dsh-crwu-workbench 0.0.10

package/skills/crwu/crwu-audit/scripts/test_fetch_review_records.py

@@ -0,0 +1,77 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""`fetch_review_records.py` 契约测试（阶段二按清单取回，只读守卫）。
+
+自洽：只依赖本技能 `scripts/` 内脚本，构造临时 fixture，可随技能安装运行。
+"""
+from __future__ import annotations
+
+import json
+import subprocess
+import sys
+import tempfile
+import unittest
+from pathlib import Path
+
+SCRIPT = Path(__file__).resolve().parent / "fetch_review_records.py"
+
+
+def _run(*args) -> subprocess.CompletedProcess:
+    return subprocess.run([sys.executable, str(SCRIPT), *args],
+                          capture_output=True, text=True)
+
+
+class FetchReviewRecordsContractTest(unittest.TestCase):
+    def setUp(self):
+        self.tmp = tempfile.TemporaryDirectory()
+        self.case = Path(self.tmp.name) / "case"
+        self.src = Path(self.tmp.name) / "all"
+        self.case.mkdir(parents=True)
+        self.src.mkdir(parents=True)
+
+    def tearDown(self):
+        self.tmp.cleanup()
+
+    def _manifest(self, entries):
+        p = self.case / "排除清单.json"
+        p.write_text(json.dumps(entries, ensure_ascii=False), encoding="utf-8")
+        return str(p)
+
+    def test_fetches_allowlisted_files_and_writes_manifest(self):
+        (self.src / "G15-4三级审核意见.docx").write_bytes(b"review" * 100)
+        manifest = self._manifest([
+            {"fileId": "f1", "name": "G15-4三级审核意见.docx", "sizeBytes": 600},
+        ])
+        r = _run("--case", str(self.case), "--manifest", manifest, "--source", str(self.src))
+        self.assertEqual(0, r.returncode, r.stderr)
+        out = self.case / "复核-人工" / "G15-4三级审核意见.docx"
+        self.assertTrue(out.is_file())
+        self.assertTrue((self.case / "复核-人工" / ".fetch-manifest.json").is_file())
+
+    def test_size_mismatch_is_refused(self):
+        (self.src / "A.docx").write_bytes(b"x" * 10)
+        manifest = self._manifest([{"fileId": "f1", "name": "A.docx", "sizeBytes": 999}])
+        r = _run("--case", str(self.case), "--manifest", manifest, "--source", str(self.src))
+        self.assertEqual(0, r.returncode)  # 守卫不因单条失败而中止整体
+        self.assertFalse((self.case / "复核-人工" / "A.docx").exists())
+        res = json.loads((self.case / "复核-人工" / ".fetch-manifest.json").read_text(encoding="utf-8"))
+        self.assertFalse(res["results"][0]["ok"])
+        self.assertIn("字节数不符", res["results"][0]["reason"])
+
+    def test_missing_source_file_is_recorded_not_fetched(self):
+        manifest = self._manifest([{"fileId": "f1", "name": "不存在.docx", "sizeBytes": 1}])
+        _run("--case", str(self.case), "--manifest", manifest, "--source", str(self.src))
+        res = json.loads((self.case / "复核-人工" / ".fetch-manifest.json").read_text(encoding="utf-8"))
+        self.assertFalse(res["results"][0]["ok"])
+        self.assertIn("未找到", res["results"][0]["reason"])
+
+    def test_output_must_not_be_inside_source_materials(self):
+        (self.src / "A.docx").write_bytes(b"x" * 10)
+        manifest = self._manifest([{"fileId": "f1", "name": "A.docx", "sizeBytes": 10}])
+        r = _run("--case", str(self.case), "--manifest", manifest, "--source", str(self.src),
