A partial type inference technique should come with a simple and precise specification, so that users predict its behavior and understand the error messages it produces. Local type inference techniques attain this simplicity by inferring missing type information only from the types of adjacent syntax nodes, without using global mechanisms such as unification variables. The paper reports on our experience with programming in a full-featured programming language including higher-order polymorphism, subtyping, parametric datatypes, and local type inference. On the positive side, our experiments on several nontrivial examples confirm previous hopes for the practicality of the type inference method. On the negative side, some proposed extensions mitigating known expressiveness problems turn out to be unsatisfactory on close examination. 1 Introduction It is widely believed that a polymorphic programming language should provide some form of type inference, to avoid discouraging programming ...

How canweintegrate interaction into a purely declarative language? This tutorial describes a solution to this problem based on a monad. The solution has been implemented in the functional language Haskell and the declarative language Escher. Comparisons are given to other approaches to interaction based on synchronous streams, continuations, linear logic, and side effects.

Despite the long list of publications on parser combinators, there does not yet exist a monadic parser combinator library that is applicable in real world situations. In particular naive implementations of parser combinators are likely to suffer from space leaks and are often unable to report precise error messages in case of parse errors. The Parsec parser combinator library described in this paper, utilizes a novel implementation technique for space and time efficient parser combinators that in case of a parse error, report both the position of the error as well as all grammar productions that would have been legal at that point in the input.

Packrat parsing is a novel technique for implementing parsers in a lazy functional programming language. A packrat parser provides the power and flexibility of top-down parsing with backtracking and unlimited lookahead, but nevertheless guarantees linear parse time. Any language defined by an LL(k) or LR(k) grammar can be recognized by a packrat parser, in addition to many languages that conventional linear-time algorithms do not support. This additional power simplifies the handling of common syntactic idioms such as the widespread but troublesome longest-match rule, enables the use of sophisticated disambiguation strategies such as syntactic and semantic predicates, provides better grammar composition properties, and allows lexical analysis to be integrated seamlessly into parsing. Yet despite its power, packrat parsing shares the same simplicity and elegance as recursive descent parsing; in fact converting a backtracking recursive descent parser into a linear-time packrat parser often involves only a fairly straightforward structural change. This paper describes packrat parsing informally with emphasis on its use in practical applications, and explores its advantages and disadvantages with respect to the more conventional alternatives.

this paper we will assume the availablity of a set of parsing combinators, that enables us to coinstruct such a mapping almost without e#ort.

Using an embedded, interpreted language to control a complicated application can have significant software-engineering benefits. But existing interpreters are designed for embedding into C code. To embed an interpreter into a different language requires an API suited to that language. This paper presents Lua-ML, a new API that is suited to languages that provide higher-order functions and types. The API exploits higher-order functions and types to reduce the amount of glue code needed to use an embedded interpreter. Where embedding in C requires a special-purpose “glue function ” for every function to be embedded, embedding in Lua-ML requires only a description of each function’s type. Lua-ML also makes it easy to define a Lua function whose behavior depends on the number and types of its arguments.

We discuss tools that aid in the development, the assessment and the reengineering of language descriptions. The assessment tools give an indication as to what is wrong with an existing language description, and give hints towards correction. From a correct and complete language description, it is possible to generate a parser, a manual, and on-line documentation. The parser is geared towards reengineering purposes, but is also used to parse the examples that are contained in the documentation. The reengineered language description is a basic ingredient for a reengineering factory that can manipulate this language. We demonstrate our approach with a proprietary language for real time embedded software systems that is used in telecommunications industry. The described tool support can also be used to develop a language standard without syntax errors in the language description and its code examples. Categories and Subject Description: D.2.6 [Software Engineering ]: Programming Environm...

A permutation phrase is a sequence of elements (possibly of di#erent types) in which each element occurs exactly once and the order is irrelevant. Some of the permutable elements may be optional. We show a way to extend a parser combinator library with support for parsing such free-order constructs. A user of the library can easily write parsers for permutation phrases and does not need to care about checking and reordering the recognised elements. Possible applications include the generation of parsers for attributes of XML tags and Haskell's record syntax.

Packrat parsing is a novel and practical method for implementing linear-time parsers for grammars defined in Top-Down Parsing Language (TDPL). While TDPL was originally created as a formal model for top-down parsers with backtracking capability, this thesis extends TDPL into a powerful general-purpose notation for describing language syntax, providing a compelling alternative to traditional context-free grammars (CFGs). Common syntactic idioms that cannot be represented concisely in a CFG are easily expressed in TDPL, such as longest-match disambiguation and "syntactic predicates," making it possible to describe the complete lexical and grammatical syntax of a practical programming language in a single TDPL grammar. Packrat parsing is an adaptation of a 30-year-old tabular parsing algorithm that was never put into practice until now. A packrat parser can recognize any string defined by a TDPL grammar in linear time, providing the power and flexibility of a backtracking recursive descent parser without the attendant risk of exponential parse time. A packrat parser can recognize any LL($k$) or LR($k$) language, as well as many languages requiring unlimited lookahead that cannot be parsed by shift/reduce parsers. Packrat parsing also provides better composition properties than LL/LR parsing, making it more suitable for dynamic or extensible languages. The primary disadvantage of packrat parsing is its storage cost, which is a constant multiple of the total input size rather than being proportional to the nesting depth of the syntactic constructs appearing in the input. Monadic combinators and lazy evaluation enable elegant and direct implementations of packrat parsers in recent functional programming languages such as Haskell. Three different packrat parsers for the Java language are presented here, demonstrating the construction of packrat parsers in Haskell using primitive pattern matching, using monadic combinators, and by automatic generation from a declarative parser specification. The prototype packrat parser generator developed for the third case itself uses a packrat parser to read its parser specifications, and supports full TDPL notation extended with ``semantic predicates,'' allowing parsing decisions to depend on the semantic values of other syntactic entities. Experimental results show that all of these packrat parsers run reliably in linear time, efficiently support ``scannerless'' parsing with integrated lexical analysis, and provide the user-friendly error-handling facilities necessary in practical applications.

We develop, in a stepwise fashion, a set of parser combinators for constructing deterministic, error-correcting parsers. The only restriction on the grammar is that it is not left recursive. Extensive use is made of lazy evaluation, and the parsers constructed "analyze themselves". Our new combinators may be used for the construction of large parsers to be used in compilers in practical use. Categories and Subject Descriptors D.1.1 [Programming techniques]: Applicative (functional) programming; D.3.3 [Programming languages]: Language Constructs and features; D.3.4 [Programming languages]: Processors---parsing,translator writing systems and compiler generators; D.3.4 [Programming languages]: Language Classification---Applicative (functional) languages General Terms parser combinators, error correction, deterministic, partial evaluation, parser generators, program analysis, lazy evaluation, polymorphism 1. INTRODUCTION There exist many di#erent implementations of the basic parse...