This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 21314 - update dependencies type="feature"
Summary: update dependencies type="feature"
Status: RESOLVED FIXED
Alias: None
Product: XPath / XQuery / XSLT
Classification: Unclassified
Component: XQuery 3 & XPath 3 Test Suite (show other bugs)
Version: Working drafts
Hardware: PC Linux
: P2 normal
Target Milestone: ---
Assignee: O'Neil Delpratt
QA Contact: Mailing list for public feedback on specs from XSL and XML Query WGs
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-03-17 13:07 UTC by Sorin Nasoi
Modified: 2013-05-01 12:10 UTC (History)
1 user (show)

See Also:


Attachments
Report generated using the tools/dependencies.xq (3.07 KB, text/xml)
2013-03-17 13:07 UTC, Sorin Nasoi
Details

Description Sorin Nasoi 2013-03-17 13:07:22 UTC
Created attachment 1341 [details]
Report generated using the tools/dependencies.xq

The attached report was generated using the tools/dependencies.xq query provided in FOTS as of March 17th.

At this moment there are a few questions I would have wrt. dependencies with type="feature":

1) shouldn't there be a 1 to 1 relation between the official optional features as defined in the spec
http://www.w3.org/TR/xquery-30/#id-conform-optional-features)
and the features defined in FOTS?

1.1) If not, IMHO the ones that are not defined in the spec should have some information added in the catalog-schema.xsd such that it's clear for everyone what they stand for.

For example what does the 'collection-stability' stands for?
How about 'xpath-1.0-compatibility' or 'directory-as-collection-uri' or 'schema-location-hint'?

2) From what I understand, the 'schemaValidation' and 'schemaImport' were merged into 'schemaAware' right?
If so the test cases need to be updated also.
Also 'moduleImport' should be updated to 'module' for consistency to the spec.

3) Should the spec add an optional feature for 'namespace-axis' ?
Comment 1 O'Neil Delpratt 2013-03-20 10:15:13 UTC
(In reply to comment #0)
> 1) shouldn't there be a 1 to 1 relation between the official optional
> features as defined in the spec
> http://www.w3.org/TR/xquery-30/#id-conform-optional-features)
> and the features defined in FOTS?

The intent was that the values under dependency type='feature' should contain the group of optional-features from the relevant section in the spec and a derived group of features which are defined throughout the spec as optional for implementations. For example in the spec (see: http://www.w3.org/TR/xquery-30/#id-basics) it states  that the namespace-axis is optional for implementations.

 
> 1.1) If not, IMHO the ones that are not defined in the spec should have some
> information added in the catalog-schema.xsd such that it's clear for
> everyone what they stand for.

The features gathered from the query dependencies.xq do all appear somewhere in the spec. Nevertheless, as you have stated we need to document them somewhere.

I hope to do this. Any suggestions?
 
We can provide some documentation in catalog-schema.xsd relating to the references of the features in the spec, but this is only possible as comment documentation. Anything more we would need to be using schema 1.1, which is not good, we would be destabilize people's test drivers.

We could also do this all in query to map the features to references in the spec.



> 
> 2) From what I understand, the 'schemaValidation' and 'schemaImport' were
> merged into 'schemaAware' right?
> If so the test cases need to be updated also.
> Also 'moduleImport' should be updated to 'module' for consistency to the
> spec.

Maybe you are right. But I am not sure. Maybe someone can comment on this.

 
> 3) Should the spec add an optional feature for 'namespace-axis' ?

No. I think I have have addressed this under point 1.
Comment 2 Michael Kay 2013-05-01 12:10:26 UTC
I have improved the documentation on dependencies in the catalog available at

http://dev.w3.org/2011/QT3-test-suite/guide/running.html

I think that's the best I can do at the moment and propose to close this as resolved; please re-open if there is still a problem.