The same errors are still in the 1.1 version of the schema. In fact, in
1.1 two more attributes than were in Dave's fix have the issue. Anybody
have insight into why Oasis hasn't fixed this?

Bob

Dave Pawson wrote:
> On 27/05/06, John Cowan <cowan@...> wrote:
>
>>Dave Pawson scripsit:
>>
>>
>>>Error reported is
>>>
>>>/files/odf/ODFWrSchema.rng:17425:19: error: conflicting ID-types for
>>>attribute "targetElement" from namespace
>>>"urn:oasis:names:tc:opendocument:xmlns:smil-compatible:1.0" of element
>>>"command" from namespace
>>>"urn:oasis:names:tc:opendocument:xmlns:animation:1.0"
>>>
>>>/files/odf/ODFWrSchema.rng:10197:22: error: conflicting ID-types for
>>>attribute "targetElement" from namespace
>>>"urn:oasis:names:tc:opendocument:xmlns:smil-compatible:1.0" of element
>>>"command" from namespace
>>>"urn:oasis:names:tc:opendocument:xmlns:animation:1.0"
>>>
>>>
>>>I take it that's an error in the schema?
>>
>>Yes, it is.  What's happened is that an attribute somewhere is being declared
>>of type ID, and then you are using an "anyAttribute" wildcard and implicitly
>>declaring all attributes to be of type "text", including the one you said was
>>of type ID earlier.  An attribute name can't be both ID and not-ID
simultaneously.
>>
>>The cure is to exclude your ID-type attribute by name from the anyAttribute
>>wildcard.
>
>
> Which is fine up to a point :-)
>
>  <define name="anyAttListOrElements">
>       <zeroOrMore>
>          <attribute>
>             <anyName>
> 	      <except>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:smil-compatible:1.0"
>
>>targetElement</name>
>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:form:1.0" >id</name>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:text:1.0">id</name>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0" >id</name>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:animation:1.0" >id</name>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0">shape-id</name>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:presentation:1.0"
>
>>master-element</name>
>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:text:1.0" >change-id</name>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0"
>
>>start-shape</name>
>
> 	 <name
ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0">end-shape</name>
> 	 <name ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0">control</name>
> 	      </except>
> 	    </anyName>
>             <text/>
>          </attribute>
>       </zeroOrMore>
>       <ref name="anyElements"/>
>    </define>
>    <define name="anyElements">
>       <zeroOrMore>
>          <element>
>             <anyName/>
>             <mixed>
>                <ref name="anyAttListOrElements"/>
>             </mixed>
>          </element>
>       </zeroOrMore>
>    </define>
>
> and
>
>   <define name="mathMarkup">
>      <oneOrMore>
>        <element>
> 	 <anyName/>
> 	 <attribute>
> 	   <anyName>
> 	     <except>
> 	       <name ns="urn:oasis:names:tc:opendocument:xmlns:smil-compatible:1.0"
>
>>targetElement</name>
>
> 	       <name ns="urn:oasis:names:tc:opendocument:xmlns:form:1.0" >id</name>
> 	       <name ns="urn:oasis:names:tc:opendocument:xmlns:text:1.0">id</name>
> 	       <name ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0"
>id</name>
> 	       <name ns="urn:oasis:names:tc:opendocument:xmlns:animation:1.0"
>
>>id</name>
>
> 	       <name
ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0">shape-id</name>
> 	       <name ns="urn:oasis:names:tc:opendocument:xmlns:presentation:1.0"
>
>>master-element</name>
>
> 	       <name ns="urn:oasis:names:tc:opendocument:xmlns:text:1.0"
>
>>change-id</name>
>
> 	       <name ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0"
>
>>start-shape</name>
>
> 	       <name
ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0">end-shape</name>
> 	       <name
ns="urn:oasis:names:tc:opendocument:xmlns:drawing:1.0">control</name>
> 	     </except>
> 	   </anyName>
> 	   <text/>
> 	 </attribute>
> 	 <text/>
> 	 <ref name="mathMarkup"/>
>        </element>
>      </oneOrMore>
>    </define>
>
> Something seriously amiss with the ODF schema!
>
>
> Thanks John.
>
> regards
>
>

Hello,

This email message is a notification to let you know that
a file has been uploaded to the Files area of the rng-users
group.

