mdstyledocx 0.1.0__py3-none-any.whl

mdstyledocx/__init__.py

@@ -0,0 +1,5 @@
+"""mdstyledocx package."""
+
+__all__ = ["__version__"]
+
+__version__ = "0.1.0"

mdstyledocx/__main__.py

@@ -0,0 +1,5 @@
+from mdstyledocx.cli import main
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

mdstyledocx/cli.py

@@ -0,0 +1,74 @@
+from __future__ import annotations
+
+import argparse
+import sys
+from pathlib import Path
+from typing import Sequence
+
+from mdstyledocx.docx_writer import build_docx
+from mdstyledocx.markdown import parse_markdown
+from mdstyledocx.presets import list_presets, load_preset, load_preset_rules
+
+
+def main(argv: Sequence[str] | None = None) -> int:
+    parser = argparse.ArgumentParser(
+        prog="mdstyledocx",
+        description="Convert convention-based Markdown into standardized DOCX using reusable presets.",
+    )
+    parser.add_argument("input", nargs="?", help="Input Markdown file path, or '-' to read from stdin.")
+    parser.add_argument("-o", "--output", help="Output DOCX file path.")
+    parser.add_argument(
+        "--preset",
+        default="default",
+        help="Built-in preset name. Use --list-presets to inspect available values.",
+    )
+    parser.add_argument(
+        "--preset-file",
+        type=Path,
+        help="Optional JSON override file to extend or replace parts of a preset.",
+    )
+    parser.add_argument(
+        "--list-presets",
+        action="store_true",
+        help="Print built-in presets and exit.",
+    )
+    parser.add_argument(
+        "--show-preset-rules",
+        metavar="PRESET",
+        help="Print the Markdown conventions and style notes for a built-in preset, then exit.",
+    )
+
+    args = parser.parse_args(list(argv) if argv is not None else None)
+
+    if args.list_presets:
+        for name, description in list_presets():
+            print(f"{name}: {description}")
+        return 0
+
+    if args.show_preset_rules:
+        print(load_preset_rules(args.show_preset_rules).rstrip())
+        return 0
+
+    if not args.input:
+        parser.error("an input Markdown file is required unless --list-presets is used")
+
+    if args.input == "-":
+        markdown_text = sys.stdin.read()
+        if not args.output:
+            parser.error("--output is required when reading Markdown from stdin")
+        output_path = Path(args.output)
+    else:
+        input_path = Path(args.input)
+        markdown_text = input_path.read_text(encoding="utf-8")
+        output_path = Path(args.output) if args.output else input_path.with_suffix(".docx")
+
+    preset = load_preset(args.preset, args.preset_file)
+    base_path = Path.cwd() if args.input == "-" else input_path.parent
+    document = parse_markdown(markdown_text, base_path=base_path)
+    output_path.write_bytes(build_docx(document, preset))
+    print(f"Wrote {output_path} with preset '{preset.name}'")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

