import importlib.util import json import sys from datetime import date from pathlib import Path import pytest _SCRIPTS_DIR = Path(__file__).parent.parent sys.path.insert(0, str(_SCRIPTS_DIR)) _MODULE_PATH = _SCRIPTS_DIR / "learning-analyzer.py" spec = importlib.util.spec_from_file_location("learning_analyzer", _MODULE_PATH) assert spec is not None learning_analyzer = importlib.util.module_from_spec(spec) assert spec.loader is not None spec.loader.exec_module(learning_analyzer) # --------------------------------------------------------------------------- # 테스트 데이터 헬퍼 # --------------------------------------------------------------------------- def make_audit_trail(tmp_path: Path, records: list[dict]) -> Path: f = tmp_path / "audit-trail.jsonl" lines = [json.dumps(r) for r in records] f.write_text("\n".join(lines), encoding="utf-8") return f def make_whitelist(tmp_path: Path, files: list[str] | None = None, patterns: list[str] | None = None) -> Path: content = "" if files: content += "files:\n" for f in files: content += f' - "{f}"\n' if patterns: content += "patterns:\n" for p in patterns: content += f' - "{p}"\n' f = tmp_path / "whitelist.yaml" f.write_text(content, encoding="utf-8") return f def make_thresholds(tmp_path: Path, min_task_count: int = 3, min_task_ratio: float = 0.3) -> Path: content = f"min_task_count: {min_task_count}\nmin_task_ratio: {min_task_ratio}\n" f = tmp_path / "thresholds.yaml" f.write_text(content, encoding="utf-8") return f def make_sample_records(file_path: str, task_ids: list[str], ts_base: str = "2026-04-07") -> list[dict]: records = [] for i, tid in enumerate(task_ids): records.append( { "ts": f"{ts_base}T{10+i:02d}:00:00Z", "bot": "dev1", "tool": "Edit", "file": f"/home/jay/workspace/{file_path}", "task_id": tid, "schema_version": 2, } ) return records # --------------------------------------------------------------------------- # 1. TestLoadWhitelist # --------------------------------------------------------------------------- class TestLoadWhitelist: def test_load_whitelist_normal(self, tmp_path: Path): wl_path = make_whitelist( tmp_path, files=["/home/jay/workspace/config/settings.yaml"], patterns=["config/*", "tests/*"], ) result = learning_analyzer.load_whitelist(wl_path) assert "files" in result or "patterns" in result def test_load_whitelist_missing(self, tmp_path: Path): missing = tmp_path / "nonexistent.yaml" result = learning_analyzer.load_whitelist(missing) assert result == {} or isinstance(result, dict) def test_load_whitelist_empty(self, tmp_path: Path): f = tmp_path / "empty.yaml" f.write_text("", encoding="utf-8") result = learning_analyzer.load_whitelist(f) assert result == {} or isinstance(result, dict) # --------------------------------------------------------------------------- # 2. TestLoadThresholds # --------------------------------------------------------------------------- class TestLoadThresholds: def test_load_thresholds_normal(self, tmp_path: Path): th_path = make_thresholds(tmp_path, min_task_count=5, min_task_ratio=0.4) result = learning_analyzer.load_thresholds(th_path) assert result["min_task_count"] == 5 assert result["min_task_ratio"] == pytest.approx(0.4) def test_load_thresholds_missing(self, tmp_path: Path): missing = tmp_path / "nonexistent.yaml" result = learning_analyzer.load_thresholds(missing) assert result["min_task_count"] == 3 assert result["min_task_ratio"] == pytest.approx(0.3) def test_load_thresholds_partial(self, tmp_path: Path): f = tmp_path / "partial.yaml" f.write_text("min_task_count: 5\n", encoding="utf-8") result = learning_analyzer.load_thresholds(f) assert result["min_task_count"] == 5 assert result["min_task_ratio"] == pytest.approx(0.3) # --------------------------------------------------------------------------- # 3. TestParseAuditTrail # --------------------------------------------------------------------------- class TestParseAuditTrail: def test_parse_normal(self, tmp_path: Path): records = make_sample_records("src/app.py", ["task-1.1", "task-2.1"]) trail_path = make_audit_trail(tmp_path, records) week_start = date(2026, 4, 7) week_end = date(2026, 4, 13) result = learning_analyzer.parse_audit_trail(trail_path, week_start, week_end) assert len(result) == 2 def test_parse_excludes_outside_range(self, tmp_path: Path): inside = make_sample_records("src/app.py", ["task-1.1"], ts_base="2026-04-08") outside = make_sample_records("src/other.py", ["task-2.1"], ts_base="2026-04-01") trail_path = make_audit_trail(tmp_path, inside + outside) week_start = date(2026, 4, 7) week_end = date(2026, 4, 13) result = learning_analyzer.parse_audit_trail(trail_path, week_start, week_end) assert len(result) == 1 assert result[0]["task_id"] == "task-1.1" def test_parse_excludes_tmp_paths(self, tmp_path: Path): records = [ { "ts": "2026-04-08T10:00:00Z", "bot": "dev1", "tool": "Edit", "file": "/tmp/scratch.py", "task_id": "task-1.1", "schema_version": 2, } ] trail_path = make_audit_trail(tmp_path, records) week_start = date(2026, 4, 7) week_end = date(2026, 4, 13) result = learning_analyzer.parse_audit_trail(trail_path, week_start, week_end) assert len(result) == 0 def test_parse_excludes_anu(self, tmp_path: Path): records = [ { "ts": "2026-04-08T10:00:00Z", "bot": "anu", "tool": "Edit", "file": "/home/jay/workspace/src/app.py", "task_id": "task-1.1", "schema_version": 2, } ] trail_path = make_audit_trail(tmp_path, records) week_start = date(2026, 4, 7) week_end = date(2026, 4, 13) result = learning_analyzer.parse_audit_trail(trail_path, week_start, week_end) assert len(result) == 0 def test_parse_excludes_unknown_task_id(self, tmp_path: Path): records = [ { "ts": "2026-04-08T10:00:00Z", "bot": "dev1", "tool": "Edit", "file": "/home/jay/workspace/src/app.py", "task_id": "unknown", "schema_version": 2, }, { "ts": "2026-04-08T11:00:00Z", "bot": "dev1", "tool": "Edit", "file": "/home/jay/workspace/src/app.py", "task_id": "", "schema_version": 2, }, ] trail_path = make_audit_trail(tmp_path, records) week_start = date(2026, 4, 7) week_end = date(2026, 4, 13) result = learning_analyzer.parse_audit_trail(trail_path, week_start, week_end) assert len(result) == 0 def test_parse_empty_file(self, tmp_path: Path): trail_path = make_audit_trail(tmp_path, []) week_start = date(2026, 4, 7) week_end = date(2026, 4, 13) result = learning_analyzer.parse_audit_trail(trail_path, week_start, week_end) assert result == [] def test_parse_missing_file(self, tmp_path: Path): missing = tmp_path / "nonexistent.jsonl" week_start = date(2026, 4, 7) week_end = date(2026, 4, 13) result = learning_analyzer.parse_audit_trail(missing, week_start, week_end) assert result == [] # --------------------------------------------------------------------------- # 4. TestComputeHotspots # --------------------------------------------------------------------------- class TestComputeHotspots: _THRESHOLDS_DEFAULT = {"min_task_count": 3, "min_task_ratio": 0.3} _WHITELIST_EMPTY = {"files": [], "patterns": []} def test_hotspot_above_threshold(self): records = make_sample_records("src/app.py", ["task-1.1", "task-2.1", "task-3.1", "task-4.1"]) thresholds = {"min_task_count": 3, "min_task_ratio": 0.1} hotspots, _ = learning_analyzer.compute_hotspots(records, self._WHITELIST_EMPTY, thresholds) assert len(hotspots) == 1 assert "app.py" in hotspots[0]["file"] def test_hotspot_below_threshold(self): records = make_sample_records("src/app.py", ["task-1.1", "task-2.1"]) hotspots, _ = learning_analyzer.compute_hotspots(records, self._WHITELIST_EMPTY, self._THRESHOLDS_DEFAULT) assert len(hotspots) == 0 def test_hotspot_ratio_check(self): records_app = make_sample_records("src/app.py", ["task-1.1", "task-2.1", "task-3.1"]) total_task_count = 30 extra_records = [] for i in range(4, total_task_count + 1): extra_records.append( { "ts": f"2026-04-07T10:00:00Z", "bot": "dev1", "tool": "Edit", "file": f"/home/jay/workspace/src/other_{i}.py", "task_id": f"task-{i}.1", "schema_version": 2, } ) thresholds = {"min_task_count": 3, "min_task_ratio": 0.5} hotspots, _ = learning_analyzer.compute_hotspots(records_app + extra_records, self._WHITELIST_EMPTY, thresholds) assert