A maximum-flow approach to anomaly isolation in unification-based incremental type inference

Johnson, Gregory F.; Walz, Janet A.

doi:10.1145/512644.512649

Cited by 47 publications

(25 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Johnson and Walz describe a contextual type checker that incrementally normalizes type constraints while type checking a program in order to precisely identify the cause of type errors [12]. Aditya and Nikhil describe an incremental Hindley/Milner type system that only supports incremental rechecking of toplevel definitions [2].…”

Section: Related Workmentioning

confidence: 99%

A co-contextual formulation of type rules and its application to incremental type checking

Erdweg

Bračevac

Kuci

et al. 2015

Proceedings of the 2015 ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applicatio

View full text Add to dashboard Cite

Type rules associate types to expressions given a typing context. As the type checker traverses the expression tree topdown, it extends the typing context with additional context information that becomes available. This way, the typing context coordinates type checking in otherwise independent subexpressions, which inhibits parallelization and incrementalization of type checking.We propose a co-contextual formulation of type rules that only take an expression as input and produce a type and a set of context requirements. Co-contextual type checkers traverse an expression tree bottom-up and merge context requirements of independently checked subexpressions. We describe a method for systematically constructing a co-contextual formulation of type rules from a regular context-based formulation and we show how co-contextual type rules give rise to incremental type checking. Using our method, we derive incremental type checkers for PCF and for extensions that introduce records, parametric polymorphism, and subtyping. Our performance evaluation shows that co-contextual type checking has performance comparable to standard contextbased type checking, and incrementalization can improve performance significantly.

show abstract

Section: Related Workmentioning

confidence: 99%

A co-contextual formulation of type rules and its application to incremental type checking

Erdweg

Bračevac

Kuci

et al. 2015

Proceedings of the 2015 ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applicatio

View full text Add to dashboard Cite

show abstract

“…Efforts on improving type-error messages in ML-like languages can be traced to the early work of Wand [35] and of Johnson and Walz [18]. These two pieces of work represent two directions in improving error messages: the former traces everything that contributes to the error, whereas the latter attempts to infer the most likely cause.…”

Section: Related Workmentioning

confidence: 99%

“…But since the error location may be used anywhere during the unification procedure, any specific order fails in some circumstance. Some prior work [16,18] builds a type graph from a more limited constraint language and infers error locations based on heuristics mostly tailored for type inference. Though the "weighted options" heuristic in [18] uses successful type unifications to distinguish abnormal types from normal ones, information about satisfiable paths is leveraged with finer-granularity in our approach, to distinguish the constraints that caused errors.…”

Section: Related Workmentioning

confidence: 99%

Toward general diagnosis of static errors

Zhang

Myers

2014

Proceedings of the 41st ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages

View full text Add to dashboard Cite

We introduce a general way to locate programmer mistakes that are detected by static analyses such as type checking. The program analysis is expressed in a constraint language in which mistakes result in unsatisfiable constraints. Given an unsatisfiable system of constraints, both satisfiable and unsatisfiable constraints are analyzed, to identify the program expressions most likely to be the cause of unsatisfiability. The likelihood of different error explanations is evaluated under the assumption that the programmer's code is mostly correct, so the simplest explanations are chosen, following Bayesian principles. For analyses that rely on programmer-stated assumptions, the diagnosis also identifies assumptions likely to have been omitted. The new error diagnosis approach has been implemented for two very different program analyses: type inference in OCaml and information flow checking in Jif. The effectiveness of the approach is evaluated using previously collected programs containing errors. The results show that when compared to existing compilers and other tools, the general technique identifies the location of programmer errors significantly more accurately.

show abstract

“…Error diagnoses for ML-like languages Efforts on improving error messages for ML-like languages can be traced to the 80's [35,16]. Most of these efforts can be categorized into three directions.…”

Section: Related Workmentioning

confidence: 99%

“…The first direction, followed by [16,19,22,14,36,26] as well as most ML-like language compilers, attempts to infer the most likely cause. One approach is to alter the order of type unification [19,22,5].…”

Section: Related Workmentioning

confidence: 99%

Diagnosing type errors with class

et al. 2015

View full text Add to dashboard Cite

Type inference engines often give terrible error messages, and the more sophisticated the type system the worse the problem. We show that even with highly expressive type system implemented by the Glasgow Haskell Compiler (GHC)-including type classes, GADTs, and type families-it is possible to identify the most likely source of the type error, rather than the first source that the inference engine trips over. To determine which are the likely error sources, we apply a simple Bayesian model to a graph representation of the typing constraints; the satisfiability or unsatisfiability of paths within the graph provides evidence for or against possible explanations. While we build on prior work on error diagnosis for simpler type systems, inference in the richer type system of Haskell requires extending the graph with new nodes. The augmentation of the graph creates challenges both for Bayesian reasoning and for ensuring termination. Using a large corpus of Haskell programs, we show that this error localization technique is practical and significantly improves accuracy over the state of the art.

show abstract

A maximum-flow approach to anomaly isolation in unification-based incremental type inference

Cited by 47 publications

References 10 publications

A co-contextual formulation of type rules and its application to incremental type checking

A co-contextual formulation of type rules and its application to incremental type checking

Toward general diagnosis of static errors

Diagnosing type errors with class

Contact Info

Product

Resources

About