   File        : /single's-lover.htm
   Uploaded by : tallnancylalay <tallnancylalay@...>
   Description : !!The best and largest dating site for tall single,admirers and
friends in the world.

You can access this file at the URL:
http://groups.yahoo.com/group/rng-users/files/single%27s-lover.htm

To learn more about file sharing for your group, please visit:
http://help.yahoo.com/help/us/groups/files

Regards,

tallnancylalay <tallnancylalay@...>

Has anyone done any work with the IDPF container format? I've just
grabbed the schema from the standard, corrected the really obvious
errors and found that it still doesn't validate. Now, I'm not the
world's most experienced RelaxNG user so that might explain my
failings but, given the following, Oxygen gives me an error that my
lack of experience is not helping with...

<?xml version="1.0" encoding="UTF-8"?>
<choice xmlns="http://relaxng.org/ns/structure/1.0">
     <element name="container"
ns='urn:oasis:names:tc:opendocument:xmlns:container'>
         <attribute name="version">
             <value>1.0</value>
         </attribute>
         <element name="rootfiles">
             <oneOrMore>
                 <element name="rootfile">
                     <attribute name="full-path">
                         <text/>
                     </attribute>
                     <attribute name="media-type">
                         <text/>
                     </attribute>
                 </element>
             </oneOrMore>
         </element>
     </element>
     <element name="signatures"
ns='urn:oasis:names:tc:opendocument:xmlns:container'>
         <oneOrMore>
             <element name="Signature"
ns="http://www.w3.org/2001/04/xmldsig#">
                 <externalRef
href="http://www.w3.org/Signature/2002/07/xmldsig-core-schema.rng"/>
             </element>
         </oneOrMore>
     </element>
     <element name="encryption"
ns='urn:oasis:names:tc:opendocument:xmlns:container'
datatypeLibrary="http://www.w3.org/2001/XMLSchema-datatypes'">
         <oneOrMore>
             <choice>
                 <element name="EncryptedData"
ns="http://www.w3.org/2001/04/xmlenc#">
                     <externalRef
href="http://www.w3.org/Encryption/2002/07/xenc-schema.rng"/>
                 </element>
                 <element name="EncryptedKey"
ns="http://www.w3.org/2001/04/xmlenc#">
                     <externalRef
href="http://www.w3.org/Encryption/2002/07/xenc-schema.rng"/>
                 </element>
             </choice>
         </oneOrMore>
     </element>
</choice>

(I hope that reads better than it pastes).

The error is:

Description: conflicting ID-types for attribute "Id" of element
"EncryptedKey" from namespace "http://www.w3.org/2001/04/xmlenc#"

To quote the divine Penelope, "heyulp"

nic gibson

> Has anyone done any work with the IDPF container format?

None whatsoever.


> I've just grabbed the schema from the standard, corrected the
> really obvious errors and found that it still doesn't validate.
> Now, I'm not the world's most experienced RelaxNG user so that
> might explain my failings but, given the following, Oxygen gives me
> an error that my lack of experience is not helping with...
> [file snipped]
> The error is:
> Description: conflicting ID-types for attribute "Id" of element
> "EncryptedKey" from namespace "http://www.w3.org/2001/04/xmlenc#"
> To quote the divine Penelope, "heyulp"

I cannot reproduce your error. My oXygen reports it is valid.
So does `jing` (using the schema that came with nxml-mode :-).

I realize that may be less than helpful, but thought more information
is better than less ...