mdstyledocx/docx_writer.py

@@ -0,0 +1,253 @@
+from __future__ import annotations
+
+import os
+import re
+from dataclasses import dataclass, field, replace
+from datetime import datetime, timezone
+from io import BytesIO
+
+from docx import Document as WordDocument
+from docx.document import Document as WordprocessingDocument
+from docx.enum.text import WD_ALIGN_PARAGRAPH, WD_LINE_SPACING
+from docx.oxml.ns import qn
+from docx.shared import Pt, Twips
+
+from mdstyledocx.model import Block, Document, ImageSpan, InlineElement, InlineSpan
+from mdstyledocx.presets import Preset, Style
+
+
+@dataclass
+class BuildState:
+    preset: Preset
+    heading_counters: dict[int, int] = field(default_factory=dict)
+
+
+def build_docx(document: Document, preset: Preset) -> bytes:
+    word_document = WordDocument()
+    _configure_document(word_document, preset)
+    _set_core_properties(word_document, _document_title(document))
+
+    state = BuildState(preset=preset)
+    for block in document.blocks:
+        _append_block(word_document, block, state)
+
+    buffer = BytesIO()
+    word_document.save(buffer)
+    return buffer.getvalue()
+
+
+def _configure_document(word_document: WordprocessingDocument, preset: Preset) -> None:
+    section = word_document.sections[0]
+    section.page_width = Twips(preset.page.width)
+    section.page_height = Twips(preset.page.height)
+    section.top_margin = Twips(preset.page.margin_top)
+    section.right_margin = Twips(preset.page.margin_right)
+    section.bottom_margin = Twips(preset.page.margin_bottom)
+    section.left_margin = Twips(preset.page.margin_left)
+    section.header_distance = Twips(preset.page.header)
+    section.footer_distance = Twips(preset.page.footer)
+    section.gutter = Twips(preset.page.gutter)
+
+
+def _set_core_properties(word_document: WordprocessingDocument, title: str) -> None:
+    properties = word_document.core_properties
+    properties.title = title
+    properties.author = "mdstyledocx"
+    properties.last_modified_by = "mdstyledocx"
+    now = datetime.now(timezone.utc).replace(microsecond=0, tzinfo=None)
+    properties.created = now
+    properties.modified = now
+
+
+def _document_title(document: Document) -> str:
+    if document.metadata.get("title"):
+        return document.metadata["title"]
+    for block in document.blocks:
+        if block.kind == "heading" and block.level == 1:
+            return "".join(span.text for span in block.spans if isinstance(span, InlineSpan)).strip()
+    return "Document"
+
+
+def _append_block(word_document: WordprocessingDocument, block: Block, state: BuildState) -> None:
+    if block.kind == "page_break":
+        word_document.add_page_break()
+        return
+
+    rendered_spans = _rendered_spans(block, state)
+    style = _resolve_style(block, state.preset)
+    if _spans_have_image(rendered_spans):
+        style = replace(style, line=240, line_rule="auto")
+
+    paragraph = word_document.add_paragraph()
+    _apply_paragraph_style(paragraph, style)
+
+    if block.kind == "list_item":
+        prefix = "• " if block.list_kind == "bullet" else f"{block.number}. "
+        _add_text_run(paragraph, InlineSpan(text=prefix), style)
+
+    for span in rendered_spans:
+        if isinstance(span, ImageSpan):
+            _add_image_run(paragraph, span, state)
+        elif span.text:
+            _add_text_run(paragraph, span, style)
+
+    if not paragraph.runs:
+        paragraph.add_run("")
+
+
+def _resolve_style(block: Block, preset: Preset) -> Style:
+    if block.kind == "heading":
+        key = {1: "title", 2: "heading1", 3: "heading2"}.get(block.level, "heading3")
+        return preset.styles[key]
+
+    base = preset.styles["body"]
+    if block.kind == "list_item":
+        left_indent = (
+            preset.list_settings.base_left_indent
+            + block.list_level * preset.list_settings.level_step
+        )
+        return replace(
+            base,
+            first_line_indent=0,
+            left_indent=left_indent,
+            hanging=preset.list_settings.hanging,
+        )
+
+    return base
+
+
+def _apply_paragraph_style(paragraph, style: Style) -> None:
+    alignment_map = {
+        "left": WD_ALIGN_PARAGRAPH.LEFT,
+        "center": WD_ALIGN_PARAGRAPH.CENTER,
+        "right": WD_ALIGN_PARAGRAPH.RIGHT,
+        "justify": WD_ALIGN_PARAGRAPH.JUSTIFY,
+        "both": WD_ALIGN_PARAGRAPH.JUSTIFY,
+    }
+    if style.align:
+        paragraph.alignment = alignment_map[style.align]
+
