From iscp9142@nus.edu.sg Tue Jan 22 10:06:29 2002
Date: Tue, 22 Jan 2002 13:53:50 +0800
From: "Sun, Jing" <iscp9142@nus.edu.sg>
To: Andrew Martin <Andrew.Martin@comlab.ox.ac.uk>
Subject: Fw: Some reading for the holidays [CZT]

    [ The following text is in the "iso-8859-1" character set. ]
    [ Your display is set for the "US-ASCII" character set. Some ]
    [ characters may be displayed incorrectly. ]


----- Original Message -----
From: "Sun, Jing" <iscp9142@nus.edu.sg>
To: "Ian Toyn" <ian@cs.york.ac.uk>
Cc: <czt-project@comlab.ox.ac.uk>
Sent: Friday, January 18, 2002 6:01 PM
Subject: Re: Some reading for the holidays [CZT]


> Dear Dr. Ian Toyn,
>
> I've read the draft of "Comparing Z Annotated Syntaxes: Towards an XML
> Markup for Standard Z", and as a matter of fact I found some similarities
> between the requirement in the draft and our fully annotated ZML approach.
> I think this is due to that we both referred to the standard Z syntax
> definition.
> The difference is that we defined it in XML Schema instead of DTD. And I
> will show some examples using XML Schema later in this message. The
> following are some of my comments toward the requirement. I also cc the
> message to the CZT mailing list, hope to get more comments on the topic.
>
> First issue I want to address is about XML Schema. In the draft, DTD was
> chosen to define Z syntax.  I agree that DTD can fulfil the job fine, but
> compare to XML Schema it has certain limitations, see at
> http://www.cse.ucsc.edu/~douglas/proximl/node5.html . XML Schema was
> originally proposed by Microsoft, then became an official W3C
recommendation
> in May 2001. Many believe that very soon XML Schemas will be used in most
> applications as a replacement and successor for DTDs due to its
simplicity,
> extensibility, datatype, namespace supports and so on, see at
> http://www.cse.ucsc.edu/~douglas/proximl/node6.html . According to Z
> standard syntax, an annotated Z markup can be specified via XML Schema
more
> easily. The attached jpg files are the screen captures of part of the Z
> schema, declaration, predicate and expression definitions in XML Schema
> represented in a diagramatic view. Sometimes I find that visulizing in
this
> way is more readable than reading those texture tag definitions. The files
> are:
>
> schema.jpg -> Z-schema definition
> declaration.jpg -> Z-declaration defintion
> predicate.jpg -> Z-predicate definition
> expression.jpg -> Z-expression defintion
> zml.jpg -> overview of ZML
>
> The above diagramatic views are generated from the XML Schema authoring
> tool. A
> complete version of ZML definition is also available at
> http://nt-appn.comp.nus.edu.sg/fm/zml/zml.xsd . It uses the standard W3C
XML
> Schema format,  and includes the definition for Z, Object-Z and TCOZ
> notations. The Z syntax was referred to the appendix in the "Z user
manual"
> and the Object-Z syntax was referred to the appendix in "The Object-Z
> Specification Language" by Graeme.
>
> The following is part (given type and state schema only) of the example
> (BirthdayBook) in ZML:
>
> ===================================================================
> <?xml version="1.0" encoding="UTF-8"?>
> <ZML xmlns:xsi="http://www.w3.org/2000/10/XMLSchema-instance"
>
xsi:noNamespaceSchemaLocation="http://nt-appn.comp.nus.edu.sg/fm/zml/zml.xsd
> ">
>  <comment>This is a simplified version of the BirthdayBook example in Z
> represented in Z Markup Language.</comment>
>  <basicTypeDef>
>   <name>DATE</name>
>   <name>NAME</name>
>  </basicTypeDef>
>  <schemaDef>
>   <name>BirthdayBook</name>
>   <declaration>
>    <variable>known</variable>
>    <dataType>
>     <unarySym>pset</unarySym>
>     <type>NAME</type>
>    </dataType>
>   </declaration>
>   <declaration>
>    <variable>birthday</variable>
>    <dataType>
>     <dataType>
>      <type>NAME</type>
>     </dataType>
>     <binarySym>tfun</binarySym>
>     <type>DATE</type>
>    </dataType>
>   </declaration>
>   <predicate>
>    <expression>
>     <varName>known</varName>
