Codebase intelligence for AI and humans: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.
Python aicode-complexitycode-healthcode-intelligencecode-qualitydead-codedeveloper-toolsdocumentationgit-analyticsmcpopen-sourcepythonrefactoringstatic-analysistechnical-debt
fix(risk): rank tests_to_run by files reached, not alphabetically (#1862)
Raghav Chamadiyaa96e4ccf9f0f
diff --git a/packages/core/src/repowise/core/analysis/pr_blast.py b/packages/core/src/repowise/core/analysis/pr_blast.pyindex af91f67a7..eba05d40f 100644--- a/packages/core/src/repowise/core/analysis/pr_blast.py+++ b/packages/core/src/repowise/core/analysis/pr_blast.py@@ -17,7 +17,8 @@ from __future__ import annotationsimport jsonimport mathimport os-from collections import defaultdict+from collections import Counter, defaultdict+from collections.abc import Iterable, Mappingfrom typing import Anyfrom sqlalchemy import select, text@@ -27,6 +28,18 @@ from repowise.core.ingestion.models import NON_DEPENDENCY_EDGE_TYPESfrom repowise.core.persistence.models import GitMetadata, GraphNode+def rank_tests_by_reach(by_file: Mapping[str, Iterable[str]]) -> list[str]:+ """Test ids ordered by how many changed files each one reaches.++ Callers cap this list and tell the agent to run the head of it first, so a+ flat alphabetical sort hands over the head of the alphabet rather than the+ tests that cover most of the change. Ties keep alphabetical order, which+ leaves a single-file change ordered exactly as before.+ """+ reach = Counter(test_id for tests in by_file.values() for test_id in tests)+ return sorted(reach, key=lambda t: (-reach[t], t))++class PRBlastRadiusAnalyzer:"""Compute blast radius for a proposed PR given its changed files."""@@ -409,7 +422,7 @@ class PRBlastRadiusAnalyzer:return {"map_present": True,"basis": "measured",- "tests_to_run": sorted(all_ids),+ "tests_to_run": rank_tests_by_reach(by_file),"by_file": by_file,}@@ -431,11 +444,10 @@ class PRBlastRadiusAnalyzer:if not reaching:return emptyby_file = {path: sorted(tests) for path, tests in reaching.items() if tests}- all_tests = sorted({t for tests in by_file.values() for t in tests})return {"map_present": empty["map_present"],"basis": "inferred",- "tests_to_run": all_tests,+ "tests_to_run": rank_tests_by_reach(by_file),"by_file": by_file,}diff --git a/packages/server/src/repowise/server/mcp_server/tool_change_risk.py b/packages/server/src/repowise/server/mcp_server/tool_change_risk.pyindex 76b42c582..692676cfd 100644--- a/packages/server/src/repowise/server/mcp_server/tool_change_risk.py+++ b/packages/server/src/repowise/server/mcp_server/tool_change_risk.py@@ -19,6 +19,7 @@ from repowise.core.analysis.change_risk import (normalize_extensions,score_live_change,)+from repowise.core.analysis.pr_blast import rank_tests_by_reachfrom repowise.core.registry import mcp_tool_registry as mcpfrom repowise.server.mcp_server._helpers import (_get_repo,@@ -447,7 +448,7 @@ async def _inferred_impacted(reaching = await tests_reaching(session, repo_id, changed_files)except Exception:reaching = {}- tests = sorted({t for group in reaching.values() for t in group})+ tests = rank_tests_by_reach(reaching)if not tests:return _empty_impacted("no_map",@@ -510,14 +511,16 @@ async def _impacted_tests_block(report = await detect_missing_tests(session, repo_id, changed)if report.map_empty:return await _inferred_impacted(session, repo_id, sorted(changed))- all_ids: set[str] = set()+ by_file: dict[str, list[str]] = {}for source_file, lines in changed.items():- for row in await tests_covering(session, repo_id, source_file, lines=lines):- all_ids.add(row["test_id"])+ rows = await tests_covering(session, repo_id, source_file, lines=lines)+ ids = sorted({row["test_id"] for row in rows})+ if ids:+ by_file[source_file] = idsexcept LookupError:return _empty_impacted("no_index", "No indexed repository; run `repowise init`.")- tests = sorted(all_ids)+ tests = rank_tests_by_reach(by_file)total = len(tests)return {"status": "map_present",diff --git a/tests/unit/server/mcp/test_risk.py b/tests/unit/server/mcp/test_risk.pyindex 23dc953bd..5d0653d8c 100644--- a/tests/unit/server/mcp/test_risk.py+++ b/tests/unit/server/mcp/test_risk.py@@ -155,6 +155,67 @@ async def test_get_risk_pr_directive_surfaces_coverage_backed_tests_to_run(setupassert "coverage-backed test(s) guard the change" in directive["summary"]+async def test_get_risk_tests_to_run_ranks_by_files_reached(setup_mcp, session):+ """The test covering both changed files leads, despite sorting last."""+ from repowise.core.analysis.health.coverage import TestCoverage+ from repowise.core.persistence.crud import save_test_coverage+ from repowise.server.mcp_server import get_risk++ def _cov(test_id: str, path: str) -> TestCoverage:+ return TestCoverage(+ test_id=test_id,+ file_path=path,+ covered_lines=[1, 2],+ source_format="coverage.py",+ test_file=test_id.split("::")[0],+ )++ await save_test_coverage(+ session,+ "repo1",+ [+ # Sorts last alphabetically, reaches both changed files.+ _cov("tests/test_zeta.py::test_both", "src/auth/service.py"),+ _cov("tests/test_zeta.py::test_both", "src/auth/token.py"),+ # Sorts first alphabetically, reaches one.+ _cov("tests/test_alpha.py::test_one", "src/auth/service.py"),+ ],+ source_format="coverage.py",+ )+ await session.flush()++ result = await get_risk(+ ["src/auth/service.py"],+ changed_files=["src/auth/service.py", "src/auth/token.py"],+ )++ assert result["directive"]["tests_to_run"] == [+ "tests/test_zeta.py::test_both",+ "tests/test_alpha.py::test_one",+ ]+++class TestRankTestsByReach:+ def test_more_files_reached_wins_over_alphabetical(self) -> None:+ from repowise.core.analysis.pr_blast import rank_tests_by_reach++ ranked = rank_tests_by_reach({"a.py": ["t_z", "t_both"], "b.py": ["t_a", "t_both"]})+ assert ranked == ["t_both", "t_a", "t_z"]++ def test_single_file_keeps_alphabetical_order(self) -> None:+ # Every test ties at one file reached, so the pre-existing ordering+ # of a single-file change is unchanged.+ from repowise.core.analysis.pr_blast import rank_tests_by_reach++ assert rank_tests_by_reach({"a.py": ["t_c", "t_a", "t_b"]}) == ["t_a", "t_b", "t_c"]++ def test_empty_mapping_is_empty(self) -> None:+ from repowise.core.analysis.pr_blast import rank_tests_by_reach++ assert rank_tests_by_reach({}) == []++@pytest.mark.asyncioasync def test_get_risk_pr_directive_falls_back_to_the_graph_without_a_map(setup_mcp):"""No per-test map -> the import graph answers, labelled as inferred.