Type Error Diagnosis for Embedded DSLs by Two-Stage Specialized Type Rules

Alejandro Serrano, J. Hage

Research output: Chapter in Book/Report/Conference proceedingConference contribution

20 Citations (Scopus)

Abstract

In the functional programming world, it is common to embed a domain specific language (DSL) in a general purpose language. Unfortunately, the extra abstraction layer provided by the DSL is lost when a type error occurs, and internals leak to users of the DSL.This paper presents specialized type rules, a way to influence which part of the program is blamed and how the particular error message is worded. These type rules match part of the programming language abstract syntax tree (AST) and guide the type checker in order to provide custom diagnostics. Our goal is to enable DSL writers to keep their high-level abstractions throughout the whole development experience. Specialized type rules have already been considered in the literature: we enhance them by providing a mechanism to allow type rules to depend on partial type information.The technique presented in this paper can be readily applied to any type engine which uses constraints to perform its duty. We refine the workings of the type engine by including a second gathering pass in case an error is found. In that second pass, partial type information can be used to select a type rule. In particular, we have implemented our techniques in a type engine based on the OutsideIn(X) framework, which underlies the Haskell GHC compiler since version 7.
Original languageEnglish
Title of host publicationProgramming Languages and Systems. ESOP 2016
PublisherSpringer
Pages672-698
Number of pages27
ISBN (Electronic)9783662494981
ISBN (Print)9783662494974
DOIs
Publication statusPublished - 22 Mar 2016

Publication series

NameLecture Notes in Computer Science
Volume9632
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Dive into the research topics of 'Type Error Diagnosis for Embedded DSLs by Two-Stage Specialized Type Rules'. Together they form a unique fingerprint.

Cite this