+    paragraph_format = paragraph.paragraph_format
+    paragraph_format.space_before = Twips(style.spacing_before)
+    paragraph_format.space_after = Twips(style.spacing_after)
+    paragraph_format.left_indent = Twips(style.left_indent)
+
+    if style.hanging:
+        paragraph_format.first_line_indent = Twips(-style.hanging)
+    else:
+        paragraph_format.first_line_indent = Twips(style.first_line_indent)
+
+    if style.line_rule == "exact":
+        paragraph_format.line_spacing = Pt(style.line / 20)
+        paragraph_format.line_spacing_rule = WD_LINE_SPACING.EXACTLY
+    elif style.line_rule == "auto":
+        paragraph_format.line_spacing = style.line / 240
+    else:
+        paragraph_format.line_spacing = style.line / 240
+
+
+def _add_text_run(paragraph, span: InlineSpan, style: Style) -> None:
+    run = paragraph.add_run(span.text)
+    font_ascii = style.font_ascii
+    font_east_asia = style.font_east_asia
+    bold = style.bold or span.bold
+    italic = style.italic or span.italic
+
+    if span.code:
+        font_ascii = "Consolas"
+        font_east_asia = "等线"
+
+    run.font.name = font_ascii
+    run.font.size = Pt(style.size_half_points / 2)
+    run.bold = bold
+    run.italic = italic
+
+    r_fonts = run._element.get_or_add_rPr().get_or_add_rFonts()
+    r_fonts.set(qn("w:ascii"), font_ascii)
+    r_fonts.set(qn("w:hAnsi"), font_ascii)
+    r_fonts.set(qn("w:eastAsia"), font_east_asia)
+
+
+def _add_image_run(paragraph, span: ImageSpan, state: BuildState) -> None:
+    run = paragraph.add_run()
+    inline_shape = run.add_picture(span.path)
+    max_width = _max_image_width(state.preset)
+    if inline_shape.width > max_width:
+        scale = max_width / inline_shape.width
+        inline_shape.width = max_width
+        inline_shape.height = int(inline_shape.height * scale)
+
+    description = span.alt_text or os.path.basename(span.path)
+    inline_shape._inline.docPr.set("descr", description)
+    inline_shape._inline.docPr.set("name", description)
+
+
+def _max_image_width(preset: Preset) -> int:
+    return Twips(preset.page.width - preset.page.margin_left - preset.page.margin_right)
+
+
+def _spans_have_image(spans: list[InlineElement]) -> bool:
+    return any(isinstance(span, ImageSpan) for span in spans)
+
+
+def _rendered_spans(block: Block, state: BuildState) -> list[InlineElement]:
+    if block.kind != "heading":
+        return block.spans
+
+    scheme = state.preset.heading_numbering.get(block.level)
+    if not scheme:
+        return block.spans
+
+    _advance_heading_counters(state, block.level)
+    if _has_number_prefix(block.spans, scheme):
+        return block.spans
+
+    prefix = _format_heading_prefix(scheme, state.heading_counters[block.level])
+    return [InlineSpan(text=prefix)] + block.spans
+
+
+def _advance_heading_counters(state: BuildState, level: int) -> None:
+    state.heading_counters[level] = state.heading_counters.get(level, 0) + 1
+    for deeper_level in list(state.heading_counters):
+        if deeper_level > level:
+            state.heading_counters[deeper_level] = 0
+
+
+def _has_number_prefix(spans: list[InlineElement], scheme: str) -> bool:
+    text = "".join(span.text for span in spans if isinstance(span, InlineSpan)).lstrip()
+    patterns = {
+        "cn-section": r"^[一二三四五六七八九十百千万零〇两]+、",
+        "cn-paren": r"^（[一二三四五六七八九十百千万零〇两]+）",
+        "arabic-dot": r"^\d+[.．]\s*",
+    }
+    return re.match(patterns[scheme], text) is not None
+
+
+def _format_heading_prefix(scheme: str, number: int) -> str:
+    if scheme == "cn-section":
+        return f"{_to_chinese_number(number)}、"
+    if scheme == "cn-paren":
+        return f"（{_to_chinese_number(number)}）"
+    if scheme == "arabic-dot":
+        return f"{number}. "
+    raise ValueError(f"Unsupported heading numbering scheme: {scheme}")
+
+
+def _to_chinese_number(number: int) -> str:
+    digits = "零一二三四五六七八九"
+    units = ["", "十", "百", "千"]
+    raw = str(number)
+    parts: list[str] = []
+
+    for index, char in enumerate(raw):
+        digit = int(char)
+        unit_index = len(raw) - index - 1
+        if digit == 0:
+            if parts and parts[-1] != "零" and any(next_char != "0" for next_char in raw[index + 1 :]):
+                parts.append("零")
+            continue
+        if not (digit == 1 and unit_index == 1 and not parts and len(raw) == 2):
+            parts.append(digits[digit])
+        parts.append(units[unit_index])
+
+    return "".join(parts) or "零"

