- From: Loren Cahlander <loren.cahlander@gmail.com>
- Date: Wed, 30 Mar 2016 11:56:31 -0400
- To: Michael Kay <mike@saxonica.com>
- Cc: "Costello, Roger L." <costello@mitre.org>, "xmlschema-dev@w3.org" <xmlschema-dev@w3.org>
- Message-Id: <9E6596F7-B575-4861-ADE5-06A745D14618@gmail.com>
Absolutely. It was just an off the cuff attempt. It would take developing and Query or XSLT program to identify the unused simple types. > On Mar 30, 2016, at 11:34 AM, Michael Kay <mike@saxonica.com> wrote: > > Shows the importance of stating requirements unambiguously! Loren has interpreted the question as meaning "simple types that are not referenced from anywhere in the schema", whereas I interpreted it as "simple types that are not used in the course of a particular validation episode". > > But even if Loren's interpretation is correct, his solution is incomplete: it doesn't allow for multiple schema documents, it doesn't allow for @type having a namespace prefix, and it doesn't allow for references to a type from another type, e.g. in @base, @itemType, or @memberTypes. > > Michael Kay > Saxonica > >> On 30 Mar 2016, at 16:24, Loren Cahlander <loren.cahlander@gmail.com <mailto:loren.cahlander@gmail.com>> wrote: >> >> Try the following (untested) snippet >> >> fn:distinct-values($schema//xs:simpleType/@name[not(.=$schema//xs:element/@type/string())]) >> >> Loren Cahlander >> >> On Mar 28, 2016, at 1:19 PM, Michael Kay <mike@saxonica.com <mailto:mike@saxonica.com>> wrote: >> >>> Saxon's schema processor has the option -stats:filename to report on the usage of schema components during a validation episode. IIRC it only lists the components that were used (not those that weren't) but you could do some post-processing to derive this information. >>> >>> Michael Kay >>> Saxonica >>> >>> >>>> On 28 Mar 2016, at 17:42, Costello, Roger L. <costello@mitre.org <mailto:costello@mitre.org>> wrote: >>>> >>>> Hi Folks, >>>> >>>> I am seeking a Java tool that finds all unused simpleTypes in an XML Schema. >>>> >>>> Does such a tool exist? >>>> >>>> /Roger >>>> >>> >>> >>> >
Received on Wednesday, 30 March 2016 15:57:02 UTC