fix(report): six parity chapters read real packet records and sit at ## (#91 fix2 F1/F2/F4)

Eleven sections printed one-column tables of 未判定 and packet key names
(summary, p5, stats ...). They now read the nested records: KP profile and
ruling planets, Tara/Chandra combined rows, lagna vargottamamsa flags, LL/7L
P5/P8/exchange checks, Yoga Pinda planets and summary, Sodhita BAV totals and
Ekadhipatya audit, friendship method and bodies, multi-reference houses,
double-transit PAC targets, congregation houses, seven/eight karaka tables.
Argala, Birth Calendar, Tithi, Tajika candidates, Goel dignity and yoga
categories no longer print bare lower-case keys or enums either.

The six chapters are now ## and their sections ###, so the reader counts
six more chapters instead of seventy.

Tests: a section passes only with a real value outside the label column; no
table cell may be a raw key; heading counts equal the base report plus six.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HwMNZEzdzmrbPhuRNmZe3C

tests/test_full_data_parity_chapters.py

@@ -32,19 +32,24 @@ _KAAL_EN = re.compile(r"\d+ of \d+")
 
 
 def _chapter_text(markdown: str, language: str) -> str:
-    marker = "# 大运附表" if language == "zh" else "# Dasha Appendix"
+    marker = "\n## 大运附表\n" if language == "zh" else "\n## Dasha Appendix\n"
     assert marker in markdown
     return markdown.split(marker, 1)[1]
 
 
 def _section_body(markdown: str, heading: str) -> str:
-    marker = f"## {heading}"
+    marker = f"### {heading}\n"
     start = markdown.index(marker) + len(marker)
     rest = markdown[start:]
-    ends = [pos for pos in (rest.find("\n## "), rest.find("\n# ")) if pos >= 0]
+    ends = [pos for pos in (rest.find("\n### "), rest.find("\n## "), rest.find("\n# ")) if pos >= 0]
     return rest[: min(ends)] if ends else rest
 
 
+def _heading_lines(markdown: str, level: int) -> list[str]:
+    prefix = "#" * level + " "
+    return [line[len(prefix):].strip() for line in markdown.splitlines() if line.startswith(prefix)]
+
+
 def test_chapter_titles_follow_the_task_groups() -> None:
     groups = _chapter_sections()
     assert len(groups) == len(CHAPTERS) == 6
@@ -73,22 +78,26 @@ def test_empty_packet_chapters_are_headed_and_do_not_print_raw_keys() -> None:
             heading = title_zh if language == "zh" else title_en
             body = _section_body(text, heading)
             assert body.strip(), heading
-            cursor_next = text.index(f"## {heading}", cursor)
+            cursor_next = text.index(f"### {heading}\n", cursor)
             assert cursor_next >= cursor
             cursor = cursor_next
             if "|---|" in body or "| --- |" in body:
                 lines = [line for line in body.splitlines() if line.startswith("|")]
                 assert len(lines) >= 3, heading
         for title in FORBIDDEN_TITLES:
+            assert f"### {title}" not in text
             assert f"## {title}" not in text
             assert f"# {title}" not in text
         assert _SNAKE.search(text) is None
+        assert _heading_lines(text, 1) == []
+        chapters = [chapter_zh if language == "zh" else chapter_en for chapter_en, chapter_zh in CHAPTERS]
+        assert _heading_lines(text, 2) == chapters
     zh = render_parity_chapters({}, "zh")
     en = render_parity_chapters({}, "en")
-    assert zh.index("### 来源感知口径") < zh.index("### 既有口径")
-    assert "来源感知口径" not in zh.split("### 既有口径", 1)[1]
-    assert en.index("### Source-aware method") < en.index("### Legacy method")
-    assert "Source-aware method" not in en.split("### Legacy method", 1)[1]
+    assert zh.index("#### 来源感知口径") < zh.index("#### 既有口径")
+    assert "来源感知口径" not in zh.split("#### 既有口径", 1)[1]
+    assert en.index("#### Source-aware method") < en.index("#### Legacy method")
+    assert "Source-aware method" not in en.split("#### Legacy method", 1)[1]
     assert "印刷页 10" in zh
     assert "printed pages 10-17" in en
     for banned in ("source_locator", "profile_id", "niryaana_shoola_strength", "source_discrepancy", "sign_idx"):
@@ -120,21 +129,21 @@ def test_three_fictional_charts_render_the_same_chapters(fictional_reports: dict
         zh_at = 0
         en_at = 0
         for title_en, title_zh in SECTIONS:
-            zh_at = zh.index(f"## {title_zh}", zh_at)
-            en_at = en.index(f"## {title_en}", en_at)
+            zh_at = zh.index(f"### {title_zh}\n", zh_at)
+            en_at = en.index(f"### {title_en}\n", en_at)
             for body, heading in ((_section_body(zh, title_zh), title_zh), (_section_body(en, title_en), title_en)):
                 assert body.strip(), f"{index} {heading}"
         for chapter_en, chapter_zh in CHAPTERS:
-            assert chinese.count(f"# {chapter_zh}") == 1
-            assert english.count(f"# {chapter_en}") == 1
+            assert _heading_lines(chinese, 2).count(chapter_zh) == 1
+            assert _heading_lines(english, 2).count(chapter_en) == 1
         for title in FORBIDDEN_TITLES:
             assert title not in zh
             assert title not in en
         assert _SNAKE.search(zh) is None
         assert _SNAKE.search(en) is None
         assert "印刷页 10" in zh and "printed pages 10-17" in en
-        assert "### 来源感知口径" in zh and "### 既有口径" in zh
-        assert "### Source-aware method" in en and "### Legacy method" in en
+        assert "#### 来源感知口径" in zh and "#### 既有口径" in zh
+        assert "#### Source-aware method" in en and "#### Legacy method" in en
         assert "再小运" in _section_body(zh, "Chara 大运时长")
         assert "Pratyantardasha" in _section_body(en, "Chara Dasha Period Durations")
         assert "marriage_windows" not in _section_body(zh, "Bhrigu 近似推进序列")
@@ -184,17 +193,40 @@ def _table_blocks(body: str) -> list[list[list[str]]]:
     return parsed
 
