Eclipse Community Forums: EMF » XML Schema and upper/lower bounds

Help

Home

Home » Modeling » EMF » XML Schema and upper/lower bounds

Show: Today's Messages :: Show Polls :: Message Navigator

XML Schema and upper/lower bounds [message #900031]

Fri, 03 August 2012 14:10

Jesper Eskilson

Messages: 134
Registered: July 2009

Senior Member

Hi,

I'm trying to write an XML schema for an existing XML file format, so
that I can load these XML files as EMF models, but I'm having problem in
getting the upper/lower bounds correct. For example:

> <xs:element name="project">
> <xs:complexType>
> <xs:choice maxOccurs="unbounded">
> <xs:element name="fileVersion" type="xs:string"
> minOccurs="1" maxOccurs="1" />
> <xs:element name="configuration" type="configuration"
> maxOccurs="unbounded" />
> <xs:element name="group" type="group" maxOccurs="unbounded" />
> <xs:element name="file" type="file" maxOccurs="unbounded" />
> </xs:choice>
> </xs:complexType>
> </xs:element>

This results in a EClass "ProjectType" with a multivalued "fileVersion"
reference of type "EList<String>", but I would like "fileVersion" to be
an *attribute* of type String. Why isn't "maxOccurs" honored by EMF?

Regards,

--
/Jesper

Report message to a moderator

Re: XML Schema and upper/lower bounds [message #900058 is a reply to message #900031]

Fri, 03 August 2012 17:49

Ed Merks

Messages: 33140
Registered: July 2009

Senior Member

Jesper,

Comments below.

On 03/08/2012 4:10 PM, Jesper Eskilson wrote:
> Hi,
>
> I'm trying to write an XML schema for an existing XML file format, so
> that I can load these XML files as EMF models, but I'm having problem
> in getting the upper/lower bounds correct. For example:
>
>> <xs:element name="project">
>> <xs:complexType>
>> <xs:choice maxOccurs="unbounded">
>> <xs:element name="fileVersion" type="xs:string"
>> minOccurs="1" maxOccurs="1" />
>> <xs:element name="configuration" type="configuration"
>> maxOccurs="unbounded" />
>> <xs:element name="group" type="group"
>> maxOccurs="unbounded" />
>> <xs:element name="file" type="file"
>> maxOccurs="unbounded" />
>> </xs:choice>
>> </xs:complexType>
>> </xs:element>
>
> This results in a EClass "ProjectType" with a multivalued
> "fileVersion" reference of type "EList<String>", but I would like
> "fileVersion" to be an *attribute* of type String.
Your schema allows an unbounded number of choices and hence there can be
an unbounded number of file version elements.
> Why isn't "maxOccurs" honored by EMF?
Why do you have an unbounded choice? Any bounds within its nested
elements are kind of meaningless from a schema point of view.
>
> Regards,
>
> --
> /Jesper

Ed Merks
Professional Support: https://www.macromodeling.com/

Report message to a moderator

Re: XML Schema and upper/lower bounds [message #900296 is a reply to message #900058]

Mon, 06 August 2012 11:55

Jesper Eskilson

Messages: 134
Registered: July 2009

Senior Member

<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
On 08/03/2012 07:49 PM, Ed Merks wrote:
<blockquote cite="mid:jvh2uk$duk$1@xxxxxxxxe.org" type="cite">Jesper,
 
 
Comments below.
 
 
On 03/08/2012 4:10 PM, Jesper Eskilson wrote:
 
<blockquote type="cite">Hi,
 
 
I'm trying to write an XML schema for an existing XML file
format, so that I can load these XML files as EMF models, but
I'm having problem in getting the upper/lower bounds correct.
For example:
 
 
<blockquote type="cite">    <xs:element name="project">
 
        <xs:complexType>
 
            <xs:choice maxOccurs="unbounded">
 
                <xs:element name="fileVersion"
type="xs:string"
 
                    minOccurs="1" maxOccurs="1" />
 
                <xs:element name="configuration"
type="configuration"
 
                    maxOccurs="unbounded" />
 
                <xs:element name="group" type="group"
maxOccurs="unbounded" />
 
                <xs:element name="file" type="file"
maxOccurs="unbounded" />
 
            </xs:choice>
 
        </xs:complexType>
 
    </xs:element>
 
</blockquote>
 
This results in a EClass "ProjectType" with a multivalued
"fileVersion" reference of type "EList<String>", but I
would like "fileVersion" to be an *attribute* of type String.
 