--- In rng-users@yahoogroups.com, Syd Bauman <Syd_Bauman@...> wrote:
>
> > Has anyone done any work with the IDPF container format?
>
> None whatsoever.
>

It occurs to me that asking if anyone could see a hole in
the schema would have been an equally valid question :)

>
> > I've just grabbed the schema from the standard, corrected the
> > really obvious errors and found that it still doesn't validate.
> > Now, I'm not the world's most experienced RelaxNG user so that
> > might explain my failings but, given the following, Oxygen gives me
> > an error that my lack of experience is not helping with...
> > [file snipped]
> > The error is:
> > Description: conflicting ID-types for attribute "Id" of element
> > "EncryptedKey" from namespace "http://www.w3.org/2001/04/xmlenc#"
> > To quote the divine Penelope, "heyulp"
>
> I cannot reproduce your error. My oXygen reports it is valid.
> So does `jing` (using the schema that came with nxml-mode :-).

Hmmmm... I think I'd better play some more.

>
> I realize that may be less than helpful, but thought more information
> is better than less ...
>

No, it's a good start. thanks :)

nic

nic.corbas scripsit:

> The error is:
>
> Description: conflicting ID-types for attribute "Id" of element
> "EncryptedKey" from namespace "http://www.w3.org/2001/04/xmlenc#"

I haven't actually looked at the schema, but the usual reason for that
problem is that the schema says EncryptedKey/@Id is of type ID, and then
somewhere else there's a pattern allowing any element in any namespace
with any attributes, either everywhere or just in certain places.

The latter pattern declares all attributes to be of type text, and since
IDness is a DTD-compatible property, it's illegal for the same attribute
in the same element to be declared as ID in one place and text in another.

The cure is to find the "any" pattern and limit it so that it doesn't
include elements in the xmlenc namespace.  Alternatively, you can give
jing the -i switch to turn off this behavior.  Technically, IDness is
not part of RELAX NG, so it's not too surprising that some programs
don't complain.

--
John Cowan    cowan@...    http://ccil.org/~cowan
Objective consideration of contemporary phenomena compel the conclusion
that optimum or inadequate performance in the trend of competitive
activities exhibits no tendency to be commensurate with innate capacity,
but that a considerable element of the unpredictable must invariably be
taken into account. --Ecclesiastes 9:11, Orwell/Brown version

On 11/09/2007, John Cowan <cowan@...> wrote:
>
> nic.corbas scripsit:
>
>  > The error is:
>  >
>  > Description: conflicting ID-types for attribute "Id" of element
>  > "EncryptedKey" from namespace "http://www.w3.org/2001/04/xmlenc#"
>
>  I haven't actually looked at the schema, but the usual reason for that
>  problem is that the schema says EncryptedKey/@Id is of type ID, and then
>  somewhere else there's a pattern allowing any element in any namespace
>  with any attributes, either everywhere or just in certain places.
>
>  The latter pattern declares all attributes to be of type text, and since
>  IDness is a DTD-compatible property, it's illegal for the same attribute
>  in the same element to be declared as ID in one place and text in another.
>
>  The cure is to find the "any" pattern and limit it so that it doesn't
>  include elements in the xmlenc namespace.  Alternatively, you can give
>  jing the -i switch to turn off this behavior.  Technically, IDness is
>  not part of RELAX NG, so it's not too surprising that some programs
>  don't complain.

Ah, bells ring. I had some sort of similar problem with DocBook and
you can tell oxygen not to check ID types.

thank ee kindly

nic

--
Nic Gibson
Director, Corbas Consulting
Editorial and Technical Consultancy
http://www.corbas.co.uk/

Hi all,

As a Localization Industry Standards Association (LISA:
http://www.lisa.org) member, I am soon to start implementing RelaxNG
as the schema for a number of document standards based there.
So I am just gathering together references, documentation, and so
forth.

Unfortunately, I have already mentioned using RelaxNG for the new
TMX 2.0 standard at LISA and have been ridiculed by the editor in
charge. Seems some people just cannot listen to reason. Has anybody
had any similar experiences or advice on how to tackle the
situation?

I am going to write an article for LISA about RelaxNG and try to
create schemas for as many of the standards they have before making
a real push. I have also read most of the common stuff around about
RelaxNG. Does anyone know of some extra information off the beaten
path that would make good ammunition?

Thanks.

Raymond Martin (OmegaT+ project manager/developer)

Hi Michael,

Thanks for the information. I'll check them out.