 
+_NOT_VALUES = {"未判定", "Not determined", "—", "-", ""}
+# A whole cell that looks like a packet key (`summary`, `p5`, `house_7`).
+_RAW_KEY_CELL = re.compile(r"^[a-z][a-z0-9]*(_[a-z0-9]+)*$")
+
+
 def _section_has_real_cell(body: str, undetermined: str) -> bool:
+    """A table must carry a value outside its first (label) column.
+
+    The first column holds row labels, so it never proves that the packet was
+    read. A one-column table therefore never passes. A body with no table at all
+    may be a plain sentence (for example "no boundary within 90 days").
+    """
     blocks = _table_blocks(body)
     if blocks:
         return any(
-            cell and cell != undetermined
+            cell not in _NOT_VALUES and cell != undetermined
             for rows in blocks
             for row in rows
-            for cell in row
+            for cell in row[1:]
         )
-    prose = [line.strip() for line in body.splitlines() if line.strip()]
-    return any(line and line != undetermined for line in prose)
+    prose = [line.strip() for line in body.splitlines() if line.strip() and not line.startswith("#")]
+    return any(line != undetermined for line in prose)
+
+
+def _all_cells(markdown: str) -> list[str]:
+    cells = []
+    for line in markdown.splitlines():
+        if not line.startswith("|"):
+            continue
+        parts = [cell.strip() for cell in line.strip().strip("|").split("|")]
+        if parts and all(set(cell) <= set("-: ") and cell for cell in parts):
+            continue
+        cells.extend(parts)
+    return cells
 
 
 def test_every_section_has_a_real_cell_except_the_four_public_gaps(fictional_reports: dict[int, tuple[str, str]]) -> None:
@@ -214,6 +246,55 @@ def test_every_section_has_a_real_cell_except_the_four_public_gaps(fictional_rep
                 assert _section_has_real_cell(body, _UNDETERMINED[language]), f"{index} {heading}"
 
