Some hints for managing the database in FLOSSMetrics

The structure of the FLOSSMetrics database has been designed to cover the various needs of the public, such as researchers or developers. For this reason the database is divided in several levels, according to the various studies that can be performed.

As it can be observed from the previous figure, the database will be divided in three levels. The first (lower) level will contain data extracted by the tools integrated into the retrieval system; the second (medium) level will unify the data of the previous level; the third (high) level will contain analyses and statistics:

• First level databases. The tools level is the lower level of the database. As its name indicates, it will store data obtained directly from executing mining or extraction tools on data sources from projects. The retrieval system is in charge of performing this work and managing the insertion of these data into the database.

• Second level databases. The main intention of this level is to establish relationships between the data stored in the tools level. One of the main difficulties for researchers is to obtain information about the relationships and the cooperation among projects. There are different levels of relationships: from developers that work in several projects, or write giving ideas to the mailing list or fixing bugs; to duplicated or reused code. Maybe, the most important characteristic of this level, is that it tries to find these correlations, and provide them to researchers and developers.

• Third level databases. The analyses level tries to minimized the time used by the researchers during their investigations. It will store a large set of distinct analyses, including common results and those who need long-time of process. If these analyses are available for the researches, they will not waste the time recalculating them. In addition, some of these analyses will be used for classifying projects according to the deliverable D4.1.

See below details about these kinds of databases.

• SCM: (CVSAnaly2)
• MLS: (MailinglistStats)
• TRK: (Bicho)

• Aggregated databases for SCM:
• Aggregated databases for MLS:
• Aggregated databases for TRK:

• Variables database for SCM:
• Variables database for MLS:
• Variables database for TRK:

All of this assumes you have direct access to the MySQL database.

You can use the “show databases” MySQL command. This will produce a list with all the databases available for users, which are those already “ready” for researchers.