Pages and files

classic Classic list List threaded Threaded
17 messages Options
Reply | Threaded
Open this post in threaded view
|

Pages and files

Eric van der Vlist
Hi,

I'd like to serve as files the common files described in the standard
page-flow except when they are located in a specific directory where I
want to serve them through a page...

Is there an other way to do so than to change the standard, top level
page flow to add a regular expression in their path-info that excludes
this specific directory?

Of course, I can do that but I prefer to keep this page flow as standard
as possible!

Thanks,

Eric
--
Weblog:
                 http://eric.van-der-vlist.com/blog?t=category&a=English
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Erik Bruchez
Administrator
Eric van der Vlist wrote:

> Hi,
>
> I'd like to serve as files the common files described in the standard
> page-flow except when they are located in a specific directory where I
> want to serve them through a page...
>
> Is there an other way to do so than to change the standard, top level
> page flow to add a regular expression in their path-info that excludes
> this specific directory?
>
> Of course, I can do that but I prefer to keep this page flow as standard
> as possible!
I think the regexp is the way to go!

The page flow is definitely there to be customized, so don't feel too
bad about it ;-)

-Erik



--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Eric van der Vlist
Le jeudi 17 novembre 2005 à 17:46 +0100, Erik Bruchez a écrit :

> Eric van der Vlist wrote:
> > Hi,
> >
> > I'd like to serve as files the common files described in the standard
> > page-flow except when they are located in a specific directory where I
> > want to serve them through a page...
> >
> > Is there an other way to do so than to change the standard, top level
> > page flow to add a regular expression in their path-info that excludes
> > this specific directory?
> >
> > Of course, I can do that but I prefer to keep this page flow as standard
> > as possible!
>
> I think the regexp is the way to go!
Short term, yes (that's what I have done) but negating pages conditions
in the files conditions looks like a hack :-) ...

A suggestion would be to allow pages and files to be interleaved and to
check them in the order in which they appear in the page flow.

That would let people put the declarations for their repertories before
the file declarations if they want them to take precedence...

> The page flow is definitely there to be customized, so don't feel too
> bad about it ;-)

Sure, but then I'll have to write an XSLT transformation to apply this
customization each time I install a new version, ie pretty often at this
time :-) ...

Eric

> -Erik
>
> pièce jointe document plein texte (message-footer.txt)
> --
> You receive this message as a subscriber of the [hidden email] mailing list.
> To unsubscribe: mailto:[hidden email]
> For general help: mailto:[hidden email]?subject=help
> ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
--
Le premier annuaire des apiculteurs 100% XML!
                                                http://apiculteurs.info/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Erik Bruchez
Administrator
Eric van der Vlist wrote:

> A suggestion would be to allow pages and files to be interleaved and to
> check them in the order in which they appear in the page flow.

> That would let people put the declarations for their repertories before
> the file declarations if they want them to take precedence...

Yes, that would be possible, although not quite trivial to do. And it
would be cleaner than using the regexp.

> Sure, but then I'll have to write an XSLT transformation to apply this
> customization each time I install a new version, ie pretty often at this
> time :-) ...

You don't really need this, do you? The basic configuration looks like this:

<config xmlns="http://www.orbeon.com/oxf/controller"
         xmlns:xu="http://www.xmldb.org/xupdate"
         xmlns:oxf="http://www.orbeon.com/oxf/processors">

     <!-- Files to serve directly -->
     <files path-info="*.gif"/>
     <files path-info="*.css"/>
     <files path-info="*.pdf"/>
     <files path-info="*.js"/>
     <files path-info="*.png"/>
     <files path-info="*.jpg"/>
     <files path-info="*.html" mime-type="text/html"/>
     <files path-info="*.java" mime-type="text/plain"/>
     <files path-info="*.txt" mime-type="text/plain"/>
     <files path-info="*.xq" mime-type="text/plain"/>

     <!-- Your stuff here -->

     <!-- "Not Found" page displayed when no page matches the request
URL -->
     <page id="not-found" path-info="/not-found"
view="/config/not-found.xml"/>

     <epilogue url="/config/epilogue.xpl"/>
     <not-found-handler page="not-found"/>

</config>

And it shouldn't change much.

