Case Study: FastAPI

Repository: fastapi/fastapi Stats: 1,118 files, 4,554 functions Drift Score: 0.690 (HIGH) | Time: 2.3s

Scope note

These numbers reflect a full-clone analysis (all branches and test files included). The Comparisons page uses a src/-only shallow-clone scope, which reports 664 files, score 0.624, and 13.1 s for the same repository.

Key Findings

499 Near-Duplicate Test Functions (MDS 0.85)

FastAPI's test suite contains hundreds of structurally identical test functions that differ only in the model name: test_read_items(), test_read_users(), test_read_events() share identical assertion structures.

This is the classic "mutant duplicate" pattern — code that was likely copied and minimally adapted rather than abstracted into a parameterized test fixture.

4 Error-Handling Patterns in Route Modules (PFS)

The Pattern Fragmentation Signal found 4 distinct approaches to error handling across route modules — try/except with re-raise, bare except with logging, HTTPException wrapping, and direct return.

Interpretation

Even well-maintained frameworks accumulate structural debt at scale. These findings don't indicate bugs — they indicate inconsistency that makes the codebase harder to modify safely.

Recommendation: Parameterized test fixtures (@pytest.mark.parametrize) could reduce the test duplication. A shared error-handling utility could unify the route patterns.