 
+def test_every_section_has_a_real_cell_outside_the_label_column_unit() -> None:
+    one_column = "| 角色 |\n|---|\n| 未判定 |\n| summary |\n"
+    assert not _section_has_real_cell(one_column, "未判定")
+    labels_only = "| 项目 | 内容 |\n|---|---|\n| 状态 | 未判定 |\n| 宫位 | — |\n"
+    assert not _section_has_real_cell(labels_only, "未判定")
+    real = "| 项目 | 内容 |\n|---|---|\n| 宫位 | 7 |\n"
+    assert _section_has_real_cell(real, "未判定")
+    assert _section_has_real_cell("参考日起 90 天内没有大运或小运交界。", "未判定")
+    assert not _section_has_real_cell("未判定", "未判定")
+
+
+def test_no_table_cell_is_a_raw_packet_key(fictional_reports: dict[int, tuple[str, str]]) -> None:
+    for index, (chinese, english) in fictional_reports.items():
+        for language, markdown in (("zh", chinese), ("en", english)):
+            chapter = _chapter_text(markdown, language)
+            keys = sorted({cell for cell in _all_cells(chapter) if _RAW_KEY_CELL.fullmatch(cell)})
+            assert keys == [], f"{index} {language} {keys[:10]}"
+
+
+def test_six_chapters_are_reader_chapters_and_sections_sit_below_them(monkeypatch: pytest.MonkeyPatch) -> None:
+    """The reader splits on `## ` and ignores `# `. The six chapters must be `##`."""
+    import scripts.pl9_full_data_export as export
+
+    zh_chapters = [chapter_zh for _chapter_en, chapter_zh in CHAPTERS]
+    en_chapters = [chapter_en for chapter_en, _chapter_zh in CHAPTERS]
+    packet = _load_packet(0)
+    full_zh = render_full_data_markdown(copy.deepcopy(packet))
+    full_en = _english_edition(copy.deepcopy(packet), True).get("markdown_en") or ""
+    monkeypatch.setattr(export, "_append_parity_chapters", lambda markdown, _packet, _language: markdown)
+    base_zh = render_full_data_markdown(copy.deepcopy(packet))
+    base_en = _english_edition(copy.deepcopy(packet), True).get("markdown_en") or ""
+    assert base_en
+    for full, base, chapters in ((full_zh, base_zh, zh_chapters), (full_en, base_en, en_chapters)):
+        h1 = _heading_lines(full, 1)
+        assert len(h1) == 1, h1
+        assert full.startswith("# " + h1[0])
+        assert _heading_lines(base, 1) == h1
+        full_h2 = _heading_lines(full, 2)
+        base_h2 = _heading_lines(base, 2)
+        assert len(full_h2) == len(base_h2) + len(chapters)
+        assert full_h2[: len(base_h2)] == base_h2
+        assert full_h2[len(base_h2):] == chapters
+        tail = full.split("\n## " + chapters[0] + "\n", 1)[1]
+        assert _heading_lines(tail, 3)[: len(_chapter_sections()[0])] == [
+            (dict(SECTIONS)[title] if chapters is zh_chapters else title) for title in _chapter_sections()[0]
+        ]
+        assert len(_heading_lines(tail, 3)) == len(SECTIONS)
+
+
 def test_parity_chapters_go_through_local_conventions() -> None:
     from scripts.pl9_full_data_export import _apply_local_report_conventions
 
@@ -236,11 +317,13 @@ def test_parity_chapters_go_through_local_conventions() -> None:
 
     english_packet = copy.deepcopy(packet)
     english_packet["report_language"] = "en"
-    english_packet["worksheets"]["strengths_and_scores"]["vimsopaka"]["Sun"]["category"] = "divorce"
+    english_packet["worksheets"]["strengths_and_scores"]["vimsopaka"]["Sun"]["category"] = "Divorce risk"
+    # A bare lower-case word is now treated as an internal key and is not
+    # printed, so the probe is a reader-style phrase (fix2 F1).
     english = render_full_data_markdown(english_packet)
     en_body = _section_body(english, "Vimsopaka Contribution")
-    en_replaced = _apply_local_report_conventions("divorce", language="en")
-    assert en_replaced == "relationship strain"
+    en_replaced = _apply_local_report_conventions("Divorce risk", language="en")
+    assert en_replaced == "relationship strain risk"
     assert en_replaced in en_body
     assert "divorce" not in en_body.lower()