+                 "--out", str(self.case / "材料-源" / "复核"))
+        self.assertEqual(2, r.returncode, "落盘目录在材料-源内必须被拒绝")
+
+
+if __name__ == "__main__":
+    unittest.main(verbosity=2)

package/skills/crwu/crwu-audit/scripts/test_media_extract.py

@@ -0,0 +1,332 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""`media_extract.py` 契约测试（图片证据通道 + H0 媒体锚点）。
+
+自洽：只用本技能 `scripts/` 内的模块与标准库造 fixture，可随技能一起安装运行。
+每条用例都对应一次**真实失效或硬口径**：
+
+1. 内嵌图必须被**导出**（此前只登记计数、无通道 → 证据在图片里就被判"为空/缺失"，
+   真实失效 2026-302150-LX9757-BG8677 的 `MKT-004`）。
+2. 独立图片必须落盘并可被宿主视觉读取（此前 note 只有"图片：无 OCR"，没有任何路径）。
+3. docx 正文内联图必须按**段落序**定位（拍品/询价截图证据位置可核）。
+4. **H0**：锚点落在隐藏行/列或隐藏 sheet 的媒体不导出、不定位，只记数量。
+5. **fail-closed**：隐藏结构不可得时不导出（不得据不可判定锚点读取隐藏区）。
+6. 锚点不可判定（`absoluteAnchor`）与被正文引用的页眉页脚图：不导出，记未核原因
+   （**未核 ≠ 缺失**）。
+7. 上限护栏：单件媒体超限即停并留痕，不静默截断。
+8. 缺依赖：`pypdf` 缺失 / `.xls` 老二进制 → 记未核原因，不静默空返。
+"""
+from __future__ import annotations
+
+import base64
+import importlib.util
+import json
+import tempfile
+import unittest
+import zipfile
+from pathlib import Path
+
+SCRIPTS_DIR = Path(__file__).resolve().parent
+
+PNG = base64.b64decode(
+    b"iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8z8DwHwAFAAH/q842iQAAAABJRU5ErkJggg==")
+
+
+def _load(name: str):
+    spec = importlib.util.spec_from_file_location(name + "_for_test", SCRIPTS_DIR / f"{name}.py")
+    module = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(module)
+    return module
+
+
+MEDIA = _load("media_extract")
+PREPARE = _load("prepare_materials")
+
+
+# ---------------------------------------------------------------- fixture 构造
+
+def _anchor(kind: str, rid: str, row: int, col: int = 0, row2: int = None, col2: int = None) -> str:
+    pic = (f'<xdr:pic><xdr:nvPicPr><xdr:cNvPr id="1" name="P1"/><xdr:cNvPicPr/></xdr:nvPicPr>'
+           f'<xdr:blipFill><a:blip r:embed="{rid}"/></xdr:blipFill><xdr:spPr/></xdr:pic>')
+    frm = (f'<xdr:from><xdr:col>{col}</xdr:col><xdr:colOff>0</xdr:colOff>'
+           f'<xdr:row>{row}</xdr:row><xdr:rowOff>0</xdr:rowOff></xdr:from>')
+    if kind == "absolute":
+        return f'<xdr:absoluteAnchor><xdr:pos x="0" y="0"/><xdr:ext cx="1" cy="1"/>{pic}</xdr:absoluteAnchor>'
+    if kind == "twoCell":
+        to = (f'<xdr:to><xdr:col>{col2 if col2 is not None else col}</xdr:col><xdr:colOff>0</xdr:colOff>'
+              f'<xdr:row>{row2 if row2 is not None else row}</xdr:row><xdr:rowOff>0</xdr:rowOff></xdr:to>')
+        return f'<xdr:twoCellAnchor>{frm}{to}{pic}</xdr:twoCellAnchor>'
+    return f'<xdr:oneCellAnchor>{frm}<xdr:ext cx="1" cy="1"/>{pic}</xdr:oneCellAnchor>'
+
+
+def make_xlsx(path: Path, anchors: list, hidden_rows=(), hidden_cols=(), sheet="表1") -> Path:
+    """造一个带内嵌图（含锚点）的 .xlsx：openpyxl 不写图，直接在 zip 层补绘图部件。"""
+    import openpyxl
+    wb = openpyxl.Workbook()
+    ws = wb.active
+    ws.title = sheet
+    ws["A1"] = "可比实例位置（见下图）"
+    for r in hidden_rows:
+        ws.row_dimensions[r].hidden = True
+    for c in hidden_cols:
+        ws.column_dimensions[c].hidden = True
+    wb.save(path)
+
+    body = "".join(_anchor(a["kind"], a["rid"], a["row"], a.get("col", 0),
+                           a.get("row2"), a.get("col2")) for a in anchors)