mdstyledocx/markdown.py

@@ -0,0 +1,187 @@
+from __future__ import annotations
+
+import re
+from pathlib import Path
+
+from mdstyledocx.model import Block, Document, ImageSpan, InlineElement, InlineSpan
+
+HEADING_RE = re.compile(r"^(#{1,6})\s+(.*?)\s*$")
+BULLET_RE = re.compile(r"^(\s*)[-*+]\s+(.*?)\s*$")
+ORDERED_RE = re.compile(r"^(\s*)(\d+)\.\s+(.*?)\s*$")
+INLINE_TOKEN_RE = re.compile(r"(!\[[^\]]*]\([^)]+\)|\*\*[^*]+\*\*|\*[^*]+\*|`[^`]+`)")
+IMAGE_RE = re.compile(r"^!\[([^\]]*)\]\(([^)]+)\)$")
+PAGEBREAK_MARKERS = {"<!-- pagebreak -->", "<!--pagebreak-->", "\f", "\\f"}
+
+
+def parse_markdown(text: str, base_path: Path | None = None) -> Document:
+    normalized = text.replace("\r\n", "\n").replace("\r", "\n").lstrip("\ufeff")
+    lines = normalized.split("\n")
+    metadata, body_lines = _parse_frontmatter(lines)
+    blocks = _parse_blocks(body_lines, base_path)
+
+    has_title = any(block.kind == "heading" and block.level == 1 for block in blocks)
+    if metadata.get("title") and not has_title:
+        blocks.insert(0, Block(kind="heading", level=1, spans=parse_inline(metadata["title"], base_path)))
+
+    return Document(metadata=metadata, blocks=blocks)
+
+
+def parse_inline(text: str, base_path: Path | None = None) -> list[InlineElement]:
+    spans: list[InlineElement] = []
+    for token in INLINE_TOKEN_RE.split(text):
+        if not token:
+            continue
+        image_match = IMAGE_RE.match(token)
+        if image_match:
+            spans.append(
+                ImageSpan(
+                    path=_resolve_asset_path(image_match.group(2), base_path),
+                    alt_text=image_match.group(1),
+                )
+            )
+        elif token.startswith("**") and token.endswith("**") and len(token) > 4:
+            spans.append(InlineSpan(text=token[2:-2], bold=True))
+        elif token.startswith("*") and token.endswith("*") and len(token) > 2:
+            spans.append(InlineSpan(text=token[1:-1], italic=True))
+        elif token.startswith("`") and token.endswith("`") and len(token) > 2:
+            spans.append(InlineSpan(text=token[1:-1], code=True))
+        else:
+            spans.append(InlineSpan(text=token))
+    return spans or [InlineSpan(text="")]
+
+
+def _parse_frontmatter(lines: list[str]) -> tuple[dict[str, str], list[str]]:
+    if not lines or lines[0].strip() != "---":
+        return {}, lines
+
+    metadata: dict[str, str] = {}
+    for index in range(1, len(lines)):
+        current = lines[index].strip()
+        if current == "---":
+            return metadata, lines[index + 1 :]
+        if ":" not in lines[index]:
+            return {}, lines
+        key, value = lines[index].split(":", 1)
+        metadata[key.strip()] = value.strip()
+
+    return {}, lines
+
+
+def _parse_blocks(lines: list[str], base_path: Path | None) -> list[Block]:
+    blocks: list[Block] = []
+    index = 0
+
+    while index < len(lines):
+        raw = lines[index]
+        stripped = raw.strip()
+
+        if not stripped:
+            index += 1
+            continue
+
+        if stripped in PAGEBREAK_MARKERS:
+            blocks.append(Block(kind="page_break"))
+            index += 1
+            continue
+
+        heading_match = HEADING_RE.match(raw)
+        if heading_match:
+            blocks.append(
+                Block(
+                    kind="heading",
+                    level=len(heading_match.group(1)),
+                    spans=parse_inline(heading_match.group(2), base_path),
+                )
+            )
+            index += 1
+            continue
+
+        bullet_match = BULLET_RE.match(raw)
+        if bullet_match:
+            blocks.append(
+                Block(
+                    kind="list_item",
+                    list_kind="bullet",
+                    list_level=len(bullet_match.group(1).replace("\t", "    ")) // 2,
+                    spans=parse_inline(bullet_match.group(2), base_path),
+                )
+            )
+            index += 1
+            continue
+
+        ordered_match = ORDERED_RE.match(raw)
+        if ordered_match:
+            blocks.append(
+                Block(
+                    kind="list_item",
+                    list_kind="ordered",
+                    list_level=len(ordered_match.group(1).replace("\t", "    ")) // 2,
+                    number=int(ordered_match.group(2)),
+                    spans=parse_inline(ordered_match.group(3), base_path),
+                )
+            )
+            index += 1
+            continue
+
+        paragraph_lines: list[str] = []
+        while index < len(lines) and not _starts_new_block(lines[index]):
+            paragraph_lines.append(_strip_blockquote(lines[index].rstrip()))
+            index += 1
+
+        blocks.append(
+            Block(kind="paragraph", spans=parse_inline(_join_paragraph_lines(paragraph_lines), base_path))
+        )
+
+    return blocks
+
+
+def _starts_new_block(line: str) -> bool:
+    stripped = line.strip()
+    if not stripped:
+        return True
+    if stripped in PAGEBREAK_MARKERS:
+        return True
+    if HEADING_RE.match(line):
+        return True
+    if BULLET_RE.match(line):
+        return True
+    if ORDERED_RE.match(line):
+        return True
+    return False
+
+
+def _strip_blockquote(line: str) -> str:
+    stripped = line.lstrip()
+    if stripped.startswith(">"):
+        return stripped[1:].lstrip()
+    return line.strip()
+
+
+def _join_paragraph_lines(lines: list[str]) -> str:
+    if not lines:
+        return ""
+
+    result = lines[0].strip()
+    for line in lines[1:]:
+        candidate = line.strip()
+        if not candidate:
+            continue
+        if _needs_space(result[-1], candidate[0]):
+            result += " " + candidate
+        else:
+            result += candidate
+    return result
+
+
+def _needs_space(previous_char: str, next_char: str) -> bool:
+    return previous_char.isascii() and next_char.isascii() and (
+        previous_char.isalnum() or previous_char in {")", "]"}
+    ) and (next_char.isalnum() or next_char in {"(", "["})
+
+
+def _resolve_asset_path(raw_path: str, base_path: Path | None) -> str:
+    candidate = raw_path.strip().strip("<>").strip()
+    path = Path(candidate)
+    if path.is_absolute() or base_path is None:
+        return str(path)
+    return str((base_path / path).resolve())

mdstyledocx/model.py

@@ -0,0 +1,36 @@
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+
+
+@dataclass
+class InlineSpan:
+    text: str
+    bold: bool = False
+    italic: bool = False
+    code: bool = False
+
+
+@dataclass
+class ImageSpan:
+    path: str
+    alt_text: str = ""
+
+
+InlineElement = InlineSpan | ImageSpan
+
+
+@dataclass
+class Block:
+    kind: str
+    spans: list[InlineElement] = field(default_factory=list)
+    level: int = 0
+    list_kind: str | None = None
+    list_level: int = 0
+    number: int | None = None
+
+
+@dataclass
+class Document:
+    metadata: dict[str, str] = field(default_factory=dict)
+    blocks: list[Block] = field(default_factory=list)

mdstyledocx/preset_specs/default.json

@@ -0,0 +1,63 @@
+{
+  "name": "default",
+  "description": "General-purpose clean document preset",
+  "page": {
+    "width": 11907,
+    "height": 16838,
+    "margin_top": 1440,
+    "margin_right": 1440,
+    "margin_bottom": 1440,
+    "margin_left": 1440
+  },
+  "styles": {
+    "title": {
+      "font_east_asia": "等线",
+      "font_ascii": "Calibri",
+      "size_half_points": 32,
+      "bold": true,
+      "align": "center",
+      "spacing_after": 240,
+      "line": 240
+    },
+    "heading1": {
+      "font_east_asia": "等线",
+      "font_ascii": "Calibri",
+      "size_half_points": 28,
+      "bold": true,
+      "spacing_before": 120,
+      "spacing_after": 120,
+      "line": 276
+    },
+    "heading2": {
+      "font_east_asia": "等线",
+      "font_ascii": "Calibri",
+      "size_half_points": 24,
+      "bold": true,
+      "spacing_before": 120,
+      "spacing_after": 80,
+      "line": 276
+    },
+    "heading3": {
+      "font_east_asia": "宋体",
+      "font_ascii": "Calibri",
+      "size_half_points": 22,
+      "bold": true,
+      "spacing_before": 80,
+      "spacing_after": 60,
+      "line": 276
+    },
+    "body": {
+      "font_east_asia": "宋体",
+      "font_ascii": "Calibri",
+      "size_half_points": 22,
+      "line": 276,
+      "line_rule": "auto",
+      "spacing_after": 120
+    }
+  },
+  "list_settings": {
+    "base_left_indent": 720,
+    "hanging": 360,
+    "level_step": 360
+  }
+}