len(hotspots) == 0 def test_hotspot_and_condition(self): records = make_sample_records("src/app.py", ["task-1.1", "task-2.1", "task-3.1"]) extra = [] for i in range(4, 20): extra.append( { "ts": "2026-04-07T10:00:00Z", "bot": "dev1", "tool": "Edit", "file": f"/home/jay/workspace/src/other_{i}.py", "task_id": f"task-{i}.1", "schema_version": 2, } ) thresholds = {"min_task_count": 3, "min_task_ratio": 0.5} hotspots, _ = learning_analyzer.compute_hotspots(records + extra, self._WHITELIST_EMPTY, thresholds) assert len(hotspots) == 0 def test_whitelist_exclusion(self): records = make_sample_records("src/app.py", ["task-1.1", "task-2.1", "task-3.1", "task-4.1"]) whitelist = {"files": {"src/app.py"}, "patterns": []} thresholds = {"min_task_count": 3, "min_task_ratio": 0.1} hotspots, whitelisted = learning_analyzer.compute_hotspots(records, whitelist, thresholds) assert len(hotspots) == 0 assert len(whitelisted) == 1 def test_whitelist_pattern_matching(self): records = make_sample_records("config/settings.yaml", ["task-1.1", "task-2.1", "task-3.1", "task-4.1"]) whitelist = {"files": [], "patterns": ["config/*"]} thresholds = {"min_task_count": 3, "min_task_ratio": 0.1} hotspots, whitelisted = learning_analyzer.compute_hotspots(records, whitelist, thresholds) assert len(hotspots) == 0 assert len(whitelisted) == 1 def test_hotspot_type_classification(self): records_py = make_sample_records("src/app.py", ["task-1.1", "task-2.1", "task-3.1", "task-4.1"]) records_yaml = make_sample_records("config/settings.yaml", ["task-1.1", "task-2.1", "task-3.1", "task-4.1"]) thresholds = {"min_task_count": 3, "min_task_ratio": 0.1} hotspots, _ = learning_analyzer.compute_hotspots(records_py + records_yaml, self._WHITELIST_EMPTY, thresholds) types = {h["file"]: h.get("type") for h in hotspots} py_file = next((h["file"] for h in hotspots if h["file"].endswith(".py")), None) yaml_file = next((h["file"] for h in hotspots if h["file"].endswith(".yaml")), None) assert py_file is not None assert yaml_file is not None assert types[py_file] == "refactor_candidate" assert types[yaml_file] == "config_hotspot" def test_hotspot_priority_classification(self): records_high = make_sample_records("src/high.py", ["task-1.1", "task-2.1", "task-3.1", "task-4.1", "task-5.1"]) records_medium = make_sample_records("src/medium.py", ["task-1.1", "task-2.1", "task-3.1"]) thresholds = {"min_task_count": 3, "min_task_ratio": 0.1} hotspots, _ = learning_analyzer.compute_hotspots( records_high + records_medium, self._WHITELIST_EMPTY, thresholds ) priority_map = {h["file"]: h.get("priority") for h in hotspots} high_file = next((h["file"] for h in hotspots if "high.py" in h["file"]), None) medium_file = next((h["file"] for h in hotspots if "medium.py" in h["file"]), None) assert high_file is not None assert medium_file is not None assert priority_map[high_file] == "high" assert priority_map[medium_file] == "medium" def test_hotspot_includes_task_list(self): task_ids = ["task-1.1", "task-2.1", "task-3.1", "task-4.1"] records = make_sample_records("src/app.py", task_ids) thresholds = {"min_task_count": 3, "min_task_ratio": 0.1} hotspots, _ = learning_analyzer.compute_hotspots(records, self._WHITELIST_EMPTY, thresholds) assert len(hotspots) == 1 assert "task_ids" in hotspots[0] or "tasks" in hotspots[0] task_list = hotspots[0].get("task_ids") or hotspots[0].get("tasks") for tid in task_ids: assert tid in task_list def test_empty_records(self): hotspots, whitelisted = learning_analyzer.compute_hotspots([], self._WHITELIST_EMPTY, self._THRESHOLDS_DEFAULT) assert hotspots == [] assert whitelisted == [] # --------------------------------------------------------------------------- # 5. TestGenerateReport # --------------------------------------------------------------------------- class TestGenerateReport: _WEEK_START = date(2026, 4, 7) _WEEK_END = date(2026, 4, 14) def _make_hotspot(self, file_path: str, task_count: int = 4, priority: str = "high", total_tasks: int = 10) -> dict: return { "file": file_path, "task_count": task_count, "task_ratio": task_count / total_tasks if total_tasks else 0, "task_ids": [f"task-{i}.1" for i in range(1, task_count + 1)], "type": "refactor_candidate", "priority": priority, "total_tasks": total_tasks, } def test_report_created(self, tmp_path: Path): hotspots = [self._make_hotspot("/home/jay/workspace/src/app.py")] output_path = learning_analyzer.generate_report(hotspots, [], self._WEEK_START, self._WEEK_END, 10, tmp_path) assert output_path.exists() def test_report_contains_header(self, tmp_path: Path): hotspots = [self._make_hotspot("/home/jay/workspace/src/app.py")] output_path = learning_analyzer.generate_report(hotspots, [], self._WEEK_START, self._WEEK_END, 10, tmp_path) content = output_path.read_text(encoding="utf-8") assert "2026-04-07" in content assert "2026-04-13" in content def test_report_contains_v1_limitation(self, tmp_path: Path): output_path = learning_analyzer.generate_report([], [], self._WEEK_START, self._WEEK_END, 0, tmp_path) content = output_path.read_text(encoding="utf-8") assert "학습 기능 v1" in content assert "에러 유사도 미지원" in content assert "v2 예정" in content def test_report_contains_hotspot_status(self, tmp_path: Path): hotspots = [self._make_hotspot("/home/jay/workspace/src/app.py")] output_path = learning_analyzer.generate_report(hotspots, [], self._WEEK_START, self._WEEK_END, 10, tmp_path) content = output_path.read_text(encoding="utf-8") assert "**status**: pending" in content def test_report_contains_summary(self, tmp_path: Path): hotspots = [self._make_hotspot("/home/jay/workspace/src/app.py")] output_path = learning_analyzer.generate_report(hotspots, [], self._WEEK_START, self._WEEK_END, 10, tmp_path) content = output_path.read_text(encoding="utf-8") assert "summary" in content.lower() or "요약" in content def test_report_filename_format(self, tmp_path: Path): output_path = learning_analyzer.generate_report([], [], self._WEEK_START, self._WEEK_END, 0, tmp_path) assert output_path.name.startswith("weekly-") assert output_path.name.endswith(".md") assert "2026" in output_path.name # --------------------------------------------------------------------------- # 6. TestCLI # --------------------------------------------------------------------------- class TestCLI: _WEEK_START = date(2026, 4, 7) _WEEK_END = date(2026, 4, 14) def test_cli_with_sample_data(self, tmp_path: Path): records = make_sample_records( "src/hot.py", ["task-1.1", "task-2.1", "task-3.1", "task-4.1"] ) + make_sample_records("src/normal.py", ["task-1.1"]) trail_path = make_audit_trail(tmp_path, records) wl_path = make_whitelist(tmp_path) th_path = make_thresholds(tmp_path) output_dir = tmp_path / "reports" output_dir.mkdir() parsed = learning_analyzer.parse_audit_trail(trail_path, self._WEEK_START, self._WEEK_END) whitelist = learning_analyzer.load_whitelist(wl_path) thresholds = learning_analyzer.load_thresholds(th_path) hotspots, whitelisted = learning_analyzer.compute_hotspots(parsed, whitelist, thresholds) total_tasks = len({r["task_id"] for r in parsed}) report_path = learning_analyzer.generate_report( hotspots, whitelisted, self._WEEK_START, self._WEEK_END, total_tasks, output_dir ) assert report_path.exists() content = report_path.read_text(encoding="utf-8") assert "hot.py" in content def test_cli_missing_audit_trail(self, tmp_path: Path): missing = tmp_path / "nonexistent.jsonl" wl_path = make_whitelist(tmp_path) th_path = make_thresholds(tmp_path) output_dir = tmp_path / "reports" output_dir.mkdir() try: parsed = learning_analyzer.parse_audit_trail(missing, self._WEEK_START, self._WEEK_END) assert parsed == [] whitelist = learning_analyzer.load_whitelist(wl_path) thresholds = learning_analyzer.load_thresholds(th_path) hotspots, whitelisted = learning_analyzer.compute_hotspots(parsed, whitelist, thresholds) report_path = learning_analyzer.generate_report( hotspots, whitelisted, self._WEEK_START, self._WEEK_END, 0, output_dir ) assert report_path.exists() except SystemExit: pass except Exception as e: pytest.fail(f"audit-trail 없을 때 예상치 못한 예외 발생: {e}")