md-to-pdf

name	md-to-pdf
description	Convert Markdown with LaTeX math, images, and tables into a polished PDF. Two rendering engines: (1) weasyprint — pure Python, no browser needed; (2) chromium — best for matrices/bmatrix/pmatrix/vmatrix and complex formulas. Supports Chinese/Japanese fonts. Can chain after ocr-md-polish for a complete OCR→clean→PDF workflow.

Markdown → PDF Converter

Converts a Markdown file with $...$ / $$...$$ LaTeX into a typeset PDF.

Two engines:

Engine	When to use	Notes
weasyprint (default)	Most documents, pure Python, no browser needed	MathML + CSS. May flatten matrices like bmatrix on some versions.
chromium	Complex formulas (bmatrix, pmatrix, vmatrix, vmatrix, det, nested fractions, …)	Requires system chromium / google-chrome to be installed. Uses native MathML rendering.

Quick Start

# 默认引擎 weasyprint
uv run python3 /home/fuurin/.claude/skills/md-to-pdf/convert.py input.md --font "LXGW WenKai"

# 矩阵/复杂公式 —— 用 chromium（需系统已装 chromium）
uv run python3 /home/fuurin/.claude/skills/md-to-pdf/convert.py input.md --font "LXGW WenKai" --engine chromium

Pipeline

[PDF] → pdf-math-convert → [raw .md] → ocr-md-polish → [clean .md] → md-to-pdf → [polished PDF]

Usage

1. Dependencies

# 默认引擎（最小方案）
uv add weasyprint marko latex2mathml

# chromium 引擎（需系统已安装 chromium/google-chrome）
uv add marko latex2mathml

2. Font preference

Name	CSS family	Style
霞鹜文楷	LXGW WenKai	楷体/手写
思源宋体	Noto Serif CJK SC	衬线/书籍
思源黑体	Noto Sans SC	无衬线/现代

The converter auto-detects installed fonts; LXGW WenKai is preferred when present, otherwise Noto Serif CJK SC.

3. Convert

Single file:

uv run python3 /home/fuurin/.claude/skills/md-to-pdf/convert.py input.md [output.pdf] [--font "LXGW WenKai"] [--engine weasyprint|chromium]

Batch:

uv run python3 /home/fuurin/.claude/skills/md-to-pdf/convert.py note1.md note2.md --font "LXGW WenKai" --engine chromium

Directory / glob:

uv run python3 /home/fuurin/.claude/skills/md-to-pdf/convert.py ./lectures/ --engine chromium
uv run python3 /home/fuurin/.claude/skills/md-to-pdf/convert.py chapter*.md --engine chromium

4. When to pick --engine chromium

Use --engine chromium when your Markdown contains any of these, because weasyprint's MathML is limited for them:

• Matrix environments: \begin{bmatrix}…\end{bmatrix}, \begin{pmatrix}…, \begin{vmatrix}…, \begin{matrix}…
• Determinants \det, column vectors, multi-line equations
• Nested \frac, \lim, \sum with sub/super-scripts

The script auto-looks for chromium / google-chrome in /usr/sbin, /usr/bin, and PATH. If none is found it exits with a clear message.

Features

Feature	Status
$...$ inline math	✅ MathML rendering
$$...$$ display math	✅ Centered block
\frac{a}{b} fractions	✅
x_0, x^2 sub/superscripts	✅
\lim, \int, \sum	✅
\begin{bmatrix}…, \begin{pmatrix}…	✅ (best with --engine chromium)
Local images, base64 images	✅
Tables	✅
Chinese / CJK text	✅ with --font
Code blocks	✅ styled CSS
Header/footer-free PDF	✅ chromium: --no-pdf-header-footer

Integration with ocr-md-polish

This skill is designed to be the final step after OCR cleanup:

1. ocr-md-polish fixes formula broken subscripts (x0 → x_0), nested dollar signs, OCR-duplicated image text, and numbering
2. md-to-pdf takes the polished Markdown and produces the final PDF

If input file is a .pdf, suggest using pdf-math-convert first to get Markdown, then follow with ocr-md-polish → md-to-pdf.