sync update from LAN repo (source dhk-1 @ 3b1ec5b4)

This commit is contained in:
gozeros committed 2026-11-05 10:56:34 +08:00
1 parent da6521b6cb
commit 264e2200d6
995 files changed
+148105 -13399

No files matched your search

@@ -8,6 +8,7 @@ import re
import sys
import zipfile
from pathlib import Path
from types import SimpleNamespace
from xml.etree import ElementTree
import pytest
@@ -97,8 +98,11 @@ def test_formal_docx_contains_required_layout_numbering_and_embedded_font(tmp_pa
assert '<w:numFmt w:val="chineseCounting"/>' in numbering
assert '<w:lvlText w:val="%1、"/>' in numbering
assert '<w:lvlText w:val="(%2)"/>' in numbering
assert '<w:lvlText w:val="%3."/>' in numbering
assert '<w:lvl w:ilvl="1">' in numbering
assert '<w:lvlRestart w:val="1"/><w:isLgl/><w:lvlText w:val="%1.%2"/>' in numbering
assert '<w:lvlRestart w:val="2"/><w:isLgl/><w:lvlText w:val="%1.%2.%3"/>' in numbering
assert '<w:lvlRestart w:val="3"/><w:isLgl/><w:lvlText w:val="%1.%2.%3.%4"/>' in numbering
assert '<w:lvlRestart w:val="4"/><w:isLgl/><w:lvlText w:val="%1.%2.%3.%4.%5"/>' in numbering
assert numbering.count('w:firstLineChars="200"') == 3
assert '<w:evenAndOddHeaders/>' in settings
assert '<w:updateFields w:val="true"/>' in settings
@@ -175,3 +179,246 @@ def test_export_rejects_wrong_font(tmp_path: Path) -> None:
)
def test_manual_heading_number_is_removed_as_one_complete_prefix(source: str, expected: str) -> None:
assert word_export._strip_manual_heading_number(source) == expected
def test_all_markdown_heading_levels_map_to_word_heading_levels() -> None:
markdown = """# 文档主标题
## 一、一级标题
### 1.1 二级标题
#### 1.1.1 三级标题
##### 1.1.1.1 四级标题
###### 1.1.1.1.1 五级标题
"""
title, blocks = word_export._parse_markdown(markdown, None)
headings = [block for block in blocks if isinstance(block, word_export.ParagraphBlock)]
assert title == "文档主标题"
assert [(block.kind, block.text, block.level) for block in headings] == [
("heading", "一级标题", 0),
("heading", "二级标题", 1),
("heading", "三级标题", 2),
("heading", "四级标题", 3),
("heading", "五级标题", 4),
]
payload = word_export.build_markdown_docx(markdown)
with zipfile.ZipFile(io.BytesIO(payload)) as archive:
document = archive.read("word/document.xml").decode("utf-8")
mapped_levels = re.findall(
r'<w:pStyle w:val="(Heading\d)"/><w:numPr><w:ilvl w:val="(\d)"/>',
document,
)
assert mapped_levels == [
("Heading1", "0"),
("Heading2", "1"),
("Heading3", "2"),
("Heading4", "3"),
("Heading5", "4"),
]
def test_tolerates_model_generated_atx_heading_variations() -> None:
markdown = """ #模型生成的标题
##一级标题 ##
###**1.1 二级标题**
#### 1.1.1 三级标题
#####1.1.1.1 四级标题
######1.1.1.1.1 五级标题
"""
title, blocks = word_export._parse_markdown(markdown, None)
headings = [block for block in blocks if getattr(block, "kind", "") == "heading"]
assert title == "模型生成的标题"
assert [(block.text, block.level) for block in headings] == [
("一级标题", 0),
("二级标题", 1),
("三级标题", 2),
("四级标题", 3),
("五级标题", 4),
]
def test_number_prefix_corrects_wrong_hash_depth_and_recovers_missing_hashes() -> None:
markdown = """# 标题
### 一、一级标题
正文。
## 1.1 二级标题
正文。
**1.2 缺少井号的二级标题**
正文。
1.2.1 缺少井号的三级标题
正文。
(2)兼容旧格式四级标题
正文。
"""
_, blocks = word_export._parse_markdown(markdown, None)
headings = [block for block in blocks if getattr(block, "kind", "") == "heading"]
assert [(block.text, block.level) for block in headings] == [
("一级标题", 0),
("二级标题", 1),
("缺少井号的二级标题", 1),
("缺少井号的三级标题", 2),
("兼容旧格式四级标题", 3),
]
def test_hierarchical_heading_without_hashes_or_blank_line_is_recovered() -> None:
markdown = "正文没有空行。\n1.1 模型漏掉井号与空行\n下一段正文。"
_, blocks = word_export._parse_markdown(markdown, "标题")
headings = [block for block in blocks if getattr(block, "kind", "") == "heading"]
assert [(block.text, block.level) for block in headings] == [
