[tim777z]

DataFactor Assessment

Score: 37.0/100 (grade F)

Summary

Ambrose is a mature, historical Twitter open-source project (2013-2015) for visualizing MapReduce workflows, built on Maven with Java backend modules (pig, hive, cascading, scalding) plus a substantial vanilla JS front end. It has a reasonable README and a basic Travis CI config, moderate JUnit test coverage on core model/service classes, but no lockfiles, no lint/format enforcement, no dependency management tooling, no structured logging/observability, and no activity since 2015, indicating an abandoned but reasonably well-organized legacy codebase.

Recommendations Addressed

• Make install, build, and test work from a fresh clone [high]
• Ship features together with their tests, in small focused commits [medium]
• Add a docker compose that brings the whole app up [medium]
• Add CI-enforced automated test run with coverage threshold [high]
• Make lint and static analysis part of the pipeline [medium]
• Introduce structured logging and remove ad-hoc LOG.fatal/println patterns [medium]
• Commit a Maven wrapper and dependency lock evidence for reproducible builds [medium]
• Exercise pig/hive/cascading integration tests without live cluster dependencies [high]