</blockquote>
Your schema allows an unbounded number of choices and hence there
can be an unbounded number of file version elements.
 
<blockquote type="cite">Why isn't "maxOccurs" honored by EMF?
 
</blockquote>
Why do you have an unbounded choice?  Any bounds within its nested
elements are kind of meaningless from a schema point of view.
 
 
</blockquote>
 
Maybe I'm hitting a limit of how XML schemas work. I want to be able
to specify child elements in any order *and* any multiplicity, but
maybe that is impossible. For example, I would like to allow a set
of child elements A, B, C in any order, such that A and B occurs
exactly once, but C occurs any number of times, but at least once. 
If I fix the order of A and B I can use a sequence, but I would like
to avoid that. 
 
<div class="moz-signature">-- 

Jesper
Eskilson
Development
Engineer 
IAR Systems AB 
Box 23051, Strandbodgatan 1 
SE-750 23 Uppsala, SWEDEN 
Phone: +46 18 16 78 00 
E-mail: <a href="mailto:jesper.eskilson@xxxxxxxx">
jesper.eskilson@xxxxxxxx</a>
Website: <a href="http://www.iar.com">
www.iar.com 
</a>
Twitter: <a href="http://www.twitter.com/iarsystems">
www.twitter.com/iarsystems</a>


</div>
</body>
</html>

Report message to a moderator

Re: XML Schema and upper/lower bounds [message #900313 is a reply to message #900296]

Mon, 06 August 2012 13:01

Ed Merks

Messages: 33140
Registered: July 2009

Senior Member

<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Jesper, 
 
XML Schema has <any> but its use is quite limited and EMF
treats it like a <sequence>.  EMF's deserializer doesn't check
that element are actually in the correct order anyway, and will
write them out in the model's feature order (except where feature
maps are used). 
 
 
<div class="moz-cite-prefix">On 06/08/2012 1:55 PM, Jesper Eskilson
wrote: 
</div>
<blockquote cite="mid:jvobas$l1o$1@xxxxxxxxe.org" type="cite">
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
On 08/03/2012 07:49 PM, Ed Merks wrote:
<blockquote cite="mid:jvh2uk$duk$1@xxxxxxxxe.org" type="cite">Jesper,

 
 
Comments below. 
 
On 03/08/2012 4:10 PM, Jesper Eskilson wrote: 
<blockquote type="cite">Hi, 
 
I'm trying to write an XML schema for an existing XML file
format, so that I can load these XML files as EMF models, but
I'm having problem in getting the upper/lower bounds correct.
For example: 
 
<blockquote type="cite">    <xs:element name="project">
 
        <xs:complexType> 
            <xs:choice maxOccurs="unbounded"> 
                <xs:element name="fileVersion"
type="xs:string" 
                    minOccurs="1" maxOccurs="1" /> 
                <xs:element name="configuration"
type="configuration" 
                    maxOccurs="unbounded" /> 
                <xs:element name="group" type="group"
maxOccurs="unbounded" /> 
                <xs:element name="file" type="file"
maxOccurs="unbounded" /> 
            </xs:choice> 
        </xs:complexType> 
    </xs:element> 
</blockquote>
 
This results in a EClass "ProjectType" with a multivalued
"fileVersion" reference of type "EList<String>", but I
would like "fileVersion" to be an *attribute* of type String.
 
</blockquote>
Your schema allows an unbounded number of choices and hence
there can be an unbounded number of file version elements. 
<blockquote type="cite">Why isn't "maxOccurs" honored by EMF? 
</blockquote>
Why do you have an unbounded choice?  Any bounds within its
nested elements are kind of meaningless from a schema point of
view. 
 
</blockquote>
 
Maybe I'm hitting a limit of how XML schemas work. I want to be
able to specify child elements in any order *and* any
multiplicity, but maybe that is impossible. For example, I would
like to allow a set of child elements A, B, C in any order, such
that A and B occurs exactly once, but C occurs any number of
times, but at least once.  If I fix the order of A and B I can use
a sequence, but I would like to avoid that. 
 
