2019 Fiscal Year Research-status Report
Tagless-final DSL embedding: how to keep extending the language and be sure it will still work
| Project/Area Number |
17K00091
|
|---|
| Research Institution | Tohoku University |
|---|
Principal Investigator |
Kiselyov Oleg 東北大学, 情報科学研究科, 助教 (50754602)
|
|---|
| Project Period (FY) |
2017-04-01 – 2021-03-31
|
| Keywords | DSL / domain-specifc language / tagless-final / denotational semantics / algebra |
|---|
| Outline of Annual Research Achievements |
As planned, in 2019 I, together with a master student, accomplished the second main project goal: using tagless-final to embed Logical-framework--like DSL for specifying typed calculi, for prototyping of functional languages and their type systems and, in particular, natural language semantic theories. We have implemented a DSL for describing derivations in AB grammars, Lambek grammars, and Kubota and Levin's Hybrid Type-Logical grammars. Not only we mechanically check that derivations are well-formed, but we also format them as LaTeX figures according to the common theoretical-linguistic conventions. That research has lead to an unexpected result in the field of Formal Grammar: an algebraic presentation of Lambek Grammars and the demonstration that under a natural restriction (satisfied in natural languages), Lambek grammars are strongly equivalent to context-free grammars, strengthening the existing solutions to this long-standing problem.
As part of disseminating research results and making them more accessible and applicable, I have delivered a three-lecture course at the International Summer School on Metaprogramming, on the tagless-final approach and its algebraic semantics.
I have investigated abstract interpretation, and denotational semantics of DSL programs with computational effects such as mutation and non-determinism, whose results are published in a journal paper (EPTCS 294).
I have used the tagless-final approach in writing code to assist in theoretical work in lambda-calculus, whose results are published as a journal paper (J. Functional Programming, v30).
|
| Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
研究実績の概要に述べた通り、2019年度に立てた目標はすべて達成されました。 そして、形式文法において重要な問題に対してもっと強く解決を提案しました。
|
| Strategy for Future Research Activity |
2019 was the last year of the project. But the research on the tagless-final approach is far from being exhausted. I'm now investigating the use of tagless-final for high-performance parallel and distributive programming and compiling of database queries. It is also interesting to further develop our result on the algebraic presentation of type-logical grammars in natural language semantics.
|
| Causes of Carryover |
当該事業の研究によって得られた結果を論文・会議で発表する予定です。論文はすでに投稿しましたが、査読が非常に遅くなってしまい、投稿の修正のため、事業期間の延長が必要となりました。また、2019年度内、参加したい適切な学会が少なかった (また、新型コロナの影響で取り消した会議もあった)ので、2020年度での参加を考えています。
|