-Erik




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Eric van der Vlist
Le jeudi 17 novembre 2005 à 18:16 +0100, Erik Bruchez a écrit :

> Eric van der Vlist wrote:
> > Sure, but then I'll have to write an XSLT transformation to apply this
> > customization each time I install a new version, ie pretty often at this
> > time :-) ...
>
> You don't really need this, do you? The basic configuration looks like this:
>
> <config xmlns="http://www.orbeon.com/oxf/controller"
>          xmlns:xu="http://www.xmldb.org/xupdate"
>          xmlns:oxf="http://www.orbeon.com/oxf/processors">
>
>      <!-- Files to serve directly -->
>      <files path-info="*.gif"/>
>      <files path-info="*.css"/>
>      <files path-info="*.pdf"/>
>      <files path-info="*.js"/>
>      <files path-info="*.png"/>
>      <files path-info="*.jpg"/>
>      <files path-info="*.html" mime-type="text/html"/>
>      <files path-info="*.java" mime-type="text/plain"/>
>      <files path-info="*.txt" mime-type="text/plain"/>
>      <files path-info="*.xq" mime-type="text/plain"/>
>
>      <!-- Your stuff here -->
>
>      <!-- "Not Found" page displayed when no page matches the request
> URL -->
>      <page id="not-found" path-info="/not-found"
> view="/config/not-found.xml"/>
>
>      <epilogue url="/config/epilogue.xpl"/>
>      <not-found-handler page="not-found"/>
>
> </config>
>
> And it shouldn't change much.
Except that I need to update the root one that contains the docs and
examples (that I prefer to keep right now) and that this one id probably
more subject to changes!

Eric

--
Freelance consulting and training.
                                            http://dyomedea.com/english/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Erik Bruchez
Administrator
Eric van der Vlist wrote:

>>And it shouldn't change much.
>
>
> Except that I need to update the root one that contains the docs and
> examples (that I prefer to keep right now) and that this one id probably
> more subject to changes!

In that case that makes sense :-)

-Erik




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Alessandro  Vernet
Administrator
In reply to this post by Eric van der Vlist
On 11/17/05, Eric van der Vlist <[hidden email]> wrote:
> A suggestion would be to allow pages and files to be interleaved and to
> check them in the order in which they appear in the page flow.
>
> That would let people put the declarations for their repertories before
> the file declarations if they want them to take precedence...

Hi Eric,

In fact this is what the page flow does. The only thing that prevents
you from interleaving files and pages is the schema. If you relax the
schema, or remove it (I did the latter for the sake of testing), you
will get the behavior you expect.

Now is this a feature that we really want? Should we really relax the
schema and allow interleaving of files and pages? What are the
benefits and drawback of this change?

Alex



--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
--
Follow Orbeon on Twitter: @orbeon
Follow me on Twitter: @avernet
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Eric van der Vlist
Hi Alex,

Le lundi 21 novembre 2005 à 16:26 -0800, Alessandro Vernet a écrit :

> On 11/17/05, Eric van der Vlist <[hidden email]> wrote:
> > A suggestion would be to allow pages and files to be interleaved and to
> > check them in the order in which they appear in the page flow.
> >
> > That would let people put the declarations for their repertories before
> > the file declarations if they want them to take precedence...
>
> Hi Eric,
>
> In fact this is what the page flow does. The only thing that prevents
> you from interleaving files and pages is the schema. If you relax the
> schema, or remove it (I did the latter for the sake of testing), you
> will get the behavior you expect.
Nice!

> Now is this a feature that we really want?

I'd say yes...

> Should we really relax the
> schema and allow interleaving of files and pages? What are the
> benefits and drawback of this change?

The risks seem to be limited : since the current schema imposes this
order, all the existing page flow will have files defined before pages
and work as before.

The benefit is that you *can* override a file definition with a page
definition by defining your page before the file. You don't have to do
it and can continue to define pages after files if you want to get the
current behavior.

The only drawback I can see is for beginners who could locate their
modifications at the top of an existing page flow without thinking that
they would override file definitions. That could be fixed by a comment
in the page flows of your standard distribution.

Eric

--
Le premier annuaire des apiculteurs 100% XML!
                                                http://apiculteurs.info/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Erik Bruchez
Administrator
Eric van der Vlist wrote:

>>Now is this a feature that we really want?
>
> I'd say yes...
>
>
>>Should we really relax the
>>schema and allow interleaving of files and pages? What are the
>>benefits and drawback of this change?
>
>
> The risks seem to be limited : since the current schema imposes this
> order, all the existing page flow will have files defined before pages
> and work as before.
>
> The benefit is that you *can* override a file definition with a page
> definition by defining your page before the file. You don't have to do
> it and can continue to define pages after files if you want to get the
> current behavior.
>
> The only drawback I can see is for beginners who could locate their
> modifications at the top of an existing page flow without thinking that
> they would override file definitions. That could be fixed by a comment
> in the page flows of your standard distribution.
I added an RFE:

http://forge.objectweb.org/tracker/index.php?func=detail&aid=304316&group_id=168&atid=350210

The next step would be to try changing the schema and making sure the
behavior is correct, then update the doc.

