- Downloads
tararchives - Sorts by Topic (roughly), Post and User
- No method exists for getting historical number of subscribers
- TODO: Use https://github.com/zapier/email-reply-parser to remove replies from raw text
- Uses askbot API for info on users and questions.
- Scrape HTML from the pages and gather
- Overall traffic (hits, bandwidth, number of visitors)
- Breakdown by url, country and operating system
- Statistics by month/year.
- Examines Github repo
- Determines type of change for each commit
- Counts the number of repositories contained in the distributions per commit
- Tracks which repos are present in each distro through time
- Statistics based on the repositories listed in the rosdistro
- Scores Github repositories based on stars, forks and subscribers.
- Gathers information on the issues/pull requests opened and closed.
- Counts the total number of commits
- TODO:
- Determine number of packages per repository/commit
- General git stats
- lines of code
- authors
- languages
- Uses resources from
scholarlyPython library - Counts citations per year
- Uses a combination of the publicly available mirror data and the DocBook format downloads.
- Tracks the number of editors, pages and revisions.
- Raw number of users is not publicly available and drawn from the official Metrics report.
- TODO:
- Special case for ROS2
- How are wiki edits spread across users?
- How big are wiki edits?
- Number of wiki tutorial pages under in any package
- Content per language/namespace
- Integrate users
The following data sources can only be crawled via use of an API key (or other special authentication)
- Crawled using Discourse API
- Retrieves info on
- users
- categories
- topics
- posts
- Analytics available for
- Uses Google Analytics API
- Overall number of unique pageviews, users and sessions gathered by month/year.
- Pageviews breakdown by url, country and operating system done by year.