SHF: PROJECT DARWIN_ Towards Principled Language Evolution

SHF：达尔文计划_迈向有原则的语言进化

• 批准号: 1544542
• 负责人: Jan Vitek
• 金额: $ 109.97万
• 依托单位: Northeastern University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-07-01 至 2022-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1544542&HistoricalAwards=false
• 关键词: SHF; PROJECT; DARWIN_; Towards; Principled

R is a large language ecosystem consisting of statistical and graphical capabilities, a programming language, and all the other artifacts present in an open source programming environment. R is used by millions of scientific/engineering programmers in academia and is used heavily in business and industry. R is hitting limits and needs major renovations if it is to keep up with, and advance, the pace of science. This proposal proposes to modernize several aspects of the R language and to integrate new infrastructure elements. The modernization effort will focus on the code base of the language implementation, the compiler and the memory subsystem. The new features will center around the addition of type annotations to improve correctness and performance of R programs as well as to assist with distributed execution. The PI is well-connected to the R core developers, and there is a clear path to releasing versions into production. The enhancements will help R support the pace of science, as problems become more computationally complex and use rapidly growing data sets. Scientific computations programmed in R will run faster, be capable of processing much larger, distributed data sets, and the implementation will be easier to maintain by modernizing the language. The experience of reimplementing a live, full-scale language like R will be interesting to the programming language research community, as well as providing a more sustainable infrastructure for science/engineering programming.

R 是一个大型语言生态系统，由统计和图形功能、编程语言以及开源编程环境中存在的所有其他工件组成。 R 被学术界数以百万计的科学/工程程序员使用，并且在商业和工业领域也被大量使用。 R 已经达到极限，如果要跟上并推进科学的步伐，就需要进行重大革新。该提案提议对 R 语言的多个方面进行现代化改造并集成新的基础设施元素。现代化工作将集中在语言实现的代码库、编译器和内存子系统上。新功能将围绕添加类型注释来提高 R 程序的正确性和性能以及协助分布式执行。 PI 与 R 核心开发人员有着良好的联系，并且有一条明确的路径将版本发布到生产环境中。随着问题的计算变得更加复杂并使用快速增长的数据集，这些增强功能将帮助 R 支持科学的步伐。 用 R 编写的科学计算将运行得更快，能够处理更大的分布式数据集，并且通过使语言现代化，实现将更容易维护。 重新实现像 R 这样的实时、全面的语言的经验将引起编程语言研究社区的兴趣，并为科学/工程编程提供更可持续的基础设施。

项目成果

CodeDJ: Reproducible Queries over Large-Scale Software Repositories

CodeDJ：大规模软件存储库的可重复查询

• 发表时间: 2021-07
• 期刊: European Conference on Object-Oriented Programming (ECOOP
• 作者: Petr Maj; Konrad Siek
• 通讯作者: Konrad Siek

Designing types for R, empirically

根据经验设计 R 类型

• DOI: 10.1145/3428249
• 发表时间: 2020-11
• 期刊: Proceedings of the ACM on Programming Languages
• 作者: Turcotte, Alexi;Goel, Aviral;Křikava, Filip;Vitek, Jan
• 通讯作者: Vitek, Jan

Julia's Efficient Algorithm for Subtyping Unions and Covariant Tuples

Julia 用于子类型联合和协变元组的高效算法

• DOI: 10.4230/lipics.ecoop.2019.24
• 发表时间: 2019-01
• 期刊: European Conference on Object-Oriented Programming (ECOOP
• 作者: Chung, Benjamin;Zappa Nardelli, Francesco;Vitek, Jan
• 通讯作者: Vitek, Jan