+    drawing = ('<xdr:wsDr xmlns:xdr="http://schemas.openxmlformats.org/drawingml/2006/spreadsheetDrawing"'
+               ' xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main"'
+               ' xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships">'
+               + body + '</xdr:wsDr>')
+    rels = "".join(
+        f'<Relationship Id="{a["rid"]}" Type="http://schemas.openxmlformats.org/officeDocument/'
+        f'2006/relationships/image" Target="../media/{a["media"]}"/>' for a in anchors)
+
+    tmp = path.with_suffix(".patch.xlsx")
+    with zipfile.ZipFile(path) as zin, zipfile.ZipFile(tmp, "w", zipfile.ZIP_DEFLATED) as zout:
+        have = set(zin.namelist())
+        for item in zin.infolist():
+            data = zin.read(item.filename)
+            if item.filename == "[Content_Types].xml":
+                data = data.decode().replace(
+                    "</Types>",
+                    '<Default Extension="png" ContentType="image/png"/>'
+                    '<Override PartName="/xl/drawings/drawing1.xml" ContentType="application/'
+                    'vnd.openxmlformats-officedocument.drawing+xml"/></Types>').encode()
+            elif item.filename == "xl/worksheets/sheet1.xml":
+                data = data.decode().replace(
+                    "</worksheet>",
+                    '<drawing xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/'
+                    'relationships" r:id="rIdImg"/></worksheet>').encode()
+            zout.writestr(item, data)
+        if "xl/worksheets/_rels/sheet1.xml.rels" not in have:
+            zout.writestr("xl/worksheets/_rels/sheet1.xml.rels",
+                          '<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/'
+                          'relationships"><Relationship Id="rIdImg" Type="http://schemas.'
+                          'openxmlformats.org/officeDocument/2006/relationships/drawing" '
+                          'Target="../drawings/drawing1.xml"/></Relationships>')
+        zout.writestr("xl/drawings/drawing1.xml", drawing)
+        zout.writestr("xl/drawings/_rels/drawing1.xml.rels",
+                      '<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/'
+                      'relationships">' + rels + '</Relationships>')
+        for a in anchors:
+            zout.writestr("xl/media/" + a["media"], PNG)
+    path.write_bytes(tmp.read_bytes())
+    tmp.unlink()
+    return path
+
+
+def make_docx(path: Path, header_image: bool = False) -> Path:
+    """造一个正文含内联图（第 3 段）的 .docx；可选在页眉也放一张图。"""
+    ct = ('<?xml version="1.0"?><Types xmlns="http://schemas.openxmlformats.org/package/2006/'
+          'content-types"><Default Extension="rels" ContentType="application/vnd.'
+          'openxmlformats-package.relationships+xml"/><Default Extension="xml" '
+          'ContentType="application/xml"/><Default Extension="png" ContentType="image/png"/>'
+          '<Override PartName="/word/document.xml" ContentType="application/vnd.'
+          'openxmlformats-officedocument.wordprocessingml.document.main+xml"/>'
+          + ('<Override PartName="/word/header1.xml" ContentType="application/vnd.'
