SQUARES - A SQL Synthesizer Using Query Reverse Engineering

Given a set of input-output examples (tables), SQUARES returns the desired query in R and in SQL. SQUARES is built on top of Trinity. Therefore, the same packages are required.

• Prerequisite:
  • python 3.6+

How to use:

• using Jupyter Notebook:

  • Go to jupyter-notebook folder and launch the jupyter notebook and select demo.ipynb:

      jupyter notebook

• using Google Colab:

  • save the google colab doc to your account and run it.

• using python3 directly:

  python3 squaresEnumerator.py [tree|lines] [flags -h, ...] input.in

  • Flags:
  • -h : help
  • -on : computing symmetries online
  • -off : computing symmetries offline
  • -nr : only SQL query

  • -d : debug info

    Default: lines enumerator and without symmetry breaking

    — Input Files (.in): Some examples can be found in tests-examples folder

— Files required to integrate SQUARES in Trinity:

• tyrell/enumerator/lines.py
• tyrell/enumerator/lattices
• tyrell/enumerator/gen_lattices.py
• squares-enumerator.py
• setup.py (modified)

Instalation: You can either use (1) anaconda or (2) python and R packages.

(1)

• Install anaconda
• run

  chmod +x config.sh
  bash config.sh
  conda activate squares
  chmod +x config_squares.sh
  bash config_squares.sh

every time before using SQUARES run:

```
conda activate squares
```

or

(2)

— Python packages (install using pip or conda):

• sqlparse
• z3-solver
• sexpdata
• click
• rpy2

— R packages (install using conda or R console):

• dplyr
• dbplyr
• tidyr
• stringr

References

• Pedro Orvalho, Miguel Terra-Neves, Miguel Ventura, Ruben Martins and Vasco Manquinho. SQUARES : A SQL Synthesizer Using Query Reverse Engineering. VLDB 2020.
• Pedro Orvalho, Miguel Terra-Neves, Miguel Ventura, Ruben Martins and Vasco Manquinho. Encodings for Enumeration-Based Program Synthesis. CP 2019.
• Pedro Orvalho. SQUARES : A SQL Synthesizer Using Query Reverse Engineering. MSc Thesis. Instituto Superior Técnico - Universidade de Lisboa. 2019.
• Ruben Martins, Jia Chen, Yanju Chen, Yu Feng, Isil Dillig. Trinity: An Extensible Synthesis Framework for Data Science. VLDB 2019.
  • Yu Feng, Ruben Martins, Osbert Bastani, Isil Dillig. Program Synthesis using Conflict-Driven Learning. PLDI 2018.