A data curation platform for data collected on open-source software
For experiments researchers regularly require real-world code that fulfils specific requirements to be eligible to be included in an evaluation dataset. These requirements can may be fairly complex to come by or compute. Thus, a multitude of code projects have to be downloaded, inspected, and most of them will be discarded as not eligible.
Delphi is designed to help in these cases by providing a search interface to find open-source projects by various pre-computed criteria.
The easiest way to Delphi will be to use our live version once it is up and running on: https://delphi.cs.uni-paderborn.de
However, as Delphi is itself open-source software you can run your own service. This source code repository bundles all of Delphi's components and provides an uniform way to compile, run, and package the system.
You need a running instance of ElasticSearch on your local machine on port 9200.
After you started ElasticSearch, you can run delphi by following the below steps
- Cloning the projects. Needs to be done only for the first time.
sbt clone-all
- After that, just use run command
sbt run
into your terminal. SBT will compile and start all necessary services to run Delphi.
- Delphi components can be deleted in case all cloned components needs to be deleted using
sbt delete
Repository: https://github.com/delphi-hub/delphi-crawler
Repository: https://github.com/delphi-hub/delphi-webapi
Repository: https://github.com/delphi-hub/delphi-management
Repository: https://github.com/delphi-hub/delphi-cli
Repository: https://github.com/delphi-hub/delphi-webapp