Hey Mukul,

Sorry I should have been more specific.  I'm sure that Xerces does indeed pass 
the XML Schema 1.1 test suite.  That said, we have run into a number of issues 
that are directly related with errors in XPath  -- which XML Schema 1.1 uses 
for assertions.  So when I say test suite, I'm talking about PsycoPath passing 
the W3C XPath / XQuery test suite. If you look at the github project 
(https://github.com/sharwell/webtools.sourceediting.xpath), you can see that  
Sam has integrated the test suite into the build process ​and you can see 
specifically which tests are failing and which tests have been corrected in the 
Rackspace branch.


-jOrGe W.



________________________________
From: Mukul Gandhi <muk...@apache.org>
Sent: Friday, June 24, 2016 3:58 AM
To: j-users@xerces.apache.org
Cc: s...@tunnelvisionlabs.com
Subject: Re: Interaction between redefine+include breaks augmenting a base 
schema's element with new attributes

Hi Jorge,
   We (that includes I think me, Jesper & Dave Carver) had done various fixes 
earlier to the originally contributed PsycoPath library that Xerces uses for 
its XML Schema 1.1 implementation. I can say with much responsibility (being a 
previous committer to Eclipse WTP source editing, and also as a PMC member of 
Xerces), that the current jar file for PsycoPath, on Xerces SVN conforms fully 
to the W3C XML Schema 1.1 test suite.

You say, '.. XPath 2 implementation reports over 800 failures in the W3C test 
suite'. I'm curious what W3C XML Schema 1.1 tests, you and Sam found that 
PsycoPath library used by Xerces does not pass? Our XML Schema 1.1 conformance 
reports are located at: 
http://wiki.apache.org/xerces/xmlschema11TestSuiteReports. I believe, this is 
also indirectly linked from the XML Schema 1.1 specification.


On 24 June 2016 at 02:49, Jorge Williams 
<jorge.willi...@rackspace.com<mailto:jorge.willi...@rackspace.com>> wrote:
Hey Guys,


If we are planning a release of Xerces, may I suggest that we also update 
version of PsycoPath that Xerces is currently using. The current version of the 
XPath 2 implementation reports over 800 failures in the W3C test suite.  Sam 
Harwell has been able to reduce this number to about 34!

I’m in the process of reviewing these fixes — which you can find here:

 
<https://github.com/sharwell/webtools.sourceediting.xpath/pulls>https://github.com/sharwell/webtools.sourceediting.xpath/pulls

Any help with these code review these PRs would be great and it would be 
awesome to include a new build of PsycoPath with the next Xerces release.

Thanks,

-jOrGe W.



> On Jun 17, 2016, at 8:54 AM, Michael Glavassevich 
> <mrgla...@ca.ibm.com<mailto:mrgla...@ca.ibm.com>> wrote:
>
> Thanks David.
>
> Michael Glavassevich
> XML Technologies and WAS Development
> IBM Toronto Lab
> E-mail: mrgla...@ca.ibm.com<mailto:mrgla...@ca.ibm.com>
> E-mail: mrgla...@apache.org<mailto:mrgla...@apache.org>
>
> David Costanzo <david_costa...@yahoo.com<mailto:david_costa...@yahoo.com>> 
> wrote on 06/17/2016 09:37:56 AM:
>
>> From: David Costanzo
>>>> I started with the JIRA release notes [1] and it doesn't look like I
>>>> have permissions to help with this. [snip]
>>
>> From: Michael Glavassevich
>>> I just add you to the list of contributors in JIRA. Hopefully that
> also
>>> granted you permission to update that field.
>>
>> Yes, that works.  I should be able to build up the release notes,
>> now.  I'll start a new thread in the "xerces-devs" list for any
>> conversation specific to the 2.12.0 release, instead of rambling
>> off-topic on this thread.
>>
>>
>>
>> From: David Costanzo
>>
>>>> As mentioned previously, the fix that my organization wants > >
>> (XERCESJ-1591, r1396551) was made in the "xml-schema-1.1-dev"
>>>> branch, not trunk, so that branch would have to be merged back to
>>>> trunk (or the fix duplicated in trunk) in order for a 2.12.0 release
>>>> to benefit my organization.
>>
>>
>> From: Michael Glavassevich
>>> I'll take a look at the original fix. Assuming it applies to XML
> Schema
>>> 1.0, it should have been committed at the same time to the trunk.
>>
>>
>> That would be appreciated, Michael.  Thanks.
>>

--
Regards,
Mukul Gandhi

Reply via email to