arXiv subjects

Peter Thiemann

Publications and source records attributed to Peter Thiemann.

33 records · Page 2Linked to original sources

arXiv2017

Static Contract Simplification

Contracts and contract monitoring are a powerful mechanism for specifying properties and guaranteeing them at run time. However, run time monitoring of contracts imposes a significant overhead. The execution time is impacted by the insertion of contract checks as well as by the introduction of proxy objects that perform delayed contract checks on demand. Static contract simplification attacks this issue using program transformation. It applies compile-time transformations to programs with contracts to reduce the overall run time while preserving the original behavior. Our key technique is to statically propagate contracts through the program and to evaluate and merge contracts where possible. The goal is to obtain residual contracts that are collectively cheaper to check at run time. We distinguish different levels of preservation of behavior, which impose different limitations on the admissible transformations: Strong blame preservation, where the transformation is a behavioral equivalence, and weak blame preservation, where the transformed program is equivalent up to the particular violation reported. Our transformations never increase the overall number of contract checks.

cs.PL↗

arXiv2017

Transaction-based Sandboxing for JavaScript

Today's JavaScript applications are composed of scripts from different origins that are loaded at run time. As not all of these origins are equally trusted, the execution of these scripts should be isolated from one another. However, some scripts must access the application state and some may be allowed to change it, while preserving the confidentiality and integrity constraints of the application. This paper presents design and implementation of DecentJS, a language-embedded sandbox for full JavaScript. It enables scripts to run in a configurable degree of isolation with fine-grained access control. It provides a transactional scope in which effects are logged for review by the access control policy. After inspection of the log, effects can be committed to the application state or rolled back. The implementation relies on JavaScript proxies to guarantee full interposition for the full language and for all code, including dynamically loaded scripts and code injected via eval. Its only restriction is that scripts must be compliant with JavaScript's strict mode.

cs.PL↗

arXiv2017

Partial Derivatives for Context-Free Languages: From $μ$-Regular Expressions to Pushdown Automata

We extend Antimirov's partial derivatives from regular expressions to $μ$-regular expressions that describe context-free languages. We prove the correctness of partial derivatives as well as the finiteness of the set of iterated partial derivatives. The latter are used as pushdown symbols in our construction of a nondeterministic pushdown automaton, which generalizes Antimirov's NFA construction.

cs.FL↗

arXiv2016

Static Trace-Based Deadlock Analysis for Synchronous Mini-Go

We consider the problem of static deadlock detection for programs in the Go programming language which make use of synchronous channel communications. In our analysis, regular expressions extended with a fork operator capture the communication behavior of a program. Starting from a simple criterion that characterizes traces of deadlock-free programs, we develop automata-based methods to check for deadlock-freedom. The approach is implemented and evaluated with a series of examples.

cs.PL↗

arXiv2016

Derivatives for Enhanced Regular Expressions

Regular languages are closed under a wealth of formal language operators. Incorporating such operators in regular expressions leads to concise language specifications, but the transformation of such enhanced regular expressions to finite automata becomes more involved. We present an approach that enables the direct construction of finite automata from regular expressions enhanced with further operators that preserve regularity. Our construction is based on an extension of the theory of derivatives for regular expressions. To retain the standard results about derivatives, we develop a derivability criterion for the compatibility of the extra operators with derivatives. Some derivable operators do not preserve regularity. Derivatives provide a decision procedure for the word problem of regular expressions enhanced with such operators.

cs.FL↗

arXiv2015

Forkable Regular Expressions

We consider forkable regular expressions, which enrich regular expressions with a fork operator, to establish a formal basis for static and dynamic analysis of the communication behavior of concurrent programs. We define a novel compositional semantics for forkable expressions, establish their fundamental properties, and define derivatives for them as a basis for the generation of automata, for matching, and for language containment tests. Forkable expressions may give rise to non-regular languages, in general, but we identify sufficient conditions on expressions that guarantee finiteness of the automata construction via derivatives.

cs.FL↗

arXiv2015

Combining behavioural types with security analysis

Today's software systems are highly distributed and interconnected, and they increasingly rely on communication to achieve their goals; due to their societal importance, security and trustworthiness are crucial aspects for the correctness of these systems. Behavioural types, which extend data types by describing also the structured behaviour of programs, are a widely studied approach to the enforcement of correctness properties in communicating systems. This paper offers a unified overview of proposals based on behavioural types which are aimed at the analysis of security properties.

cs.PL↗

arXiv2015

Transparent Object Proxies for JavaScript

Proxies are the swiss army knives of object adaptation. They introduce a level of indirection to intercept select operations on a target object and divert them as method calls to a handler. Proxies have many uses like implementing access control, enforcing contracts, virtualizing resources. One important question in the design of a proxy API is whether a proxy object should inherit the identity of its target. Apparently proxies should have their own identity for security-related applications whereas other applications, in particular contract systems, require transparent proxies that compare equal to their target objects. We examine the issue with transparency in various use cases for proxies, discuss different approaches to obtain transparency, and propose two designs that require modest modifications in the JavaScript engine and cannot be bypassed by the programmer. We implement our designs in the SpiderMonkey JavaScript interpreter and bytecode compiler. Our evaluation shows that these modifications of have no statistically significant impact on the benchmark performance of the JavaScript engine. Furthermore, we demonstrate that contract systems based on wrappers require transparent proxies to avoid interference with program execution in realistic settings.