Cheers,

Raymond

On 18 September 2007 05:26:37 you wrote:
>
> > As a Localization Industry Standards Association (LISA:
> > http://www.lisa.org) member, I am soon to start implementing RelaxNG
> > as the schema for a number of document standards based there.
> > So I am just gathering together references, documentation, and so
> > forth.
> >
> > Unfortunately, I have already mentioned using RelaxNG for the new
> > TMX 2.0 standard at LISA and have been ridiculed by the editor in
> > charge. Seems some people just cannot listen to reason. Has anybody
> > had any similar experiences or advice on how to tackle the
> > situation?
>
> A few years back, as part of a discussion thread on an IETF
> mailing list, James Clark posted an e-mail message detailing why
> RELAXNG "should receive at least equal billing" with the W3C XML
> Schema language as a recommended formalism for use in XML-related
> IETF specs. At that time, I wrote up an item for XMLHack about it:
>
>   http://www.xmlhack.com/read.php?item=1677
>
> His main points about the advantages of RELAXNG are summarized in
> the bulleted list in there. His original posting is here:
>
>   http://www.imc.org/ietf-xml-use/mail-archive/msg00217.html
>
> The "A Triumph of Simplicity: James Clark on Markup Languages and
> XML" interview that Eugene Eric Kim did with James back in 2001 is
> also still worth reading and pointing others to:
>
>   http://www.ddj.com/184404686
>
> > I am going to write an article for LISA about RelaxNG and try to
> > create schemas for as many of the standards they have before making
> > a real push. I have also read most of the common stuff around about
> > RelaxNG. Does anyone know of some extra information off the beaten
> > path that would make good ammunition?
>
> Along with the above, I guess it's also worth reminding people
> that RELAXNG is a full and final ISO International Standard, as
> part of ISO/IEC 19757, "DSDL Document Schema Definition
> Languages":
>
>   http://dsdl.org/
>   http://dsdl.org/relaxng-is.pdf
>
> It also should be remembered that RELAXNG has been used as a
> formalism in a number of specifications from W3C working groups,
> including the XHTML working group. Masayasu Ishikawa once posted
> an explanation of the rationale behind choosing RELAXNG for the
> XHTML spec and others -
>
>   http://lists.w3.org/Archives/Public/www-svg/2003May/0069.html
>
> One comment he made in that posting was (speaking from his
> perspective as a spec editor) "I chose to start from RELAX NG
> primarily to make *my* life easier".
>
>   --Mike
>

All,

I am very sorry for not taking actions against spam.  I have
just finished three actions.  First, this ML is now moderated.
Every message has to be approved by the moderators.  Second,
I have removed spammers from the list of participants.  Third,
Dave Pawson kindly volunteered to become a moderator, and he
is now a moderator.

Again, my apologies for poor management.  But everything
will change.

Cheers,


--
MURATA Makoto (FAMILY Given) <EB2M-MRT@...>

Folks,

I have created schemas for the combination of Atom and its extensions.
There are two versions.  One version is solely in RELAX NG, while the
other uses NVDL as well as RELAX NG.

Even when you are not interested in atom, these schemas might be useful
in studying schema customization techniques.

Cheers,

Makoto

Forwarded by "MURATA Makoto (FAMILY Given)" <eb2m-mrt@...>
----------------------- Original Message -----------------------
  From:    "MURATA Makoto (FAMILY Given)" <eb2m-mrt@...>
  To:      "Atom WG" <atom-syntax@...>
  Date:    Mon, 03 Sep 2007 02:15:05 +0900
  Subject: Schema for the combination of Atom and its extensions
----


I find that no schemas examine constraints specified
in RFC 4287 (The Atom Syndication Format),
RFC 4685 (Atom Threading Extensions), "Feed Paging and Archiving",
and Open Search, all together.

I gave a lecture of schema customization at Tsukuba University and
asked the students to try.  On the basis of their result, I created a
schema for the combination of the above specifications.  Furthermore,
other extensions can be easily added to this schema.