mdstyledocx/preset_specs/default.md

@@ -0,0 +1,35 @@
+# `default` preset
+
+## 适用场景
+
+通用正式文档，不强调行业模板约束，适合先把内容稳定导出成干净的 Word 版式。
+
+## 样式说明
+
+- 页面大小：A4
+- 标题：居中、等线、较大字号
+- 正文：宋体，常规正式文档风格
+- 一级到三级标题：按层级逐步缩小字号
+
+## Markdown 约定
+
+- `#`：文档主标题
+- `##`：一级标题
+- `###`：二级标题
+- 普通段落：空行分段
+- `1.`：有序列表
+- `-` / `*` / `+`：无序列表
+- `<!-- pagebreak -->`：分页
+
+## 示例
+
+```md
+# 项目阶段性汇报
+
+本周工作进展如下。
+
+## 一、已完成事项
+
+1. 完成需求梳理。
+2. 完成文档输出规范。
+```

mdstyledocx/preset_specs/gov-cn-hei.json

@@ -0,0 +1,69 @@
+{
+  "name": "gov-cn-hei",
+  "description": "Chinese government document preset with Hei title, Kai headings, and FangSong body",
+  "page": {
+    "width": 11907,
+    "height": 16838,
+    "margin_top": 1440,
+    "margin_right": 1800,
+    "margin_bottom": 1440,
+    "margin_left": 1800
+  },
+  "styles": {
+    "title": {
+      "font_east_asia": "黑体",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 44,
+      "align": "center",
+      "spacing_after": 240,
+      "line": 560,
+      "line_rule": "exact"
+    },
+    "heading1": {
+      "font_east_asia": "黑体",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 32,
+      "spacing_before": 80,
+      "spacing_after": 80,
+      "line": 560,
+      "line_rule": "exact"
+    },
+    "heading2": {
+      "font_east_asia": "楷体",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 32,
+      "spacing_before": 60,
+      "spacing_after": 60,
+      "line": 560,
+      "line_rule": "exact"
+    },
+    "heading3": {
+      "font_east_asia": "仿宋",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 32,
+      "bold": true,
+      "spacing_before": 40,
+      "spacing_after": 40,
+      "line": 560,
+      "line_rule": "exact"
+    },
+    "body": {
+      "font_east_asia": "仿宋",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 32,
+      "first_line_indent": 640,
+      "line": 560,
+      "line_rule": "exact"
+    }
+  },
+  "list_settings": {
+    "base_left_indent": 720,
+    "hanging": 360,
+    "level_step": 360
+  },
+  "heading_numbering": {
+    "2": "cn-section",
+    "3": "cn-paren",
+    "4": "arabic-dot"
+  }
+}

mdstyledocx/preset_specs/gov-cn-hei.md

@@ -0,0 +1,57 @@
+# `gov-cn-hei` preset
+
+## 适用场景
+
+中文政府公文、正式通知、情况报告、请示类材料的另一套常见字体方案。
+
+## 当前版式基线
+
+- 页面大小：A4
+- 页边距：Word 默认“常规”（上/下 `2.54 cm`，左/右 `3.18 cm`）
+- 大标题：`黑体`
+- 一级标题：`黑体`
+- 二级标题：`楷体`
+- 正文：`仿宋`
+- 正文字号：三号
+- 正文行距：固定值 `28 磅`
+- 正文首行：缩进两字符
+- 三级标题：默认使用 `仿宋` 加粗
+- 一级、二级、三级标题：自动编号
+
+## 推荐 Markdown 写法
+
+- `#`：公文标题
+- 标题下方直接写主送机关或称谓段落
+- `##`：一级标题，导出为 `一、XXXX`
+- `###`：二级标题，导出为 `（一）XXXX`
+- `####`：三级标题，导出为 `1. XXXX`
+- 普通段落：正文内容，每段空一行
+- `![说明](./image.png)`：插入本地图片，按嵌入式输出；图片所在段落改为单倍行距
+- `<!-- pagebreak -->`：需要分页时显式插入
+
+编号由导出器自动补齐，Markdown 里只写标题内容，不要手写 `一、`、`（一）`、`1.`。
+
+## 与 `gov-cn` 的区别
+
+- 不使用 `方正小标宋简体`
+- 不使用 `楷体_GB2312`
+- 不使用 `仿宋_GB2312`
+- 更适合系统内仅提供通用 `黑体 / 楷体 / 仿宋` 字体的环境
+
+## 示例
+
+```md
+# 关于开展示例工作的通知
+
+各有关单位：
+
+为统一输出格式，现将有关事项通知如下。
+
+## 工作目标
+
+### 总体要求
+
+#### 任务分工
+
+请各单位提高认识，认真落实。
+```