+             'openxmlformats-officedocument.wordprocessingml.header+xml"/>' if header_image else '')
+          + '</Types>')
+    rels = ('<?xml version="1.0"?><Relationships xmlns="http://schemas.openxmlformats.org/package/'
+            '2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/'
+            'officeDocument/2006/relationships/officeDocument" Target="word/document.xml"/>'
+            + ('<Relationship Id="rIdH" Type="http://schemas.openxmlformats.org/officeDocument/'
+               '2006/relationships/header" Target="word/header1.xml"/>' if header_image else '')
+            + '</Relationships>')
+    doc = ('<?xml version="1.0"?><w:document xmlns:w="http://schemas.openxmlformats.org/'
+           'wordprocessingml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/'
+           '2006/relationships" xmlns:wp="http://schemas.openxmlformats.org/drawingml/2006/'
+           'wordprocessingDrawing" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main">'
+           '<w:body>'
+           '<w:p><w:r><w:t>三、评估结论：本次采用市场法。</w:t></w:r></w:p>'
+           '<w:p><w:r><w:t>以下为询价截图：</w:t></w:r></w:p>'
+           '<w:p><w:r><w:drawing><wp:inline><wp:extent cx="1" cy="1"/><wp:docPr id="1" name="P1"/>'
+           '<a:graphic><a:graphicData uri="http://schemas.openxmlformats.org/drawingml/2006/picture">'
+           '<a:blip r:embed="rId10"/></a:graphicData></a:graphic></wp:inline></w:drawing></w:r></w:p>'
+           '<w:p><w:r><w:t>截图显示询价 1,234 元/平方米。</w:t></w:r></w:p>'
+           '</w:body></w:document>')
+    doc_rels = ('<?xml version="1.0"?><Relationships xmlns="http://schemas.openxmlformats.org/package/'
+                '2006/relationships"><Relationship Id="rId10" Type="http://schemas.openxmlformats.org/'
+                'officeDocument/2006/relationships/image" Target="media/image1.png"/></Relationships>')
+    with zipfile.ZipFile(path, "w", zipfile.ZIP_DEFLATED) as z:
+        z.writestr("[Content_Types].xml", ct)
+        z.writestr("_rels/.rels", rels)
+        z.writestr("word/document.xml", doc)
+        z.writestr("word/_rels/document.xml.rels", doc_rels)
+        z.writestr("word/media/image1.png", PNG)
+        if header_image:
+            z.writestr("word/header1.xml",
+                       '<?xml version="1.0"?><w:hdr xmlns:w="http://schemas.openxmlformats.org/'
+                       'wordprocessingml/2006/main" xmlns:r="http://schemas.openxmlformats.org/'
+                       'officeDocument/2006/relationships"><w:p><w:r><w:drawing>'
+                       '<a:blip xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" '
+                       'r:embed="rId9"/></w:drawing></w:r></w:p></w:hdr>')
+            z.writestr("word/_rels/header1.xml.rels",
+                       '<?xml version="1.0"?><Relationships xmlns="http://schemas.openxmlformats.org/'
+                       'package/2006/relationships"><Relationship Id="rId9" Type="http://schemas.'
+                       'openxmlformats.org/officeDocument/2006/relationships/image" '
+                       'Target="media/image2.png"/></Relationships>')
+            z.writestr("word/media/image2.png", PNG)
+    return path
+
+
+class MediaExtractTest(unittest.TestCase):
+    def setUp(self):
+        self.tmp = tempfile.TemporaryDirectory()
+        self.case = Path(self.tmp.name)
+        self.out = self.case / "媒体证据"
+
+    def tearDown(self):
+        self.tmp.cleanup()
+
+    def _plan(self, path: Path, ext: str, kind: str, hidden=None, budget=None):
+        return MEDIA.plan_media(str(path), ext, kind, hidden, str(self.out), str(self.case),
+                                str(path.relative_to(self.case)), "定稿", budget=budget)
+
+    # ---- 1 xlsx 内嵌图必须导出（含锚点定位）-----------------------------
+    def test_xlsx_embedded_image_is_exported_with_anchor(self):
+        import openpyxl  # noqa: F401
+        p = make_xlsx(self.case / "计算表.xlsx", [{"kind": "oneCell", "rid": "rId1",
