Categories

In a world where data analytics skills have become an indispensable part of any education and modern workforce, are you still struggling with your relational queries?

HNRQ is about automatically constructing small, simple database instances (counterexamples) to illustrate why queries return wrong results, allowing users to trace query execution over these instances, and leading users to generalize from specific counterexamples to semantic descriptions of what cause wrong results.

HNRQ is still under construction and we will release the prototype for SQL queries in the future.

Key Features

Small Counterexample

Finding a small subinstance from the test database illustrating the error

Trace Query Execution

Tracing intermediate results of (correlated) subqueries in a way consistent with how the query was written

Demo

A video demonstrating our prototype for debugging SQL queries

The Team

HNRQ is being developed by a team of undergraduate and graduate students advised by Prof. Jun Yang, Prof. Sudeepa Roy, and Prof. Kristin Stephens-Martinez.

Graduate Students

Zhengjie Miao, Tiangang Chen, Yihao Hu

Undergraduate Students

Alexander Bendeck, Kevin Day, Jeffrey Luo

Publications

Zhengjie Miao, Tiangang Chen, Alexander Bendeck, Kevin Day, Sudeepa Roy, Jun Yang. “I-Rex: An Interactive Relational Query Explainer for SQL”. In Proceedings of the VLDB Endowment (PVLDB), Vol 13, Demonstration Track. Paper
Zhengjie Miao, Sudeepa Roy, Jun Yang. “Explaining Wrong Queries Using Small Examples.” In ACM SIGMOD International Conference on Management of Data (SIGMOD) 2019. Paper
Zhengjie Miao, Sudeepa Roy, Jun Yang. “RATest: Explaining Wrong Relational Queries Using Small Examples”. In ACM SIGMOD International Conference on Management of Data (SIGMOD) 2019, Demonstration track. Paper