Sun 23 Jun 2019 10:00 - 10:30 at 212B - Session I Chair(s): Neville Grech

In the past 15 years, Datalog has found increasing use in declarative program analyses. While Datalog itself is small and accessible, the tooling required to extract the analysis facts from a program remains complex and requires specialized knowledge in compilers or intermediate representations.In our MetaDL framework (SOAP’19) we have extended Datalog to allow source-level analysis of Datalog within Datalog. To that end, we have introduced syntactic patterns and a relational representation of the program’s AST. Our approach has the potential to facilitate the quick prototyping and development of custom program analyses by general software practitioners. The only requirements are a good command of the object language and first order logic, which many developers already possess.In this talk we describe our MetaDL system, give examples of its usage, and sketch how our approach can be adapted to other programming languages whose frontend is implemented on top of a (reference) attribute grammar system.

Sun 23 Jun
Times are displayed in time zone: (GMT-07:00) Tijuana, Baja California change

09:30 - 11:00: DPA - Session I at 212B
Chair(s): Neville GrechUniversity of Athens
dpa-2019-papers09:30 - 10:00
Sandeep DasguptaUniversity of Illinois at Urbana-Champaign, USA
dpa-2019-papers10:00 - 10:30
Christoph ReichenbachLund University
dpa-2019-papers10:30 - 11:00
Michael ArntzeniusUniversity of Birmingham, UK