XML Schema Analysis View for TPR2

Jeri Lockhart
14 December 2005
revised 13 January 2006

Purpose
Screens
Features
Implementation
Other Useful Features that are Outside the Scope of the Analysis View
Resources

Purpose

• To answer questions that a user may have about his or her Schema in the form of a filtered graph.
• To help a Schema developer understand the structure of a Schema, especially a large Schema, by providing a graph showing the relationship between Schema components, such as where a certain global component is reused in a Schema.
• To help a Schema developer identify possible problems in his or her Schema by providing predefined queries, such as "Show all Complex Types that do not contain an Any element?", which could, for example,  indicate limited extensibility.

Screens

This is the "Show Usages" query from TPR1 in the new columnar format.  The predefined queries will be listed in the "Queries" column.  The column to the right of the Queries column will be used for customizing the query.  The third column contains the graph.





This illustrates the format of query list and query customizer column.  The content of the list is still to be determined.

Features

Provide a list of predefined queries.   At least three of these queries will be implemented for TPR2.

• Show "Red Flags" in a Schema - [see Resources] 

• Show relationships between Schema components

• Show relationships between files comprising a Schema
  

• Provide customization of the queries.

• Filter by name "starts with" or "contains", case-sensitive/insensitive

• Provide linking from the Schema Column View (Explore View) for appropriate queries.
  
• Provide linking to Schema Column View from the vertices in the graph (
  
• Provide linking to Schema Source from the vertices in the graph
• Activate property sheets for vertices in the graph

Implementation

The open-source prefuse graphing toolkit (http://prefuse.sourceforge.net/) was used for the TPR1.   At some point, a hierarchical directed graph layout, such as one that implements the Sugiyama, Tagawa, Toda 1981 algorithm, will be needed to graph the relationships between the files comprising a multifile Schema.  The prefuse toolkit does not have such a layout, so it will have to be implemented in prefuse. 

The visdev 2.0 module that is used by the SOA orchestration module is another possibility, but this hasn't been tested.

Other Useful Features that are Outside the Scope of the Analysis View

• Help developer to create "venetian blind" style schemas.
• Nest repeated elements in their own container.  Giving each repeated element a wrapper element worked best with tools which mapped XML to data structures and help avoid invalidating the UPA rules.
• Find violations of UPA constraint.  Optionally ignore UPA constraint violations.

Resources

These two documents discuss best practices for writing Schemas:

• Kohsuke -- http://www.xml.com/pub/a/2001/06/06/schemasimple.html
• MSDN -- http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnxml/html/xmlschemacomplex.asp?frame=true&hidetoc=true
• HP -- http://devresource.hp.com/drc/resources/xmlSchemaBestPractices.jsp
  

These documents from the W3C  User Experiences Workshop, June 2005  (http://www.w3.org/2005/03/xml-schema-user-cfp.html) discuss pitfalls to avoid in authoring Schemas:

• BT -- http://www.w3.org/2005/05/25-schema/BT.html
• Microsoft -- http://www.w3.org/2005/05/25-schema/microsoft.html

Sketches for TPR2

Find Global Elements that are substitutable

(Sketched with Graffle)