cs.PL↗

arXiv2015

TreatJS: Higher-Order Contracts for JavaScript

TreatJS is a language embedded, higher-order contract system for JavaScript which enforces contracts by run-time monitoring. Beyond providing the standard abstractions for building higher-order contracts (base, function, and object contracts), TreatJS's novel contributions are its guarantee of non-interfering contract execution, its systematic approach to blame assignment, its support for contracts in the style of union and intersection types, and its notion of a parameterized contract scope, which is the building block for composable run-time generated contracts that generalize dependent function contracts. TreatJS is implemented as a library so that all aspects of a contract can be specified using the full JavaScript language. The library relies on JavaScript proxies to guarantee full interposition for contracts. It further exploits JavaScript's reflective features to run contracts in a sandbox environment, which guarantees that the execution of contract code does not modify the application state. No source code transformation or change in the JavaScript run-time system is required. The impact of contracts on execution speed is evaluated using the Google Octane benchmark.

cs.PL↗

arXiv2015

A Falsification View of Success Typing

Dynamic languages are praised for their flexibility and expressiveness, but static analysis often yields many false positives and verification is cumbersome for lack of structure. Hence, unit testing is the prevalent incomplete method for validating programs in such languages. Falsification is an alternative approach that uncovers definite errors in programs. A falsifier computes a set of inputs that definitely crash a program. Success typing is a type-based approach to document programs in dynamic languages. We demonstrate that success typing is, in fact, an instance of falsification by mapping success (input) types into suitable logic formulae. Output types are represented by recursive types. We prove the correctness of our mapping (which establishes that success typing is falsification) and we report some experiences with a prototype implementation.

cs.PL↗

arXiv2014

Symbolic Solving of Extended Regular Expression Inequalities

This paper presents a new solution to the containment problem for extended regular expressions that extends basic regular expressions with intersection and complement operators and consider regular expressions on infinite alphabets based on potentially infinite character sets. Standard approaches deciding the containment do not take extended operators or character sets into account. The algorithm avoids the translation to an expression-equivalent automaton and provides a purely symbolic term rewriting systems for solving regular expressions inequalities. We give a new symbolic decision procedure for the containment problem based on Brzozowski's regular expression derivatives and Antimirov's rewriting approach to check containment. We generalize Brzozowski's syntactic derivative operator to two derivative operators that work with respect to (potentially infinite) representable character sets.

cs.FL↗

arXiv2013

On the Proxy Identity Crisis

A proxy, in general, is an object mediating access to an arbitrary target object. The proxy is then intended to be used in place of the target object. Ideally, a proxy is not distinguishable from other objects. Running a program with a proxy leads to the same outcome as running the program with the target object. Even though the approach provides a lot of power to the user, proxies come with a limitation. Because a proxy, wrapping a target object, is a new object and different from its target, the interposition changes the behaviour of some core components. For distinct proxies the double == and triple === equal operator returns false, even if the target object is the same. More precisely, the expected result depends on use case. To overcome this limitation we will discuss alternatives.

cs.PL↗

arXiv2013

Efficient Dynamic Access Analysis Using JavaScript Proxies

JSConTest introduced the notions of effect monitoring and dynamic effect inference for JavaScript. It enables the description of effects with path specifications resembling regular expressions. It is implemented by an offline source code transformation. To overcome the limitations of the JSConTest implementation, we redesigned and reimplemented effect monitoring by taking advantange of JavaScript proxies. Our new design avoids all drawbacks of the prior implementation. It guarantees full interposition; it is not restricted to a subset of JavaScript; it is self-maintaining; and its scalability to large programs is significantly better than with JSConTest. The improved scalability has two sources. First, the reimplementation is significantly faster than the original, transformation-based implementation. Second, the reimplementation relies on the fly-weight pattern and on trace reduction to conserve memory. Only the combination of these techniques enables monitoring and inference for large programs.

cs.PL↗

arXiv2013

Towards Tree Automata-based Success Types

Error detection facilities for dynamic languages are often based on unit testing. Thus, the advantage of rapid prototyping and flexibility must be weighed against cumbersome and time consuming test suite development. Lindahl and Sagonas' success typings provide a means of static must-fail detection in Erlang. Due to the constraint-based nature of the approach, some errors involving nested tuples and recursion cannot be detected. We propose an approach that uses an extension of model checking for pattern-matching recursion schemes with context-aware ranked tree automata to provide improved success typings for a constructor-based first-order prototype language.

cs.PL↗

arXiv2013

Type-based Dependency Analysis for JavaScript

Dependency analysis is a program analysis that determines potential data flow between program points. While it is not a security analysis per se, it is a viable basis for investigating data integrity, for ensuring confidentiality, and for guaranteeing sanitization. A noninterference property can be stated and proved for the dependency analysis. We have designed and implemented a dependency analysis for JavaScript. We formalize this analysis as an abstraction of a tainting semantics. We prove the correctness of the tainting semantics, the soundness of the abstraction, a noninterference property, and the termination of the analysis.

cs.PL↗