The schemas are available at:

	 http://www.asahi-net.or.jp/~eb2m-mrt/atom/schemaForAtomExtensions.zip

The schema "atomRev.rnc" is a rewrite of the schema in RFC 4287. It is
created by Mr. Fujisaki at Tsukuba University, and is slightly modified
by me.


--
MURATA Makoto (FAMILY Given) <EB2M-MRT@...>



--------------------- Original Message Ends --------------------


Forwarded by "MURATA Makoto (FAMILY Given)" <eb2m-mrt@...>
----------------------- Original Message -----------------------
  From:    "MURATA Makoto (FAMILY Given)" <eb2m-mrt@...>
  To:      Atom WG <atom-syntax@...>
  Date:    Tue, 11 Sep 2007 12:19:30 +0900
  Subject: Re: Schema for the combination of Atom and its extensions
----


The first version of the schema is written solely in RELAX NG.  I
have created the second version using NVDL and RELAX NG.  Simply put,
NVDL decomposes atom feeds in to pieces so that tiny schemas can be used
for the validation of the pieces.

I believe that the NVDL + RELAX NG version is easier to undestand and
maintain.

Both versions are available at

	 http://www.asahi-net.or.jp/~eb2m-mrt/atom/schemaForAtomExtensions.zip

Cheers,

--
MURATA Makoto (FAMILY Given) <EB2M-MRT@...>



--------------------- Original Message Ends --------------------

I'm experimenting with documentation of an rng schema that I'm
working on.

The only tool I could find to produce documentation from rng directly
is
         http://www.techquila.com/rng-tools.html

It seems to work okay, but it's slow and does not embed figures.

So I've been converting rng -> w3c schema and using oxygen's xsl for
documentation - which produced nice output with embedded figures.

It seems to work with the DTD compatibility <a:documentation>  tags,
but not with plain comments.

How does one embed URLs into documentation like that?

What are people using to comment and generate documentation about
relaxng schema?

thanks!

Brad

> I'm experimenting with documentation of an rng schema that I'm
>  working on.
>
>  The only tool I could find to produce documentation from rng directly
>  is
>          http://www.techquila.com/rng-tools.html
>
>  It seems to work okay, but it's slow and does not embed figures.
What figures are there in your schema?

>  What are people using to comment and generate documentation about
>  relaxng schema?


Norm Walsh has something he's developed for the definitive guide,
docbook v5, rng.

I don't think he's released the code though.

regards


--
Dave Pawson
XSLT XSL-FO FAQ.
http://www.dpawson.co.uk

> >  What are people using to comment and generate documentation
> >  about relaxng schema?

There is, of course, the TEI Guidelines and its chapter on
"Documentation Elements", aka ODDs:
http://www.tei-c.org/release/doc/tei-p5-doc/html/TD.html.

But the TEI is not about documenting a RELAX NG schema, rather it is
about writing an abstract encoding system in such a way that through
tangle and weave processes both prose documentation and a formal
schema, likely in RELAX NG, can be produced. The major drawback is
that there are RELAX NG constructs (like co-constraint) that cannot
be expressed in ODD.

(The TEI also limits itself to deterministic content models and won't
use interleave, but that isn't required by the ODD language, and
personally I hope we drop these restrictions soon.)

On Thu, 2007-09-27 at 18:28 +0100, Dave Pawson wrote:
> > I'm experimenting with documentation of an rng schema that I'm
> >  working on.
> >
> >  The only tool I could find to produce documentation from rng directly
> >  is
> >          http://www.techquila.com/rng-tools.html
> >
> >  It seems to work okay, but it's slow and does not embed figures.
> What figures are there in your schema?

the xs3p xsl + modifications from oxygen makes
little figures like the one i'm attaching here:

>
> >  What are people using to comment and generate documentation about
> >  relaxng schema?
>
>
> Norm Walsh has something he's developed for the definitive guide,
> docbook v5, rng.

searched his site... no luck though.
any hint about where?

brad

On 28/09/2007, Brad Langhorst <brad@...> wrote:


>  > Norm Walsh has something he's developed for the definitive guide,
>  > docbook v5, rng.
>
>  searched his site... no luck though.
>  any hint about where?

