JAVALI - JAVA Libraries and Interfaces

Everyday, libraries are updated, created or removed, and so their APIs. Such changes may impact stakeholders such as APIs providers and clients. In this context, it is important for APIs providers to know about strategic information like APIs change impact on clients and popularity over competitors. On the client side, it is interesting to compare APIs in order to select the most recommended in the market.To address these challenges, we propose JAVALI, a tool to analyze the popularity of Java APIs. Our database is composed of around 260K Java projects and 131M APIs. Our Web interface provides features to view and to export the results. We also report usage examples of JAVALI to solve real world API issues.
Aline Brito
M.Sc. Student in Computer Science. Software Developer.
Department of Computer Science. UFMG.
Lattes | LinkedIn | nbrito.aline@gmail.com
André Hora
Postdoc at the ASERG group lead by Marco Túlio Valente.
Department of Computer Science. UFMG.
Lattes | LinkedIn | andrehoraa@gmail.com
Marco Tulio Valente
Assistant Professor.
Department of Computer Science. UFMG.
Lattes | LinkedIn | mtov@dcc.ufmg.br
Address:
Av. Antonio Carlos, 6627 - ICEX
31270-010 – Belo Horizonte, MG. Brazil.
Phone: 55-31-3409-5860 – Fax: 55-31-3409-5858
We use Boa Language and Infrastructure (GitHub dataset). We analyze projects with at least one valid Java file and one import.

Our dataset contains:
  • 263.425 Java projects;
  • 16.386.193 Java files;
  • 131.147.733 imports;
  • 4.780.469 distinct imports.

apiwave: Developed by André Hora. It allows to analyze on the most used APIs in top 500 GitHub Java projects.