-Erik



--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Alessandro  Vernet
Administrator
On 11/25/05, Erik Bruchez <[hidden email]> wrote:
> I added an RFE:
>
> http://forge.objectweb.org/tracker/index.php?func=detail&aid=304316&group_id=168&atid=350210
>
> The next step would be to try changing the schema and making sure the
> behavior is correct, then update the doc.

Of course Eric, if you or anyone here wants to do those changes, we
can then include them in the build.

Alex



--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
--
Follow Orbeon on Twitter: @orbeon
Follow me on Twitter: @avernet
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Eric van der Vlist
Le lundi 28 novembre 2005 à 18:30 -0800, Alessandro Vernet a écrit :

> On 11/25/05, Erik Bruchez <[hidden email]> wrote:
> > I added an RFE:
> >
> > http://forge.objectweb.org/tracker/index.php?func=detail&aid=304316&group_id=168&atid=350210
> >
> > The next step would be to try changing the schema and making sure the
> > behavior is correct, then update the doc.
>
> Of course Eric, if you or anyone here wants to do those changes, we
> can then include them in the build.
OK, I'll see what I can do!

Thanks,

Eric
--
Freelance consulting and training.
                                            http://dyomedea.com/english/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Eric van der Vlist
In reply to this post by Alessandro Vernet
Le lundi 28 novembre 2005 à 18:30 -0800, Alessandro Vernet a écrit :

> On 11/25/05, Erik Bruchez <[hidden email]> wrote:
> > I added an RFE:
> >
> > http://forge.objectweb.org/tracker/index.php?func=detail&aid=304316&group_id=168&atid=350210
> >
> > The next step would be to try changing the schema and making sure the
> > behavior is correct, then update the doc.
>
> Of course Eric, if you or anyone here wants to do those changes, we
> can then include them in the build.
That won't be possible without loosing some error detection :-( ...

Page flows are currently described with a W3C XML Schema.

They have a model where, under a root element, you have zero or more
files elements, zero or more page elements, an optional epilogue and an
optional not-found-handler element.

With WXS, if you want to allow these elements in any order, you can't
use the "all" compositor since some of these elements can happen several
tiles and you have to replace your sequence by a choice.

Unfortunately, that means that you don't control the number of
occurrences of the epilogue and not-found-handler elements and that you
can have several of them.

If the PFC controller can easily cope with that that's OK. Otherwise
that's a problem...

You could replace your schema by a RNG schema (which hasn't these
restrictions) but then you would miss the integrity checks (unique and
keyref).

RNG supports ID/IDREF checks and that can be used for most of your the
integrity checks but that would be a problem for:

      * path-info-unique (that doesn't seem very useful anyway since it
        doesn't catch regexp overlaps)
      * result-when-unique and action-when-unique which do not catch
        XPath overlap either.

I think that moving to RNG would probably be the best way to go...

What do you think?

Eric

--
Carnet web :
           http://eric.van-der-vlist.com/blog?t=category&a=Fran%C3%A7ais
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Eric van der Vlist
Le mercredi 30 novembre 2005 à 21:54 +0100, Eric van der Vlist a écrit :

> RNG supports ID/IDREF checks and that can be used for most of your the
> integrity checks but that would be a problem for:
>
>       * path-info-unique (that doesn't seem very useful anyway since it
>         doesn't catch regexp overlaps)
>       * result-when-unique and action-when-unique which do not catch
>         XPath overlap either.
>
> I think that moving to RNG would probably be the best way to go...

I had forgotten that you're using MSV to validate your documents.
Another option is to use its ability to accept Schematron rules embedded
in schemas. That would allow to implement all the integrity checks (and
more if needed).

Eric

--
Weblog:
                 http://eric.van-der-vlist.com/blog?t=category&a=English
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Erik Bruchez
Administrator
In reply to this post by Eric van der Vlist
Eric van der Vlist wrote:

> That won't be possible without loosing some error detection :-( ...
>
> Page flows are currently described with a W3C XML Schema.
>
> They have a model where, under a root element, you have zero or more
> files elements, zero or more page elements, an optional epilogue and an
> optional not-found-handler element.
>
> With WXS, if you want to allow these elements in any order, you can't
> use the "all" compositor since some of these elements can happen several
> tiles and you have to replace your sequence by a choice.
>
> Unfortunately, that means that you don't control the number of
> occurrences of the epilogue and not-found-handler elements and that you
> can have several of them.
>
> If the PFC controller can easily cope with that that's OK. Otherwise
> that's a problem...
>
> You could replace your schema by a RNG schema (which hasn't these
> restrictions) but then you would miss the integrity checks (unique and
> keyref).
>
> RNG supports ID/IDREF checks and that can be used for most of your the
> integrity checks but that would be a problem for:
>
>       * path-info-unique (that doesn't seem very useful anyway since it
>         doesn't catch regexp overlaps)
>       * result-when-unique and action-when-unique which do not catch
>         XPath overlap either.
>
> I think that moving to RNG would probably be the best way to go...
>
> What do you think?
We used to have an RNG schema, but I we got rid of it because XML Schema
was required by editors such as Eclipse / Orbeon Studio, IntelliJ, etc.,
and maintaining two schemas was not optimal.