I don't think I've seen the code.
http://www.docbook.org/tdg5/en/html/chapter.html shows the html output though.

Content Model

chapter ::=

etc.

regards




--
Dave Pawson
XSLT XSL-FO FAQ.
http://www.dpawson.co.uk

I'm trying to build some high quality schemas to define an interchange
format for analytical ultracentrifuge data.

The instrument acquires optical (or other) scans of a sample...

So I have a scan object defined

     <define name="Scan">
         <optional>
             <element name="ExperimentURI">
                 <data type="anyURI"/>
             </element>
         </optional>
  ...
         <element name="ScanData">
             <data type="hexBinary"/>
         </element>
     </define>

Thinking that I'm going to have it refer back to the experiment it was
acquired in using that uri field.

Is this reasonable?

The experiment schema refers to a set of scans
the analysis schema also refers to a set of scans.

Usually software that must deal with experiments does not need to deal
with analyses.


Should I embed the definition of a scan in both rng files or have them
refer to a shared file?

Or maybe I should put all three constructs into one file?

On 10/10/2007, Brad Langhorst <brad@...> wrote:
>
> I'm trying to build some high quality schemas to define an interchange
>  format for analytical ultracentrifuge data.
>
>  The instrument acquires optical (or other) scans of a sample...
>
>  So I have a scan object defined
>
>  <define name="Scan">
>          <optional>
>              <element name="ExperimentURI">
>                  <data type="anyURI"/>
>              </element>
>          </optional>
>   ...
>          <element name="ScanData">
>              <data type="hexBinary"/>
>          </element>
>      </define>
>
>  Thinking that I'm going to have it refer back to the experiment it was
>  acquired in using that uri field.
>
>  Is this reasonable?

Yes, I talk about X now, I want to refer back to X later.


>  Should I embed the definition of a scan in both rng files or have them
>  refer to a shared file?
>
>  Or maybe I should put all three constructs into one file?

Or use cross references to link from one to the other?
If you wrote the XML, the annoying part would be re-typing
the url two or three times. It would be easier to write it once,
identify it with an id value, then cross reference to it when
you want to talk about it again?

HTH




--
Dave Pawson
XSLT XSL-FO FAQ.
http://www.dpawson.co.uk

Hi,

A new major version of oXygen is available from our website
http://www.oxygenxml.com

There are a few Relax NG related additions in oXygen XML editor 9.0:

* Better error reporting for XML validation against Relax NG schemas

Incomplete error reports like "required attributes missing", "required
elements missing" or "unfinished element" are now updated to report
exactly what XML attributes or elements are missing and what is expected
to finish the element.

* Better error reporting for Relax NG schema validation

The errors about overlapping names inside interleave are reporting now
specifying the actual names that are overlapping.

The main new addition in version 9.0 is a CSS-based visual XML editor
allowing WYSIWYG-like editing of XML documents. Support for major
document frameworks like DocBook, DITA, TEI and XHTML in built-in and
users can configure and share support for other document types.

For a quick preview of the visual editing support watch the following 5
minutes video demonstration:
http://www.oxygenxml.com/demo/WYSIWYG/WYSIWYG_XML_Editing.html

The complete list with the new additions and more details are available from
http://www.oxygenxml.com/index.html#new-version

Best Regards,
George
---------------------------------------------------------------------
George Cristian Bina - http://aboutxml.blogspot.com/
<oXygen/> XML Editor, Schema Editor and XSLT Editor/Debugger
http://www.oxygenxml.com

Hi!

I've been googling for RNG versions of the DITA schemas without finding
anything.

Does anyone know of a place where RNG schemas for DITA can be found?

I've tried creating RNC schemas from the DITA DTDs using trang, but they
fail in nxml.  There's a complaint of duplicate definition, like eg. this:

File mode specification error: (error "Duplicate definition of task-info-types
at position 3505 in
~/workspaces/project/users_guide/schemas/relaxng/task.mod.rnc")

Line 3505 is the last line of task.mod.rnc, it contains only one line
with "element task", and according to grep, that's the only line with
"element task" in the entire directory.