+                                                   "row": 3, "media": "image1.png"}])
+        hidden = PREPARE._hidden_metadata(openpyxl.load_workbook(p),
+                                          openpyxl.load_workbook(p, data_only=False), path=str(p))
+        res = self._plan(p, ".xlsx", "zip-ooxml", hidden)
+
+        self.assertEqual(1, res["summary"]["exportedCount"], res["summary"])
+        entry = res["entries"][0]
+        self.assertEqual("xlsx内嵌图", entry["kind"])
+        self.assertEqual("表1!图片#1（锚点 4:4）", entry["locator"])
+        self.assertEqual([4, 4], entry["anchorRows"])
+        exported = self.case / entry["localPath"]
+        self.assertTrue(exported.exists(), "必须真的落盘，不能只登记计数")
+        self.assertEqual(PNG, exported.read_bytes())
+        self.assertEqual("host-vision", entry["evidenceChannel"])
+
+    def test_xlsx_two_cell_anchor_reports_row_span(self):
+        import openpyxl  # noqa: F401
+        p = make_xlsx(self.case / "跨行.xlsx",
+                      [{"kind": "twoCell", "rid": "rId1", "row": 5, "row2": 8, "media": "image1.png"}])
+        hidden = PREPARE._hidden_metadata(openpyxl.load_workbook(p),
+                                          openpyxl.load_workbook(p, data_only=False), path=str(p))
+        res = self._plan(p, ".xlsx", "zip-ooxml", hidden)
+        self.assertEqual([6, 9], res["entries"][0]["anchorRows"], "twoCellAnchor 必须给出起止行")
+
+    # ---- 2 H0：隐藏锚点不导出 -------------------------------------------
+    def test_hidden_anchored_media_is_skipped_not_exported(self):
+        import openpyxl
+        p = make_xlsx(self.case / "隐藏锚点.xlsx",
+                      [{"kind": "oneCell", "rid": "rId1", "row": 0, "media": "image1.png"},
+                       {"kind": "oneCell", "rid": "rId2", "row": 3, "media": "image2.png"}],
+                      hidden_rows=[4])
+        hidden = PREPARE._hidden_metadata(openpyxl.load_workbook(p),
+                                          openpyxl.load_workbook(p, data_only=False), path=str(p))
+        res = self._plan(p, ".xlsx", "zip-ooxml", hidden)
+
+        self.assertEqual(1, res["summary"]["exportedCount"], "可见锚点的图必须导出")
+        self.assertEqual(1, res["summary"]["hiddenSkippedCount"], "隐藏行锚点的图必须跳过")
+        self.assertEqual([0], [e["anchorRows"][0] - 1 for e in res["entries"]])
+        blob = json.dumps(res, ensure_ascii=False)
+        self.assertNotIn("image2.png", blob, "隐藏锚点媒体不得进入导出清单（不定位）")
+
+    def test_hidden_column_anchored_media_is_skipped(self):
+        import openpyxl
+        p = make_xlsx(self.case / "隐藏列锚点.xlsx",
+                      [{"kind": "oneCell", "rid": "rId1", "row": 0, "col": 1, "media": "image1.png"}],
+                      hidden_cols=["B"])
+        hidden = PREPARE._hidden_metadata(openpyxl.load_workbook(p),
+                                          openpyxl.load_workbook(p, data_only=False), path=str(p))
+        res = self._plan(p, ".xlsx", "zip-ooxml", hidden)
+        self.assertEqual(0, res["summary"]["exportedCount"])
+        self.assertEqual(1, res["summary"]["hiddenSkippedCount"])
+
+    def test_hidden_structure_unavailable_fails_closed(self):
+        """隐藏结构不可得（如原件包用其他工具解析失败）→ 整件不导出，记未核原因。"""
+        import openpyxl  # noqa: F401
+        p = make_xlsx(self.case / "无隐藏元数据.xlsx",
+                      [{"kind": "oneCell", "rid": "rId1", "row": 0, "media": "image1.png"}])
+        res = self._plan(p, ".xlsx", "zip-ooxml", None)
+        self.assertEqual(0, res["summary"]["exportedCount"], "H0 fail-closed：不得导出")
+        self.assertIn("隐藏结构不可得", " ".join(res["summary"]["unresolvedReasons"]))
+
+    def test_unresolvable_anchor_is_unresolved_not_missing(self):
+        import openpyxl  # noqa: F401
+        p = make_xlsx(self.case / "绝对锚点.xlsx",
+                      [{"kind": "absolute", "rid": "rId1", "row": 0, "media": "image1.png"}])
+        hidden = PREPARE._hidden_metadata(openpyxl.load_workbook(p),
