fix: P1 - complete_tests now feeds DataWriter; P2 - loop syncs complete_tests; P5 - machine_json gets coverage fields

orchestrator.py

@@ -48,10 +48,20 @@ def run_pipeline(cfg: Config, cpath: str, cbl: str, java: str, map_path: str) ->
         if vr.llm_cost > cfg.max_llm_cost:
             return _done(vr, t0, "BLOCKED", 3)
 
-            # ── Phase 1+2: cobol_testgen + HINA Agent + 策略 Agent + 质量门禁 ──
+        # ── Phase 1+2: cobol_testgen + HINA Agent + 策略 Agent + 质量门禁 ──
         try:
             cobol_src_text = Path(cbl).read_text(encoding="utf-8")
-            structure = extract_structure(cobol_src_text)
+            structure = extract_structure(cobol_src_text, source_dir=str(Path(cbl).parent))
+
+            # cobol_testgen 路径枚举 + 基础数据生成
+            base_records = generate_data(cobol_src_text, structure, source_dir=str(Path(cbl).parent))
+            vr.debug["cobol_testgen_records"] = len(base_records)
+            vr.debug["total_branches"] = structure.get("total_branches", 0)
+
+            # 转换为 TestCase 列表（增强管线的基础数据集）
+            complete_tests = []
+            for i, rec in enumerate(base_records):
+                complete_tests.append(TestCase(id=f"CTG-{i+1:04d}", fields=dict(rec)))
 
             # HINA Agent 类型判定
             hina_result = {}
@@ -68,18 +78,13 @@ def run_pipeline(cfg: Config, cpath: str, cbl: str, java: str, map_path: str) ->
                 vr.debug["hina_agent_error"] = str(e)
                 logger.warning(f"[orchestrator] HINA Agent 判定失败: {e}")
 
-            # cobol_testgen 路径枚举 + 基础数据生成
-            base_records = generate_data(cobol_src_text, structure)
-            vr.debug["cobol_testgen_records"] = len(base_records)
-            vr.debug["total_branches"] = structure.get("total_branches", 0)
-
-            base_testcases = []
-            for i, rec in enumerate(base_records):
-                base_testcases.append(TestCase(id=f"CTG-{i+1:04d}", fields=dict(rec)))
-
-            # 策略 Agent 补充
-            strategy_tests = strategy_supplement(base_testcases, hina_result)
-            complete_tests = base_testcases + strategy_tests
+            # 策略 Agent 补充（追加标记记录，统一为 TestCase 格式）
+            for m in strategy_supplement([], hina_result):
+                complete_tests.append(TestCase(
+                    id=m.get("id", f"STG-{len(complete_tests)+1:04d}"),
+                    fields=m.get("fields", {}),
+                    coverage_targets=m.get("coverage_targets", []),
+                ))
 
             # 质量门禁循环
             cov = check_coverage(structure, base_records)
@@ -95,11 +100,17 @@ def run_pipeline(cfg: Config, cpath: str, cbl: str, java: str, map_path: str) ->
                 if gaps and structure.get("branch_tree_obj"):
                     delta = incremental_supplement(structure["branch_tree_obj"], gaps)
                     base_records.extend(delta)
+                    # 同步更新 complete_tests
+                    for i, d in enumerate(delta):
+                        complete_tests.append(TestCase(
+                            id=f"CTG-S{attempt+1}-{i+1:04d}",
+                            fields=dict(d),
+                        ))
                     cov = check_coverage(structure, base_records)
                 else:
                     break
 
-            vr.paragraph_rate = cov.get("paragraph_rate", 0.0)
+            vr.paragraph_rate = 0.0  # Phase 3 通过 gcov 获取精确值
             vr.branch_rate = cov.get("branch_rate", 0.0)
             vr.decision_rate = cov.get("decision_rate", 0.0)
 
@@ -112,6 +123,7 @@ def run_pipeline(cfg: Config, cpath: str, cbl: str, java: str, map_path: str) ->
 
         suite = Agent2Data(llm).design(tree, cfg.coverage_default, cfg.runner_mode == "spark")
         vr.llm_cost += 0.002
+        suite.test_cases = complete_tests  # 替换为增强管线数据（P1/P2 修复）
         vr.debug["test_cases"] = [{"id":tc.id,"fields":tc.fields,"targets":tc.coverage_targets} for tc in suite.test_cases]
         vr.debug["spark_config"] = {"records":suite.spark_config.num_records} if suite.has_spark else None