Thanx!


- Steinar

Steinar Bang wrote:

   [ sorry for the multi-post on RNG List and nXML List,
     where this question was asked ]

> File mode specification error: (error "Duplicate
> definition of task-info-types at position 3505 in
> ~/workspaces/project/users_guide/schemas/relaxng/task.mod.rnc")

> Line 3505 is the last line of task.mod.rnc, it contains
> only one line with "element task", and according to grep,
> that's the only line with "element task" in the entire
> directory.

   Actually, that's _position_ 3505.  That means the 3505-th
character in the buffer.  That's the position of the first
character after the definition of task-info-types.  Below is
what I thought about your problem:

   1/ the Dita DTD modules use entity calls to modularize the
      DTDs, with a default definition of the entity

   2/ task.mod uses (and defines) the entity task-info-types,
      this entity can be "overriden" by an including DTD

   3/ ditabase.dtd includes task.mod without defining the
      entity, so in the resulting tasl.mod.rnc generated by
      Trang, you have the definition of an identifier named
      task-info-types

   4/ task.dtd includes task.mod and defines the entity.
      This is legal with DTD, this is the way to override
      default values in DTD composition.  Because the
      corresponding identifier is already defined in
      task.rnc, Trang doesn't define it again in task.mod.rnc

   5/ so if you use separated sets of Trang-generated RNC
      files (one set per top-level generation: one for
      ditabase.dtd, one for task.dtd...), everything's ok

   6/ But I guess you generated in the same output directory,
      once for task.dtd then once for ditabasse.dtd (for
      instance), so the result was a task.rnc including
      task.mod.rnc, but this later was overriden by the
      subsequent generation and contains now the definition
      of task-info-types

   Could you please try to generate again your RNC files with
Trang, but this time putting each output in a different
directory, and modifying your schemas.xml in accordance?

   Regards,

--drkm





















      
_____________________________________________________________________________
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail

* Steinar Bang wrote:
>I've tried creating RNC schemas from the DITA DTDs using trang, but they
>fail in nxml.  There's a complaint of duplicate definition, like eg. this:

http://lists.w3.org/Archives/Public/www-archive/2005Feb/0029.html has a
xmllint-based script that should expand all entities; that should make
it a lot simpler for trang to do the conversion (at the expense of some
structure).
--
Björn Höhrmann · mailto:bjoern@... · http://bjoern.hoehrmann.de
Weinh. Str. 22 · Telefon: +49(0)621/4309674 · http://www.bjoernsworld.de
68309 Mannheim · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/

Greetings all
(blame Tommie Usdin for directing me here ...).

I'm always happy to try out a new technology like Relax,
especially when people I respect have cogent arguments
for favoring it over something previous like the W3C schema language.

I'm waist-deep in a project which started out as an independent system
so we have a collection of workable relax ng schemas in testing phase.
Then the additional requirement came down the road to look at
compliance with the ISO 19139 standard, geographic metadata in XML
encoding, and that document supplies a collection of .xsd schemas in
an ISO namespace.

I hate to take the easy road, abandon Relax and translate everything
back into .xsd.  but I need to do various things like include or
import elements and complex types from the ISO schemas into the Relax
schemas and make the whole system validate.  Can this be done?  Is
there a good tutorial available which discusses the nuts and bolts?

thanks,
John

chistletoe1 wrote:
>  but I need to do various things like include or
> import elements and complex types from the ISO schemas into the Relax
> schemas and make the whole system validate.  Can this be done?
you can't mix and match XSD and RELAXNG, no.
you'd have to statically convert the XSD to RELAX,
which is not always easy.

alternatively (and now it gets tricksy), you
can validate difference pieces of the document
with different schemas in different languages.
so you write an NVDL schema which says
"validate stuff in namespace X
against schema Y, stuff in namespace A against
schema B".

some people get this straightaway and love it,
others recoil in horror.