+                                          openpyxl.load_workbook(p, data_only=False), path=str(p))
+        res = self._plan(p, ".xlsx", "zip-ooxml", hidden)
+        self.assertEqual(0, res["summary"]["exportedCount"])
+        self.assertIn("锚点不可判定", " ".join(res["summary"]["unresolvedReasons"]))
+
+    # ---- 3 docx 内联图 ---------------------------------------------------
+    def test_docx_inline_image_is_exported_with_paragraph_index(self):
+        p = make_docx(self.case / "说明.docx")
+        res = self._plan(p, ".docx", "zip-ooxml")
+        self.assertEqual(1, res["summary"]["exportedCount"], res["summary"])
+        entry = res["entries"][0]
+        self.assertEqual(3, entry["paragraphIndex"])
+        self.assertIn("第3段图片#1", entry["locator"])
+        self.assertIn("以下为询价截图", entry["locator"], "定位须带前文，便于人工核对")
+        self.assertTrue((self.case / entry["localPath"]).exists())
+
+    def test_docx_header_image_is_unresolved_not_exported(self):
+        p = make_docx(self.case / "带页眉.docx", header_image=True)
+        res = self._plan(p, ".docx", "zip-ooxml")
+        self.assertEqual(1, res["summary"]["exportedCount"], "正文图照常导出")
+        reasons = " ".join(res["summary"]["unresolvedReasons"])
+        self.assertIn("页眉页脚", reasons, "页眉页脚图按版式证据记未核，不导出")
+        blob = json.dumps(res, ensure_ascii=False)
+        self.assertNotIn("image2.png", blob)
+
+    # ---- 4 独立图片 ------------------------------------------------------
+    def test_standalone_image_is_copied_into_evidence_dir(self):
+        p = self.case / "现场照片.png"
+        p.write_bytes(PNG)
+        res = self._plan(p, ".png", "png")
+        self.assertEqual(1, res["summary"]["exportedCount"])
+        self.assertEqual("独立图片", res["entries"][0]["kind"])
+        self.assertEqual("现场照片.png（整图）", res["entries"][0]["locator"])
+        self.assertEqual(PNG, (self.case / res["entries"][0]["localPath"]).read_bytes())
+
+    # ---- 5 缺依赖 / 老格式：未核原因而非静默 -----------------------------
+    def test_pdf_without_pypdf_records_reason(self):
+        p = self.case / "扫描件.pdf"
+        p.write_bytes(b"%PDF-1.4\n%\xe2\xe3\xcf\xd3\n")
+        res = self._plan(p, ".pdf", "pdf")
+        try:
+            import pypdf  # noqa: F401
+            self.assertTrue(res["summary"]["unresolvedReasons"] or
+                            res["summary"]["exportedCount"] >= 0)
+        except ImportError:
+            self.assertEqual(0, res["summary"]["exportedCount"])
+            self.assertIn("pypdf", " ".join(res["summary"]["unresolvedReasons"]))
+
+    def test_legacy_xls_records_media_gap(self):
+        p = self.case / "老表.xls"
+        p.write_bytes(b"\xd0\xcf\x11\xe0\xa1\xb1\x1a\xe1dummy")
+        res = self._plan(p, ".xls", "ole")
+        self.assertEqual(0, res["summary"]["exportedCount"])
+        self.assertIn("未核", " ".join(res["summary"]["unresolvedReasons"]))
+
+    # ---- 6 上限护栏 ------------------------------------------------------
+    def test_budget_cap_stops_and_records(self):
+        p = make_xlsx(self.case / "超限.xlsx",
+                      [{"kind": "oneCell", "rid": "rId1", "row": 0, "media": "image1.png"},
+                       {"kind": "oneCell", "rid": "rId2", "row": 1, "media": "image2.png"}])
+        budget = MEDIA._Budget(items=1, nbytes=10 ** 6)
+        res = self._plan(p, ".xlsx", "zip-ooxml",
+                         {"hiddenSheets": [], "hiddenRows": {}, "hiddenCols": {}}, budget=budget)
+        self.assertEqual(1, res["summary"]["exportedCount"], "超限即停，不静默截断")
+        self.assertIn("超上限", " ".join(res["summary"]["unresolvedReasons"]))
+
+    # ---- 7 非媒体文件不产生噪声 ------------------------------------------
+    def test_plain_text_file_yields_no_media_summary(self):
+        p = self.case / "说明.txt"
+        p.write_text("正文", encoding="utf-8")
+        res = self._plan(p, ".txt", "other")
+        self.assertEqual([], res["entries"])
+        self.assertIsNone(res["summary"])
+
+
+if __name__ == "__main__":
+    unittest.main(verbosity=2)