mdstyledocx/preset_specs/gov-cn.json

@@ -0,0 +1,69 @@
+{
+  "name": "gov-cn",
+  "description": "Chinese government document baseline preset",
+  "page": {
+    "width": 11907,
+    "height": 16838,
+    "margin_top": 1440,
+    "margin_right": 1800,
+    "margin_bottom": 1440,
+    "margin_left": 1800
+  },
+  "styles": {
+    "title": {
+      "font_east_asia": "方正小标宋简体",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 44,
+      "align": "center",
+      "spacing_after": 240,
+      "line": 560,
+      "line_rule": "exact"
+    },
+    "heading1": {
+      "font_east_asia": "黑体",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 32,
+      "spacing_before": 80,
+      "spacing_after": 80,
+      "line": 560,
+      "line_rule": "exact"
+    },
+    "heading2": {
+      "font_east_asia": "楷体_GB2312",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 32,
+      "spacing_before": 60,
+      "spacing_after": 60,
+      "line": 560,
+      "line_rule": "exact"
+    },
+    "heading3": {
+      "font_east_asia": "仿宋_GB2312",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 32,
+      "bold": true,
+      "spacing_before": 40,
+      "spacing_after": 40,
+      "line": 560,
+      "line_rule": "exact"
+    },
+    "body": {
+      "font_east_asia": "仿宋_GB2312",
+      "font_ascii": "Times New Roman",
+      "size_half_points": 32,
+      "first_line_indent": 640,
+      "line": 560,
+      "line_rule": "exact"
+    }
+  },
+  "list_settings": {
+    "base_left_indent": 720,
+    "hanging": 360,
+    "level_step": 360
+  },
+  "heading_numbering": {
+    "2": "cn-section",
+    "3": "cn-paren",
+    "4": "arabic-dot"
+  }
+}

mdstyledocx/preset_specs/gov-cn.md

@@ -0,0 +1,60 @@
+# `gov-cn` preset
+
+## 适用场景
+
+中文政府公文、正式通知、情况报告、请示类材料的基线样式。
+
+## 当前版式基线
+
+- 页面大小：A4
+- 页边距：Word 默认“常规”（上/下 `2.54 cm`，左/右 `3.18 cm`）
+- 正文：`仿宋_GB2312`
+- 正文字号：三号
+- 正文行距：固定值 `28 磅`
+- 正文首行：缩进两字符
+- 标题：居中，默认使用 `方正小标宋简体`
+- 一级标题：默认使用黑体
+- 二级标题：默认使用 `楷体_GB2312`
+- 三级标题：默认使用 `仿宋_GB2312` 加粗
+- 一级、二级、三级标题：自动编号
+
+## 推荐 Markdown 写法
+
+- `#`：公文标题
+- 标题下方直接写主送机关或称谓段落
+- `##`：一级标题，导出为 `一、XXXX`
+- `###`：二级标题，导出为 `（一）XXXX`
+- `####`：三级标题，导出为 `1. XXXX`
+- 普通段落：正文内容，每段空一行
+- `![说明](./image.png)`：插入本地图片，按嵌入式输出；图片所在段落改为单倍行距
+- `<!-- pagebreak -->`：需要分页时显式插入
+
+编号由导出器自动补齐，Markdown 里只写标题内容，不要手写 `一、`、`（一）`、`1.`。
+
+## 示例
+
+```md
+# 关于开展示例工作的通知
+
+各有关单位：
+
+为统一输出格式，现将有关事项通知如下。
+
+## 工作目标
+
+### 总体要求
+
+#### 任务分工
+
+请各单位提高认识，认真落实。
+```
+
+## 当前边界
+
+当前只是“公文样式基线”，还不是完整公文排版规范。以下规则后续可以继续补：
+
+- 发文字号
+- 签发人
+- 主送机关的专门样式
+- 成文日期与落款位置
+- 附件说明和抄送规则

