Synthesizing Patches in Dockerfiles: Base Dataset

Docker containers are a standardized way of packaging applications and their execution environment in a reproducible manner. This dataset is an extension of an existing docker dataset with over 100,000 Dockerfiles in 15,000 projects (https://zenodo.org/record/1200869/). 

This dataset was used to extract patching patterns for Dockerfiles with the goal of improving quality in an automatic fashion.

The extension of the original dataset includes: 

• Static analysis results of every version of every Dockerfile
• Vulnerability data from analysing a limited amount of built Docker images
• A second database containing quality patches based on the static analysis results 

Files:

• msr18_extended
  A compressed, binary PostgreSQL database dump of the docker dataset extended with analysis results
• patch_database
  A compressed, binary PostgreSQL database dump of extracted patches
• patch_datbase.sql
  A PostgreSQL plain SQL database dump of extracted patches

Furhter information on the artifact used to extract and apply patches and instructions to import the database dumps are provided here: https://github.com/mandoway/dfp