It looks like we have to revive the RNG schema and use that at runtime,
and go back to maintaining two schemas, assuming we want the extra error
detection.

-Erik



--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Erik Bruchez
Administrator
In reply to this post by Eric van der Vlist
Eric van der Vlist wrote:

> Le mercredi 30 novembre 2005 ? 21:54 +0100, Eric van der Vlist a ?crit :
>
>
>>RNG supports ID/IDREF checks and that can be used for most of your the
>>integrity checks but that would be a problem for:
>>
>>      * path-info-unique (that doesn't seem very useful anyway since it
>>        doesn't catch regexp overlaps)
>>      * result-when-unique and action-when-unique which do not catch
>>        XPath overlap either.
>>
>>I think that moving to RNG would probably be the best way to go...
>
>
> I had forgotten that you're using MSV to validate your documents.
Would you recommend another RNG validator? We actually may be using Jing
right now.

RNG didn't use to support ID/IDREF constraints, or checking for unique
keys either, right?

> Another option is to use its ability to accept Schematron rules embedded
> in schemas. That would allow to implement all the integrity checks (and
> more if needed).

Is this a feature of MSV?

-Erik



--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Eric van der Vlist
In reply to this post by Erik Bruchez
Le jeudi 01 décembre 2005 à 12:14 +0100, Erik Bruchez a écrit :
> Eric van der Vlist wrote:

> > RNG supports ID/IDREF checks and that can be used for most of your the
> > integrity checks but that would be a problem for:
> >
> >       * path-info-unique (that doesn't seem very useful anyway since it
> >         doesn't catch regexp overlaps)
> >       * result-when-unique and action-when-unique which do not catch
> >         XPath overlap either.
> >
> > I think that moving to RNG would probably be the best way to go...
> >
> > What do you think?
>
> We used to have an RNG schema, but I we got rid of it because XML Schema
> was required by editors such as Eclipse / Orbeon Studio, IntelliJ, etc.,
> and maintaining two schemas was not optimal.
Yes, that's an issue.

I am using oXygen (either standalone or in Eclipse) which nicely
supports RNG, but I understand that this is not an option for everyone.

> It looks like we have to revive the RNG schema and use that at runtime,
> and go back to maintaining two schemas, assuming we want the extra error
> detection.

Or keep the current limitation that files elements need to be included
before page elements, yes.

Eric
--
Did you know it? Python has now a Relax NG (partial) implementation.
                                          http://advogato.org/proj/xvif/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws
Reply | Threaded
Open this post in threaded view
|

Re: Pages and files

Eric van der Vlist
In reply to this post by Erik Bruchez
Le jeudi 01 décembre 2005 à 12:20 +0100, Erik Bruchez a écrit :

> Eric van der Vlist wrote:
> > Le mercredi 30 novembre 2005 à 21:54 +0100, Eric van der Vlist a écrit :
> >
> >
> >>RNG supports ID/IDREF checks and that can be used for most of your the
> >>integrity checks but that would be a problem for:
> >>
> >>      * path-info-unique (that doesn't seem very useful anyway since it
> >>        doesn't catch regexp overlaps)
> >>      * result-when-unique and action-when-unique which do not catch
> >>        XPath overlap either.
> >>
> >>I think that moving to RNG would probably be the best way to go...
> >
> >
> > I had forgotten that you're using MSV to validate your documents.
>
> Would you recommend another RNG validator? We actually may be using Jing
> right now.
I'd suspect Jing of being faster but haven't made any test to support
this statement...

> RNG didn't use to support ID/IDREF constraints, or checking for unique
> keys either, right?

RNG supports ID/IDREF constraints (that's part of the "DTD compatibility
feature) but doesn't support unique/key/keyref which are specific to
WXS.

> > Another option is to use its ability to accept Schematron rules embedded
> > in schemas. That would allow to implement all the integrity checks (and
> > more if needed).
>
> Is this a feature of MSV?

Yes, through an addon :
http://www.sun.com/software/xml/developers/schematronaddon/index.xml

With Jing, you'd have to use the NRL support to do something similar but
that seems to be an overkill.

Eric

--
Le premier annuaire des apiculteurs 100% XML!
                                                http://apiculteurs.info/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------




--
You receive this message as a subscriber of the [hidden email] mailing list.
To unsubscribe: mailto:[hidden email]
For general help: mailto:[hidden email]?subject=help
ObjectWeb mailing lists service home page: http://www.objectweb.org/wws