Overview

Overview

Most of the content in this Wiki page was extracted from the Graduate Work pager of Flávio Juvenal (@fjsj), who was Groundhog's initial developer.

This document can certainly help you understand Groundhog, what things it does and how it works.

In Groundhog, the integration flow is linear - that is - for every project to be analyzed, first the Search module is executed, then Crawler, then CodeHistory and ultimately, Parser. In some case it's also necessary to execute another module: the Extractor. Below is a brief, overview-ish explanation of the responsibilities of each module:

• Search – browses the forges' web pages or official APIs and fetches the information about the projects for further download by the Crawler module.

• Crawler – uses the project's information obtained by the Search module and performs the download of the source code of the projects through the forges' pages or its VCSs.

• CodeHistory – obtains the most recent version of the source code of a project within a date of interest to have its metrics extracted by the Parser module.

• Parser – analyzes each file in the source code of a project and builds an AST and browses through it to extract metrics.

• Extractor – in some cases the source code files are compressed and this module performs the extraction of these files in order to make the source code analysis possible.

Other resources

• http://commons.apache.org/proper/commons-bcel/
• http://flossmole.org/
• http://sourcerer.ics.uci.edu/index.html