--
Sebastian Rahtz

Information Manager, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

chistletoe1 wrote:

> I'm waist-deep in a project which started out as an independent system
> so we have a collection of workable relax ng schemas in testing phase.
> Then the additional requirement came down the road to look at
> compliance with the ISO 19139 standard, geographic metadata in XML
> encoding, and that document supplies a collection of .xsd schemas in
> an ISO namespace.

You can't combine RNG and WXS in one schema, you have to convert WXS to
RNG (or vice versa) before combining them together.

> I hate to take the easy road, abandon Relax and translate everything
> back into .xsd.  but I need to do various things like include or
> import elements and complex types from the ISO schemas into the Relax
> schemas and make the whole system validate.  Can this be done?  Is
> there a good tutorial available which discusses the nuts and bolts?

If ISO 19139 defines its elements in a separate namespace you can use
NVDL (see http://nvdl.org for links to implementations). NVDL allows you
to validate different parts of document against separate schemas -- each
schema can be in different schema language.

(And yes NVDL is also ISO standard, so you can be compliant with three
ISO standards -- RELAX NG, NVDL and ISO 19139 ;-D)

				 Jirka

--
------------------------------------------------------------------
   Jirka Kosek      e-mail: jirka@...      http://xmlguru.cz
------------------------------------------------------------------
        Professional XML consulting and training services
   DocBook customization, custom XSLT/XSL-FO document processing
------------------------------------------------------------------
  OASIS DocBook TC member, W3C Invited Expert, ISO JTC1/SC34 member
------------------------------------------------------------------

>>>>> Bjoern Hoehrmann <derhoermi@...>:

> * Steinar Bang wrote:
>> I've tried creating RNC schemas from the DITA DTDs using trang, but they
>> fail in nxml.  There's a complaint of duplicate definition, like eg. this:

> http://lists.w3.org/Archives/Public/www-archive/2005Feb/0029.html has
> a xmllint-based script that should expand all entities; that should
> make it a lot simpler for trang to do the conversion (at the expense
> of some structure).

I first tried the approach suggested by Florent Georges, ie. generate
the RNC for each DTD into separate directories.

Then I tried the script suggested by Björn Höhrmann (a bourne shell
version of the script is at the end of the message), used it to generate
DTDs with expanded entities, that then were run through trang to create
single-file RNC schemas.

The resulting schemas from both approaches gave me the same problem:

The schemas loaded without error messages (which was an improvement :-) ).

But instead of <task> or <concept> (the only two document types I
tested) as the top directory, both DTDs had <no-topic-nesting> with no
children as the only legal element.

So... I'm open for more suggestions...:-)


- Steinar


Here's a bourne shell version of Björn Höhrmann's script:

#!/bin/sh
#
# Expand entities in DTD files
#
DTDFILE=$1
perl -e 'print qq{<!DOCTYPE x [<!ENTITY \x25 x SYSTEM "$ARGV[0]">\x25x;]><x/>}'
$DTDFILE > $DTDFILE.tmp
xmllint --noent $DTDFILE.tmp | perl -ne 'local$/;$_=<>;s/^<!DOCTYPE x
\[//;s/]>\s*<x\/>$//;s/<!--.*?-->//sg;s/<!ENTITY.*?>//sg; s/\n+/\n/g; print'
rm $DTDFILE.tmp

Check out the Source Forge page: http://sourceforge.net/projects/relaxng/


Daniel C wrote:
> i've become a fan of RNG....   Are there any RNG to XSD conversion
> tools out there that work on the fly in .Net?
>
> i am aware of Trang.   Are there any alternatives, or ports to .Net?
>
> Thanks!
>
>    - Daniel
>

On Dec 3, 2007 9:26 AM, Daniel C <danielcor@...> wrote:

> I am in fact a user of Tenuto.   I use it for it's intended purpose to
> validate XML, and it works well for me.   It doesn't however, do the
> conversion to .xsd, which is what I am looking for.   Any other suggestions?

Use Trang and don't worry about converting "on-the-fly". Is there any
particular reason why you require that?

Bruce