>    </expression>
>    <relationSym>=</relationSym>
>    <expression>
>     <prefixExpr>dom</prefixExpr>
>     <expression>
>      <varName>birthday</varName>
>     </expression>
>    </expression>
>   </predicate>
>  </schemaDef>
> ......
> </ZML>
> ============================================
>
> As I mentioned earlier since we both refer to the standard Z syntax, the
> specs are quite similar to that in the requirement draft. One obvious
> difference might be the use of  "WORD" in the draft and the "name" in ZML.
>
> I also attached the complete example file of the BirthdayBook in XML
> (birthdaybook.xml). This XML file can be validated by XML parsers against
> its XML Schema syntax definition. For the visualized BirthdayBook example
in
> Latex please refer to the "birthdaybook.ps" (attached).
>
> Another issue is that I agree that the validation function of XML parser
can
> be used as a facility for checking syntactical errors in Z specs. And for
> this purpose XML Schema also has its advantage. Let's look at a simple
> example. If we agreed that the Z schema name can only be starting by an
> alphabetical letter following by the others, then we could define a
nameType
> in XML Schema which extense the string type and with restrictions as
> follows.
>
>  <xsd:simpleType name="nameType">
>   <xsd:restriction base="xsd:string">
>    <xsd:pattern value="[A-Z a-z].*"/>
>   </xsd:restriction>
>  </xsd:simpleType>
>
> Then defining the syntax for schema names as "<xsd:element name="name"
> type="nameType"/>". Any violation of mis-type value of the name will be
> checked by XML validator parser as "value does not match facet
pattern="[A-Z
> a-Z}.* in element name". This type of validating process will certainly
> beyond the capablity of DTD.
>
> Finally, as Ian mentioned in the draft is the encoding of relation between
> declarations and reference to them. We treat this as using the annotation
of
> "type" and "name" (see the datatype in the declaration.jpg). The name of a
> given type, schema type will become future type refernce to other type
> definitions, which annotated as "type" in a data declaration section. It
> might be one way of tackle the to the problem.
>
> In the above message we presented a similar annotation of Z in XML using
XML
> Schema, which we though would be more flexible and extensiable. Any
comment
> is welcome. Thanks.
>
>
> Best regards.
>
> Jing Sun
>
>
> ----- Original Message -----
> From: "Ian Toyn" <ian@cs.york.ac.uk>
> To: "Sun, Jing" <iscp9142@nus.edu.sg>
> Sent: Tuesday, January 15, 2002 6:29 PM
> Subject: Re: Some reading for the holidays [CZT]
>
>
> > Dear Jing,
> >
> > > We have been using XML to represent Z/Object-Z languages, attached as
> > > follows.
> > >
> > > 1) light annotated version (upto predicate level): objectZschema.xml
> > > 2) fully annotated version (downto expression level): zml.xsd
> >
> > I looked at zml.xsd and found your mark-up of Z expressions,
> > but I didn't see any mark-up for annotations on those expressions.
> >
> > > As our project is still in the initial stage, we would be grateful to
> have
> > > more comments.
> > > And I'd also like to help with any questions on XML/DTD too.
> >
> > Thanks for your offer to help.
> > I've attached my latest draft to this message.
> > That draft sets out my requirements for an interchange mark-up for Z.
> > I suspect the requirements for your XML mark-up are different from mine.
> > What do you think?
> >
> > Regards,
> > Ian Toyn
>
>
>
>

    [ Part 2, Image/JPEG  34KB. ]
    [ Unable to print this part. ]


    [ Part 3, Image/JPEG  54KB. ]
    [ Unable to print this part. ]


    [ Part 4, Image/JPEG  60KB. ]
    [ Unable to print this part. ]


    [ Part 5, Image/JPEG  83KB. ]
    [ Unable to print this part. ]


    [ Part 6, Image/JPEG  30KB. ]
    [ Unable to print this part. ]


    [ Part 7, Text/XML (Name: "birthdaybook.xml")  198 lines. ]
    [ Unable to print this part. ]


    [ Part 8, Application/POSTSCRIPT  108KB. ]
    [ Unable to print this part. ]