mdstyledocx/presets.py

@@ -0,0 +1,128 @@
+from __future__ import annotations
+
+import json
+from copy import deepcopy
+from dataclasses import dataclass, field
+from functools import lru_cache
+from importlib.resources import files
+from pathlib import Path
+from typing import Any
+
+
+@dataclass
+class PageSettings:
+    width: int
+    height: int
+    margin_top: int
+    margin_right: int
+    margin_bottom: int
+    margin_left: int
+    header: int = 708
+    footer: int = 708
+    gutter: int = 0
+
+
+@dataclass
+class Style:
+    font_east_asia: str
+    font_ascii: str
+    size_half_points: int
+    bold: bool = False
+    italic: bool = False
+    align: str = "left"
+    first_line_indent: int = 0
+    left_indent: int = 0
+    hanging: int = 0
+    spacing_before: int = 0
+    spacing_after: int = 0
+    line: int = 240
+    line_rule: str = "auto"
+
+
+@dataclass
+class ListSettings:
+    base_left_indent: int = 720
+    hanging: int = 360
+    level_step: int = 360
+
+
+@dataclass
+class Preset:
+    name: str
+    description: str
+    page: PageSettings
+    styles: dict[str, Style]
+    list_settings: ListSettings
+    heading_numbering: dict[int, str] = field(default_factory=dict)
+
+
+def list_presets() -> list[tuple[str, str]]:
+    return [
+        (name, definition["description"])
+        for name, definition in sorted(_builtin_definitions().items(), key=lambda item: item[0])
+    ]
+
+
+def load_preset(name: str, preset_file: Path | None = None) -> Preset:
+    if preset_file:
+        override_data = json.loads(preset_file.read_text(encoding="utf-8"))
+        base_name = override_data.pop("extends", name)
+        base = _base_definition(base_name)
+        merged = _deep_merge(base, override_data)
+        return _build_preset(merged)
+
+    return _build_preset(_base_definition(name))
+
+
+def load_preset_rules(name: str) -> str:
+    rules_path = _preset_specs_dir().joinpath(f"{name}.md")
+    if not rules_path.is_file():
+        available = ", ".join(sorted(_builtin_definitions()))
+        raise ValueError(f"Unknown preset '{name}'. Available presets: {available}")
+    return rules_path.read_text(encoding="utf-8")
+
+
+def _base_definition(name: str) -> dict[str, Any]:
+    definitions = _builtin_definitions()
+    if name not in definitions:
+        available = ", ".join(sorted(definitions))
+        raise ValueError(f"Unknown preset '{name}'. Available presets: {available}")
+    return deepcopy(definitions[name])
+
+
+@lru_cache(maxsize=1)
+def _builtin_definitions() -> dict[str, dict[str, Any]]:
+    definitions: dict[str, dict[str, Any]] = {}
+    for definition_path in _preset_specs_dir().iterdir():
+        if definition_path.suffix != ".json":
+            continue
+        raw = json.loads(definition_path.read_text(encoding="utf-8"))
+        definitions[raw["name"]] = raw
+    return definitions
+
+
+def _preset_specs_dir():
+    return files("mdstyledocx").joinpath("preset_specs")
+
+
+def _build_preset(data: dict[str, Any]) -> Preset:
+    return Preset(
+        name=data["name"],
+        description=data["description"],
+        page=PageSettings(**data["page"]),
+        styles={name: Style(**style) for name, style in data["styles"].items()},
+        list_settings=ListSettings(**data["list_settings"]),
+        heading_numbering={
+            int(level): scheme for level, scheme in data.get("heading_numbering", {}).items()
+        },
+    )
+
+
+def _deep_merge(base: dict[str, Any], override: dict[str, Any]) -> dict[str, Any]:
+    merged = deepcopy(base)
+    for key, value in override.items():
+        if isinstance(value, dict) and isinstance(merged.get(key), dict):
+            merged[key] = _deep_merge(merged[key], value)
+        else:
+            merged[key] = value
+    return merged

mdstyledocx-0.1.0.dist-info/METADATA

@@ -0,0 +1,147 @@
+Metadata-Version: 2.4
+Name: mdstyledocx
+Version: 0.1.0
+Summary: Convert convention-based Markdown into standardized DOCX with reusable style presets.
+Author: YANG Zijie
+License: MIT
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: python-docx>=1.2.0
+Dynamic: license-file
+
+# mdstyledocx
+
+一个按约定编写 Markdown、再一键导出标准化 Word (`.docx`) 的小工具。
+
+当前设计重点不是“完整支持所有 Markdown 语法”，而是“稳定地把结构化 Markdown 落成统一版式的 Word 文档”。它适合做：
+
+- 政府公文预设
+- 单位通知 / 简报 / 汇报材料
+- 团队内部统一模板
+
+## 核心思路
+
+把 Markdown 当成“内容源”，把版式规范抽成“preset”。
+
+你只要按固定约定写 Markdown：
+
+- `#`：文档标题
+- `##`：一级标题
+- `###`：二级标题
+- 空行分段
+- `-` / `*` / `+`：无序列表
+- `1.`：有序列表
+- `<!-- pagebreak -->`：分页
+
+然后执行一次命令，就能得到带统一字体、字号、缩进、页边距的 `.docx`。
+
+## 内置预设
+
+各 preset 的详细说明都放在 `src/mdstyledocx/preset_specs/` 下：
+
+- [default](src/mdstyledocx/preset_specs/default.md)：通用正式文档版式
+- [gov-cn](src/mdstyledocx/preset_specs/gov-cn.md)：中文政府公文风格基线
+- [gov-cn-hei](src/mdstyledocx/preset_specs/gov-cn-hei.md)：黑体标题、楷体二级标题、仿宋正文
+
+目录约定：
+
+- `*.json`：样式参数、页边距、字体、缩进等机器可读定义
+- `*.md`：该模板的写作约定、推荐语法和边界说明
+
+## 使用方式
+
+推荐直接用 `uv`：
+
+```bash
+uv sync
+uv run python -m unittest
+uv run mdstyledocx --list-presets
+uv run mdstyledocx --show-preset-rules gov-cn
+uv run mdstyledocx examples/gov_notice.md -o examples/gov_notice.docx --preset gov-cn
+```
+
+如果不使用 `uv`，也可以用传统方式：
+
+```bash
+pip install -e .
+python3 -m unittest
+mdstyledocx examples/gov_notice.md -o examples/gov_notice.docx --preset gov-cn
+```
+
+## 自定义预设
+
+可以在内置 preset 基础上再叠加一个 JSON 覆盖文件：
+
+```bash
+mdstyledocx input.md -o output.docx --preset gov-cn --preset-file my-preset.json
+```
+
+如果想先看某个模板要求什么 Markdown 写法：
+
+```bash
+mdstyledocx --show-preset-rules gov-cn
+```
+
+示例：
+
+```json
+{
+  "extends": "gov-cn",
+  "styles": {
+    "body": {
+      "line": 520,
+      "line_rule": "exact"
+    },
+    "title": {
+      "size_half_points": 40
+    }
+  }
+}
+```
+
+## Markdown 约定
+
+README 只保留通用约定。某个 preset 的专用写法，以对应的 `preset_specs/*.md` 为准。
+
+通用写法：
+
+```md
+# 关于开展示例工作的通知
+
+各有关单位：
+
+为统一输出格式，现将有关事项通知如下。
+
+## 一、工作目标
+
+1. 统一内容源。
+2. 统一输出格式。
+
+## 二、工作要求
+
+请各单位按要求执行。
+```
+
+## 当前边界
+
+当前版本优先保证：
+
+- 标题、段落、列表、分页可稳定导出
+- 预设版式可复用
+- 产物是标准 `.docx`
+
+暂未覆盖：
+
+- 表格
+- 图片
+- 脚注
+- 复杂嵌套列表
+- 目录自动生成
+
+如果后面继续做，这个工具可以自然扩展成：
+
+- 多个行业 preset 集合
+- frontmatter 驱动的页面元信息
+- 更完整的 Markdown 语法支持
+- GUI 或 Web 包装层

