Taught by Priscilla Walmsley.

A lot of existing XSLT stylesheets are not optimal. They may be performing poorly, hard to maintain, or not robust enough to handle all use cases. There are many reasons for this: they might have been developed under time constraints, built up gradually over time into something overly complex, written by less experienced developers, or generated from a tool.

In this class, we will take a look at what constitutes a well-designed XSLT stylesheet, and how to apply good design techniques to existing XSLT stylesheets to make them more robust and easier to maintain. We will talk about the pitfalls that developers new to XSLT encounter that could cause problems later on, and how to spot them. We will look at tools to identify potential stylesheet weaknesses and automate some of the cleanup.

As a group, we will examine specific examples of XSLT stylesheets and discuss how to improve them. Students are encouraged to submit XSLT examples that can be used for this discussion, in order to get constructive feedback.

Taught by Adam Retter.

This class will provide an overview of the capabilities and use cases of XML databases, examining some of the database products that support XML and how they are being used. It will then cover the role of XQuery among other XML technologies in the querying of XML databases.

As a group, the class will build a simple search application using XQuery and an XML database (eXist). This will provide an opportunity for attendees to learn the syntax and capabilities of XQuery, as well as see it in action. Major features of the XQuery language such as FLWOR expressions, XML constructors, and user-defined functions will be explained.

Taught by Florent Georges.

XSLT 3.0 and XQuery 3.0 come with a lot of new, powerful features.

XPath 3.0 itself comes with one of the most useful of them: function items and higher-order functions; that is, the ability to manipulate functions and pass them around, calling them dynamically and define new anonymous inline functions. XQuery 3.0 has numerous additions on its own, among them grouping, windowing, try/catch, and private functions. For XSLT 3.0, the biggest addition is the support for streaming transformation. Florent will also introduce the packages, the evaluation of dynamically computed XPath expressions, the extension of template rules to atomic values, the try/catch mechanism and the new xsl:assert instruction.

Taught by Jirka Kosek.

XSLT 3.0 brings a lot of broad new features, such as streaming processing and high-order functions. But many existing transformations will not benefit from these features because they are operating on moderately sized documents and do not implement complex algorithms.

Instead, they will benefit more from a lot of small XSLT 3.0 features that improve productivity and readability. Building on the previous session, some of these incremental, practical improvements will be demonstrated and applied, including:

• producing more readable complex XPath expressions by using new operators (simple map operator (!), arrow operator (=>), string concatenation (||))
• writing more readable templates using text value templates
• using xsl:evaluate and shadow attributes to make stylesheets easily parameterized
• replacing recursive template calls and functions with xsl:iterate
• writing identity transformations without the identity template (xsl:mode)

During the class, attendees will also be refactoring provided simple XSLT 1.0/2.0 transformations into more readable code using XSLT 3.0 in order to get new features under the skin.