("模型漏掉井号与空行", 1)
]
def test_heading_tolerance_does_not_consume_routes_lists_or_fenced_code() -> None:
markdown = """#/public/scheduled-tasks/123
1. 普通有序列表
```markdown
###1.1 代码块里的示例
```
"""
title, blocks = word_export._parse_markdown(markdown, "显式标题")
assert title == "显式标题"
assert not any(getattr(block, "kind", "") == "heading" for block in blocks)
assert any(getattr(block, "kind", "") == "ordered_list" for block in blocks)
assert any(getattr(block, "kind", "") == "code" and "###1.1" in block.text for block in blocks)
@pytest.mark.parametrize(
("style", "markdown", "expected"),
[
(
"hybrid_decimal",
"## 一、一级\n### 1.1 二级\n#### 1.1.1 三级",
[("一级", 0), ("二级", 1), ("三级", 2)],
),
(
"official_chinese",
"## 一、一级\n### (一)二级\n#### 1. 三级\n##### (1)四级\n###### 1)五级",
[("一级", 0), ("二级", 1), ("三级", 2), ("四级", 3), ("五级", 4)],
),
(
"decimal",
"## 1 一级\n### 1.1 二级\n#### 1.1.1 三级",
[("一级", 0), ("二级", 1), ("三级", 2)],
),
(
"chapter_section",
"## 第一章 一级\n### 第一节 二级\n#### 一、三级\n##### (一)四级\n###### 1. 五级",
[("一级", 0), ("二级", 1), ("三级", 2), ("四级", 3), ("五级", 4)],
),
(
"none",
"## 一级\n### 二级\n#### 三级",
[("一级", 0), ("二级", 1), ("三级", 2)],
),
],
)
def test_heading_prefixes_are_parsed_according_to_selected_preset(
style: str,
markdown: str,
expected: list[tuple[str, int]],
) -> None:
_, blocks = word_export._parse_markdown(
markdown,
None,
heading_numbering_style=style,
)
headings = [block for block in blocks if getattr(block, "kind", "") == "heading"]
assert [(block.text, block.level) for block in headings] == expected
@pytest.mark.parametrize(
("style", "expected_level_texts"),
[
("hybrid_decimal", ["%1、", "%1.%2", "%1.%2.%3", "%1.%2.%3.%4", "%1.%2.%3.%4.%5"]),
("official_chinese", ["%1、", "(%2)", "%3.", "(%4)", "%5)"]),
("decimal", ["%1", "%1.%2", "%1.%2.%3", "%1.%2.%3.%4", "%1.%2.%3.%4.%5"]),
("chapter_section", ["第%1章", "第%2节", "%3、", "(%4)", "%5."]),
],
)
def test_heading_numbering_presets_emit_expected_word_patterns(
style: str,
expected_level_texts: list[str],
) -> None:
numbering = word_export._numbering_xml(style)
for expected in expected_level_texts:
assert f'<w:lvlText w:val="{expected}"/>' in numbering
ElementTree.fromstring(numbering)
def test_no_numbering_preset_keeps_heading_styles_without_numpr() -> None:
payload = word_export.build_markdown_docx(
"# 文档标题\n\n## 一级标题\n\n### 二级标题",
heading_numbering_style="none",
)
with zipfile.ZipFile(io.BytesIO(payload)) as archive:
document = archive.read("word/document.xml").decode("utf-8")
assert '<w:pStyle w:val="Heading1"/>' in document
assert '<w:pStyle w:val="Heading2"/>' in document
assert "<w:numPr>" not in document
def test_unknown_heading_numbering_preset_is_rejected() -> None:
with pytest.raises(word_export.WordExportError, match="未知的 Word 标题编号格式"):
word_export.build_markdown_docx("正文", heading_numbering_style="unknown")
@pytest.mark.asyncio
async def test_export_route_uses_current_heading_numbering_setting(monkeypatch) -> None:
from app.gateway.routers import writing
captured: dict[str, object] = {}
def fake_build(markdown, title, *, embedded_images, heading_numbering_style):
captured.update(
markdown=markdown,
title=title,
embedded_images=embedded_images,
heading_numbering_style=heading_numbering_style,
)
return b"docx"
monkeypatch.setattr(
writing,
"load_system_settings",
lambda: SimpleNamespace(
prompt_prefix=SimpleNamespace(word_heading_numbering_style="official_chinese")
),
)
monkeypatch.setattr(writing, "build_markdown_docx", fake_build)
response = await writing.export_markdown_docx(
writing.MarkdownDocxExportRequest(markdown="# 标题", title="测试")
)
assert response.body == b"docx"
assert captured == {
"markdown": "# 标题",
"title": "测试",
"embedded_images": None,
"heading_numbering_style": "official_chinese",
}