global-benchmark-database-tool 2.1.1

A tool for global benchmark management

gbd (Global Benchmark Database)

A tool for global benchmark management. Sometimes, it is hard to get a overview about all solutions for different problems and how fast/complex the algorithms are and how they can compete with your solution. What if all your benchmark files (.csv) would be managed by a database? This is what GBD does: Searching for files on your computer (path given), hashes (and resolves) them and add schemes for different attributes of benchmarks (clauses, variables, etc.).

Installation

Programming Language

Python 3

Python Packages (Requirements)

• tatsu (install with pip3)
• setuptools
• flask

Getting Started

• setup python3
• make sure Python was compiled with SSL support
• when coding, use package 'gbd_tool' for importing components

Usage

• The ConfigManager constructor needs an absolute path to your wished configuration directory - it does not have to exist
• Optional, you can give this constructor a path to a existing configuration file which you want to use. Note that you have to define all keys from the standard configuration file
• By default GBD works with a local sqlite3 database local.db which you have to define and give to the parameters of the constructor of the gbd_tool.GbdApi class. Important: You have to provide absolute paths to the api constructor, not relative paths!!!

  TODO: ConfigManager to provide relative path function

• You can set up your database by using methods in gbd_api.py
• If you want to provide this management system on a server, there will be a package for doing this

Documenation

GBD was initially presented at the Pragmatics of SAT (POS) Workshop 2018 hosted at FLoC 2018 in Oxford, UK. Thus, two resources can now be used as documentation of the system.