this paper we describe a statically typed XML processing language called XDuce (o#cially pronounced "transduce"). XDuce is a functional language whose primitive data structures represent XML documents and whose types---called regular expression types---correspond to document schemas. The motivating principle behind its design is that a simple, clean, and powerful type system for XML processing can be based directly on the theory of regular tree automata

Abstract not found

We describe the summary graph lattice for dataflow analysis of programs that dynamically construct XML documents. Summary graphs have successfully been used to provide static guarantees in the JWIG language for programming interactive Web services. In particular, the JWIG compiler is able to check validity of dynamically generated XHTML documents and to type check dynamic form data. In this paper we present summary graphs and indicate their applicability for various scenarios. We also show that the expressive power of summary graphs is similar to that of the regular expression types from XDuce, but that the extra structure in summary graphs makes them more suitable for certain program analyses.

XML is successful as a machine processable data interchange format, but it is often too verbose for human use. For this reason, many XML languages permit an alternative more legible non-XML syntax. XSLT stylesheets are often used to convert from the XML syntax to the alternative syntax; however, such transformations are not reversible since no general tool exists to automatically parse the alternative syntax back into XML. We present XSugar, which makes it possible to manage dual syntax for XML languages. An XSugar specification is built around a context-free grammar that unifies the two syntaxes of a language. Given such a specification, the XSugar tool can translate from alternative syntax to XML and vice versa. Moreover, the tool statically checks that the transformations are reversible and that all XML documents generated from the alternative syntax are valid according to a given XML schema.

Abstract not found

... In this chapter, we shall see that the suspicion is easy dispelled. We look at techniques now used in practice for dealing with XML trees, and we note how they depart from old-fashioned uses. Since trees are objects that are very complicated to manipulate directly through pointer updates, declarative techniques are becoming increasingly important, especially when it comes to exploring, mining, and constructing tree-shaped data. In particular, we will contrast conventional concepts of database theory such as relational calculus with that of more procedural notations for trees. We explore why the essential problem of locating data in trees is intimately linked with tree automata and decidable logics, somewhat in parallel to the link between query algebras and first-order logic in relational database theory. So, we shall see why logic and automata create interesting new research opportunities.

Abstract not found

In this paper, we present a type system for typing Web applications in SMLserver, an e#cient multi-threaded Web server platform for Standard ML scriptlets. The type system guarantees that only conforming XHTML documents are sent to clients and that forms are used consistently and in a type-safe way. The type system is encoded in the type system of Standard ML using so-called phantom types.

Abstract not found

Abstract. Abstraction is the cornerstone of high-level programming; HTML forms are the principal medium of web interaction. However, most web programming environments do not support abstraction of form components, leading to a lack of compositionality. Using a semantics based on idioms, we show how to support compositional form construction and give a convenient syntax. 1