<div class="moz-signature">-- 
 Jesper
Eskilson Development

Engineer 
IAR Systems AB 
Box 23051, Strandbodgatan 1 
SE-750 23 Uppsala, SWEDEN 
Phone: +46 18 16 78 00 
E-mail: <a moz-do-not-send="true"
href="mailto:jesper.eskilson@xxxxxxxx"> jesper.eskilson@xxxxxxxx</a>
Website: <a moz-do-not-send="true"
href="http://www.iar.com"> www.iar.com 
</a> Twitter: <a moz-do-not-send="true"
href="http://www.twitter.com/iarsystems"> www.twitter.com/iarsystems</a>
 
</div>
</blockquote>
 
</body>
</html>

Ed Merks
Professional Support: https://www.macromodeling.com/

Report message to a moderator

Re: XML Schema and upper/lower bounds [message #900323 is a reply to message #900313]

Mon, 06 August 2012 13:36

Jesper Eskilson

Messages: 134
Registered: July 2009

Senior Member

<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
On 08/06/2012 03:01 PM, Ed Merks wrote:
<blockquote cite="mid:jvof7s$sj0$1@xxxxxxxxe.org" type="cite">
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
Jesper, 
 
XML Schema has <any> but its use is quite limited and EMF
treats it like a <sequence>.  EMF's deserializer doesn't
check that element are actually in the correct order anyway, and
will write them out in the model's feature order (except where
feature maps are used). 
</blockquote>
 
Does this mean that even if I specify <sequence>, EMF will not
complain when parsing an XML file where the child element order does
not match? There is a lot of XML code out in the wild, and I want to
avoid getting validation and/or parse errors on them. 
 
<div class="moz-signature">-- 

Jesper
Eskilson
Development
Engineer 
IAR Systems AB 
Box 23051, Strandbodgatan 1 
SE-750 23 Uppsala, SWEDEN 
Phone: +46 18 16 78 00 
E-mail: <a href="mailto:jesper.eskilson@xxxxxxxx">
jesper.eskilson@xxxxxxxx</a>
Website: <a href="http://www.iar.com">
www.iar.com 
</a>
Twitter: <a href="http://www.twitter.com/iarsystems">
www.twitter.com/iarsystems</a>


</div>
</body>
</html>

Report message to a moderator

Re: XML Schema and upper/lower bounds [message #900339 is a reply to message #900323]

Mon, 06 August 2012 14:41

Ed Merks

Messages: 33140
Registered: July 2009

Senior Member

<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Jesper, 
 
Yes, it will not complain. 
 
<div class="moz-cite-prefix">On 06/08/2012 3:36 PM, Jesper Eskilson
wrote: 
</div>
<blockquote cite="mid:jvoh86$oa$1@xxxxxxxxe.org" type="cite">
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
On 08/06/2012 03:01 PM, Ed Merks wrote:
<blockquote cite="mid:jvof7s$sj0$1@xxxxxxxxe.org" type="cite">
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
Jesper, 
 
XML Schema has <any> but its use is quite limited and EMF
treats it like a <sequence>.  EMF's deserializer doesn't
check that element are actually in the correct order anyway, and
will write them out in the model's feature order (except where
feature maps are used). 
</blockquote>
 
Does this mean that even if I specify <sequence>, EMF will
not complain when parsing an XML file where the child element
order does not match? There is a lot of XML code out in the wild,
and I want to avoid getting validation and/or parse errors on
them. 
 
<div class="moz-signature">-- 
 Jesper
Eskilson Development

Engineer 
IAR Systems AB 
Box 23051, Strandbodgatan 1 
SE-750 23 Uppsala, SWEDEN 
Phone: +46 18 16 78 00 
E-mail: <a moz-do-not-send="true"
href="mailto:jesper.eskilson@xxxxxxxx"> jesper.eskilson@xxxxxxxx</a>
Website: <a moz-do-not-send="true"
href="http://www.iar.com"> www.iar.com 
</a> Twitter: <a moz-do-not-send="true"
href="http://www.twitter.com/iarsystems"> www.twitter.com/iarsystems</a>
 
</div>
</blockquote>
 
</body>
</html>

Ed Merks
Professional Support: https://www.macromodeling.com/

Report message to a moderator

Previous Topic:	Automatically validate all referenced models
Next Topic:	EMF Code generation problem

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

]

Current Time: Thu Apr 25 09:57:25 GMT 2024

.:: Contact :: Home ::.

Breadcrumbs

Sign up to our Newsletter