mdstyledocx-0.1.0.dist-info/RECORD

@@ -0,0 +1,19 @@
+mdstyledocx/__init__.py,sha256=4vSWxWbz4mSgVO8CfC3mqnjArLUjXbX2FWulkpYTh1o,77
+mdstyledocx/__main__.py,sha256=_LB_VxImJA0ewYU5y_x9UsqOECJFom3PZHAsFwe9GR0,91
+mdstyledocx/cli.py,sha256=0dst4k-WOHizwMIioaV14TypEze6isFbKn_sOYVMUY4,2583
+mdstyledocx/docx_writer.py,sha256=PggklNBsOcyytZq1T9DHdqCLCTrfxEgUPojr-Hki7u8,8848
+mdstyledocx/markdown.py,sha256=MKEra2G5rmbBAP9BfcK3mEqxRWHlETwSJN6Hr_hIM1U,6178
+mdstyledocx/model.py,sha256=R9KpQ8ig3jX371Gwa95V4JF6x0pDuM2s-jdZTjp7NyI,650
+mdstyledocx/presets.py,sha256=WpBKiJCQXHFpkUQF6E1KK4DcDP3-lEANFfVA2hZhcto,3649
+mdstyledocx/preset_specs/default.json,sha256=yXwkf2BOiCvQSjPJKROlemCvDKbDWkjGPL7j5XLY4H0,1395
+mdstyledocx/preset_specs/default.md,sha256=3pxQpz1pM0e_KZIViyP3Rt-LAx0DuqEd6OjFSJw9KME,690
+mdstyledocx/preset_specs/gov-cn-hei.json,sha256=VX_qZXZvqW-FlJjF5rxA8YmGKNoXlz5CbfaV8CAlx38,1633
+mdstyledocx/preset_specs/gov-cn-hei.md,sha256=1cSPh8wOor_BToO73sWa94OH3R6TQhAdVkXhlk5gM3s,1547
+mdstyledocx/preset_specs/gov-cn.json,sha256=tmpKL3-iMgWB3N9U3NiLiRpCCiM6BFfADpLF2G2XL_Y,1626
+mdstyledocx/preset_specs/gov-cn.md,sha256=4rX3S_eR1WYdWD2lR8FVCJuYr-96qN5vDwwC7uIMiFY,1655
+mdstyledocx-0.1.0.dist-info/licenses/LICENSE,sha256=43iGV-F6hCXbP392AGPVFkJOX-eB4aZ5MmGUiZyYovE,1079
+mdstyledocx-0.1.0.dist-info/METADATA,sha256=Bu1a7xVVoqYMzk7RK01OsqRaccLzNgpvSF2Z2qQvEfA,3352
+mdstyledocx-0.1.0.dist-info/WHEEL,sha256=aeYiig01lYGDzBgS8HxWXOg3uV61G9ijOsup-k9o1sk,91
+mdstyledocx-0.1.0.dist-info/entry_points.txt,sha256=FIQJ7HcdSo9BelxMVuipTJID0Qjcev5224vPD6UQy-I,53
+mdstyledocx-0.1.0.dist-info/top_level.txt,sha256=gQ8aSe-ISLHZ1CcQ2rqJD1BF_FGSKnSJmIil6hOyvNU,12
+mdstyledocx-0.1.0.dist-info/RECORD,,

mdstyledocx-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (82.0.1)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

mdstyledocx-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+mdstyledocx = mdstyledocx.cli:main

mdstyledocx-0.1.0.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 YANG Zijie (杨子杰)
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

mdstyledocx-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+mdstyledocx