feat: Phase 2 complete — 13 Phases of COBOL type classification and test benchmark

P0.6: gcov infrastructure P1: extract_structure output expansion (11 new feature fields) P2: Confusion group rule engine (8 pairs + contradiction + backtrack) P3: 4-factor confidence calculation + quality gate update P4: 33+2 COBOL program type test samples (22 files, 7 categories) P5: parametrized/ test data generation engine P6: japanese_data.py lookup tables P7-10: Type-specific test suites (~159 parametrized tests) P11: Full classification pipeline (classify_program) + orchestrator integration P12: Documentation (module-interfaces, test-plan v3.0, coverage-matrix) Architecture decisions: - classification_pipeline/ merged to hina/pipeline/ - parametrized/ as independent module - japanese_data.py as root-level file - hina/__all__ only exports classify_program() Co-Authored-By: Claude <noreply@anthropic.com>
2026-06-19 23:51:55 +08:00
parent 63b5284715
commit bc1d56d1a4
129 changed files with 19378 additions and 261 deletions
@@ -0,0 +1,152 @@
+"""gcov 覆盖率采集全链路测试
+
+测试内容:
+  1. cobc --coverage 编译含 IF 分支的简单 COBOL 程序
+  2. 运行生成 .gcda 文件
+  3. collect_gcov() 解析 line_rate > 0
+  4. 清理中间产物
+"""
+
+import sys, os, subprocess, tempfile
+from pathlib import Path
+
+sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "..")))
+
+import pytest
+from hina.gcov_collector import collect_gcov
+
+
+HAVE_COBC = None
+
+
+def _check_cobc() -> bool:
+    """检查 cobc 是否在 PATH 且支持 --coverage"""
+    global HAVE_COBC
+    if HAVE_COBC is not None:
+        return HAVE_COBC
+    try:
+        r = subprocess.run(["cobc", "--version"], capture_output=True, text=True, timeout=15)
+        HAVE_COBC = r.returncode == 0
+    except FileNotFoundError:
+        HAVE_COBC = False
+    return HAVE_COBC
+
+
+# ── 嵌入一个简单的 COBOL 程序 (IF 分支) ──
+
+SAMPLE_COBOL = """\
+       IDENTIFICATION DIVISION.
+       PROGRAM-ID. test-gcov.
+       DATA DIVISION.
+       WORKING-STORAGE SECTION.
+       01 WS-X PIC 9(2) VALUE 0.
+       01 WS-Y PIC 9(2) VALUE 0.
+       PROCEDURE DIVISION.
+           MOVE 10 TO WS-X.
+           IF WS-X > 5 THEN
+               MOVE 1 TO WS-Y
+           ELSE
+               MOVE 2 TO WS-Y
+           END-IF.
+           DISPLAY "Y=" WS-Y.
+           STOP RUN.
+"""
+
+
+# ── 夹具: 创建临时目录存放 COBOL 源和编译产物 ──
+
+
+@pytest.fixture
+def work_dir() -> Path:
+    """创建临时工作目录"""
+    with tempfile.TemporaryDirectory(prefix="gcov_test_") as tmp:
+        yield Path(tmp)
+
+
+# ── 辅助函数 ──
+
+
+def _compile_with_coverage(src_path: Path, out_dir: Path) -> bool:
+    """用 cobc --coverage 编译, 返回是否成功"""
+    r = subprocess.run(
+        ["cobc", "-x", "--coverage", str(src_path), "-o", str(out_dir / "test-gcov.exe")],
+        capture_output=True, text=True, timeout=30,
+        cwd=str(out_dir),
+    )
+    if r.returncode != 0:
+        print(f"[compile] stderr: {r.stderr[:300]}")
+    return r.returncode == 0
+
+
+def _run_executable(exe_path: Path, run_dir: Path) -> bool:
+    """运行可执行文件, 返回是否成功"""
+    r = subprocess.run(
+        [str(exe_path)],
+        capture_output=True, text=True, timeout=15,
+        cwd=str(run_dir),
+    )
+    if r.returncode != 0:
+        print(f"[run] stderr: {r.stderr[:300]}")
+    print(f"[run] stdout: {r.stdout.strip()}")
+    return r.returncode == 0
+
+
+# ── 测试用例 ──
+
+
+@pytest.mark.skipif(not _check_cobc(), reason="cobc 未安装或不在 PATH 中")
+def test_gcov_basic_collect(work_dir: Path) -> None:
+    """全链路: 编译 → 运行 → collect_gcov → 验证 line_rate"""
+
+    # 1. 写入 COBOL 源文件
+    src = work_dir / "test-gcov.cbl"
+    src.write_text(SAMPLE_COBOL, encoding="utf-8")
+
+    # 2. 编译 (--coverage)
+    assert _compile_with_coverage(src, work_dir), "cobc --coverage 编译失败"
+
+    # 3. 确认 .gcno 已生成
+    gcno_files = list(work_dir.glob("*.gcno"))
+    assert len(gcno_files) > 0, "编译后未生成 .gcno 文件"
+
+    # 4. 运行程序 (生成 .gcda)
+    exe = work_dir / "test-gcov.exe"
+    assert _run_executable(exe, work_dir), "程序运行失败"
+
+    # 5. 确认 .gcda 已生成
+    gcda_files = list(work_dir.glob("*.gcda"))
+    assert len(gcda_files) > 0, "运行后未生成 .gcda 文件"
+
+    # 6. 调用 collect_gcov() 采集覆盖率
+    result = collect_gcov(cobol_src=src, work_dir=work_dir)
+    print(f"[gcov] collect_gcov returned: {result}")
+
+    # 7. 验证结果
+    assert result["available"] is True, f"覆盖率采集失败: {result.get('reason', 'unknown')}"
+    assert result["line_rate"] > 0, f"line_rate 应为正值, 实际: {result['line_rate']}"
+    assert result["total_lines"] > 0, f"total_lines 应为正值, 实际: {result['total_lines']}"
+    assert result["executed_lines"] > 0, f"executed_lines 应为正值, 实际: {result['executed_lines']}"
+
+    # 8. 验证分支覆盖 (IF 的两路应至少覆盖了一路)
+    assert result["line_rate"] <= 1.0, f"line_rate 不应超过 1.0"
+    print(f"[gcov] ✅ line_rate={result['line_rate']} ({result['executed_lines']}/{result['total_lines']})")
+
+
+@pytest.mark.skipif(not _check_cobc(), reason="cobc 未安装或不在 PATH 中")
+def test_gcov_no_gcda_graceful(work_dir: Path) -> None:
+    """无 .gcda 文件时 collect_gcov 应优雅降级"""
+
+    src = work_dir / "test-gcov.cbl"
+    src.write_text(SAMPLE_COBOL, encoding="utf-8")
+
+    # 编译但不运行, 所以没有 .gcda
+    subprocess.run(
+        ["cobc", "-x", "--coverage", str(src), "-o", str(work_dir / "test-gcov.exe")],
+        capture_output=True, text=True, timeout=30,
+        cwd=str(work_dir),
+    )
+
+    result = collect_gcov(cobol_src=src, work_dir=work_dir)
+    # 没有 .gcda 时应 graceful 返回 {available: False}
+    assert result["available"] is False
+    print(f"[gcov] 无 .gcda 降级正常: {result}")