com
Tel. 01962-816742
Internal tel. 37246742

From: Mike Beckerle <mbeckerle.dfdl@gmail.com>
To: Mark Frost/UK/IBM@IBMGB,
Cc: "dfdl-wg@ogf.org" <dfdl-wg@ogf.org>
Date: 11/04/2014 13:23
Subject: Re: [DFDL-WG] validating expressions on elements in a choice or unordered sequence
Sent by: dfdl-wg-bounces@ogf.org

Comments inline

On Fri, Apr 11, 2014 at 6:22 AM, Mark Frost <FROSTMAR@uk.ibm.com> wrote:
When we were implementing unordered sequences, this raised some questions around evaluating relative paths in expressions, for elements in a choice or unordered sequence :

DFDL spec: (gwdrp-dfdl-v1.0.4 section 15)
"When processing a choice group the parser validates
any contained path expressions. If a path
expression contained inside a choice branch refers
to any other branch of the choice, then it is a
schema definition error."

1. I'm not clear what benefit this restriction on path expressions gives.
It seems redundant since in any single instance of a choice group, if the branch being processed exists, then by definition none of it's sibling branches exist. Any expression path referring to a non-existent branch would correctly return <empty sequence>

Typically in XPath, such paths would just be empty-sequence at runtime. Making it an SDE hoists the error to (hopefully) compile time, and making it SDE (non-recoverable) changes the way one must write expressions. You can't write utter nonsense paths and have them be runnable.

If the choice group is inside a repeating structure, then expressions referring to choice branches within other instances of the choice could be useful.
Should an expression referring to branches in other instances of a choice cause a schemadef error?

Should be no issue if you are looking at say, position() - n. If you reach to something that doesn't exist, then you'll get empty sequence.

My experience so far with XPath is that this notion that non-existance returns empty sequence is painful at best and a nightmare at worst. Expressions that are utterly nonsense are accepted executed, and silently fail by returning empty sequence. The most common mistake is writing /a/b/c when you needed /ns1:a/ns2:b/ns3:c.

Example
expression on el_b could be{ fn:count(../../el_choice/el_a) }

- parent
[sequence]
- el_choice [minOccurs=5 maxOccurs=5]
[choice]
- el_a
- el_b

2. Should an expression that potentially refers to branches in the choice cause a schemadef error?

Example
identically named elements in and out of a choice
expression on el_c could be { fn:count(../el_a) }

- parent
[sequence]
- el_a
- el_b
- [embedded choice group]
- el_a
- el_c

I'd love to restrict this, because we're looking at having to create a DFDL expression language implementation for performance reasons, and complex things like this require a very complex implementation tantamount to a query-engine.

I would claim that these two el_a elements are different, and we could choose to restrict a DFDL path expression to return only nodes described by the same schema component, with "same schema component" meaning same path from document element to the schema component where an element or group or type reference counts as part of that path. So two different element references to the same global element would be two different schema components.

But I suspect that this is too restrictive, and implementations are just going to have to be sophisticated enough to execute queries like this one, and a good implementation will optimize simpler cases for faster execution.

...mikeb-- dfdl-wg mailing list dfdl-wg@ogf.orghttps://www.ogf.org/mailman/listinfo/dfdl-wg

Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU-- dfdl-wg mailing list dfdl-wg@ogf.org https://www.ogf.org/mailman/listinfo/dfdl-wg

Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU