code-to-images

Convert code files to A4-ratio PNG/SVG images with line numbers and syntax highlighting, then merge to PDF. 支持中英文双语说明。

Install

openclaw skills install @yon-gjun/code-to-images

Code → A4 图片 + PDF / A4 Images + PDF

v2.2.0 — 49 种语言 · 多语言自动识别 · 忠实缩进 · Auto word-wrapping

📖 中文 · English

中文说明 / Chinese

🇨🇳 中文 | 中文

将源代码文件转换为多页 A4 比例图片（SVG+PNG），然后合并为一个 PDF。

功能

✅ 49 种语言自动识别 — 按扩展名 / 文件名 / 内容自动选择语法规则：Python、C/C++、Arduino、Java、JavaScript、TypeScript、Go、Rust、Ruby、Shell、Lua、SQL、PHP、C#、CSS、HTML、JSON、YAML、仓颉 Cangjie、Kotlin、Swift、Dart、R、Perl、MATLAB、Objective-C、Scala、Haskell、Julia、Zig、Groovy、PowerShell、Batch、Visual Basic、Makefile、Dockerfile、INI/TOML、Markdown、LaTeX、CMake、Solidity、Erlang、Elixir、Clojure、F#、Pascal、Fortran、汇编
✅ 多语言注释与字符串 — 跨行块注释（/* */、、--[[ ]]、{- -}、#= =#、(* *)、{ } 等）、三引号字符串（Python/Julia 的 """ '''）、#////--/;/!/' 行注释
✅ 语言特性高亮 — HTML/XML 标签、@ 注解与指令（Java/Python/ObjC/Swift）、$ 变量（Shell/PHP/Perl/PowerShell）、Makefile 目标行与 YAML 键、Markdown 标题
✅ 行号 — 自适应装订线宽度
✅ 语法高亮 — 关键字、寄存器、宏、数字、字符串、变量、标签、注释分别着色
✅ 6 种等宽字体 — Cascadia Code / Fira Code / JetBrains Mono / Consolas / Courier New / Source Code Pro
✅ 字号自动适配 — 根据最长行自动缩放，填满 A4 宽度
✅ 每页行数自适应 — 根据字号和行高自动计算
✅ 忠实缩进 — 保留源码实际前导空格 / Tab，C 风格大括号与 Python 风格空格缩进都正确呈现
✅ 代码超宽自动换行 — 在 ;,(){}[] 标点后折行，续行不产生新行号
✅ Tab 制表位 — 正确按制表位对齐
✅ 固定 A4 比例 — 794×1123px (96 DPI)
✅ PDF 输出 — 每个源文件一份 PDF

💡 .m 文件自动嗅探：含 @interface/#import/NSString 识别为 Objective-C，否则识别为 MATLAB；Makefile/Dockerfile/CMakeLists.txt 按文件名识别。

依赖

bash

npm install -g @resvg/resvg-js
pip install img2pdf

使用

编辑 gen_code_pdfs.py 顶部的 FILES 列表，填入源文件名
按需修改 FONT_KEY、FONT_SIZE 等配置
运行：python gen_code_pdfs.py
输出：文件名.pdf + 文件名_images/ 目录（中间 SVG/PNG）

配置

参数	默认值	说明
`FONT_KEY`	`cascadia`	字体：`cascadia`/`firacode`/`jetbrains`/`consolas`/`courier`/`sourcecode`
`FONT_SIZE`	`14`	基础字号 (px)，超宽代码自动缩小
`LINE_HEIGHT_RATIO`	`1.65`	行高系数
`TAB_WIDTH`	`4`	Tab 制表位宽度
`MIN_FONT_SIZE`	`7`	自动缩小时的最小字号
`WRAP_IDENT`	`2`	续行额外缩进字符数

English / English

🇬🇧 English | 中文

Convert source code files into multi-page A4-ratio images (SVG+PNG), then merge into a single PDF.

Features

✅ 49-language auto-detection — Picks syntax rules by extension / basename / content sniff: Python, C/C++, Arduino, Java, JavaScript, TypeScript, Go, Rust, Ruby, Shell, Lua, SQL, PHP, C#, CSS, HTML, JSON, YAML, Cangjie, Kotlin, Swift, Dart, R, Perl, MATLAB, Objective-C, Scala, Haskell, Julia, Zig, Groovy, PowerShell, Batch, Visual Basic, Makefile, Dockerfile, INI/TOML, Markdown, LaTeX, CMake, Solidity, Erlang, Elixir, Clojure, F#, Pascal, Fortran, Assembly
✅ Rich comments & strings — Cross-line block comments (/* */, , --[[ ]], {- -}, #= =#, (* *), { }…), triple-quoted strings (Python/Julia """ '''), #////--/;/!/' line comments
✅ Language-specific tokens — HTML/XML tags, @ annotations & directives (Java/Python/ObjC/Swift), $ variables (Shell/PHP/Perl/PowerShell), Makefile targets & YAML keys, Markdown headings
✅ Line numbers — Adaptive gutter width
✅ Syntax highlighting — Keywords, registers, macros, numbers, strings, variables, tags, comments
✅ 6 monospace fonts — Cascadia Code / Fira Code / JetBrains Mono / Consolas / Courier New / Source Code Pro
✅ Auto font size — Scales to fill A4 width based on longest line
✅ Auto lines per page — Computed from font size and line height
✅ Faithful indentation — Preserves actual leading whitespace / tabs for both brace-based and space-based languages
✅ Auto word-wrapping — Breaks after ;,(){}[], continuation lines get no new line number
✅ Tab stops — Proper tab alignment
✅ Fixed A4 ratio — 794×1123px (96 DPI)
✅ PDF output — One PDF per source file

💡 .m content sniffing: files with @interface/#import/NSString are treated as Objective-C, otherwise MATLAB; Makefile/Dockerfile/CMakeLists.txt are detected by basename.

Prerequisites

bash

npm install -g @resvg/resvg-js
pip install img2pdf

Usage

Edit the FILES list at the top of gen_code_pdfs.py
Configure FONT_KEY, FONT_SIZE, etc.
Run: python gen_code_pdfs.py
Output: filename.pdf + filename_images/ directory (intermediate SVG/PNG)

Configuration

Parameter	Default	Description
`FONT_KEY`	`cascadia`	Font: `cascadia`/`firacode`/`jetbrains`/`consolas`/`courier`/`sourcecode`
`FONT_SIZE`	`14`	Base font size (px), auto-reduces for wide code
`LINE_HEIGHT_RATIO`	`1.65`	Line height multiplier
`TAB_WIDTH`	`4`	Tab stop width
`MIN_FONT_SIZE`	`7`	Minimum font size when auto-reducing
`WRAP_IDENT`	`2`	Extra indent for continuation lines

Python Batch Script

Save as gen_code_pdfs.py and run with python gen_code_pdfs.py:

python

#!/usr/bin/env python3
"""Batch convert code files -> A4 images + PDF. Multi-language auto-detection."""
import os, sys, subprocess, json

# ===== CONFIG =====
# Language is auto-detected from each file's extension / basename / content (see LANG / EXT2LANG / BASE2LANG below).
FILES = [/* List source files here, e.g. 'main.c', 'game.py', 'app.ts', 'Makefile' */]
FONT_KEY = '***'   # cascadia | firacode | jetbrains | consolas | courier | sourcecode
FONT_SIZE = 14; LINE_HEIGHT_RATIO = 1.65; TAB_WIDTH = 4; MIN_FONT_SIZE = 7; WRAP_IDENT = 2
NODE_EXE = os.environ.get('NODE_EXE', 'node')   # full path to node if not on PATH

FONT_NAMES = {'cascadia':'Cascadia Code','firacode':'Fira Code','jetbrains':'JetBrains Mono',
    'consolas':'Consolas','courier':'Courier New','sourcecode':'Source Code Pro'}
A4_W, A4_H = 794, 1123; HDR_H = 60; PAD_R = 36; PAD_T = 72; PAD_B = 28
CONTENT_H = A4_H - PAD_T - PAD_B; CHAR_RATIO = 0.60; WRAP_PUNCT = set(';,(){}[]')
CO = {'keyword':'#d63384','register':'#e8590c','macro':'#9c36b5',
      'number':'#2b8a3e','string':'#099268','comment':'#868e96','var':'#1971c2','tag':'#0b7285','text':'#212529'}
BG='#ffffff'; HDR_BG='#f1f3f5'; HDR_BD='#dee2e6'; GT_BG='#f8f9fa'; GT_BD='#e9ecef'; LN_COLOR='#868e96'

# Per-language rules (keyword set, comment styles, string quotes, ...).
LANG = {
    'python': {'name':'Python','kw':{'def','class','import','from','return','if','for',...},
               'sp':set(),'hash':'comment','slash':False,'dash':False,'block':None,'quotes':'"\'','backtick':False,
               'triple':('"""',"'''"),'atword':True},
    'c':      {'name':'C','kw':{'void','char','int','for','if','while',...},'sp':{'P0',...,'P7','RST'},
               'hash':'macro','slash':True,'dash':False,'block':('/*','*/'),'quotes':'"\'','backtick':False},
    # ... 40+ languages: see gen_code_pdfs.py (LANG / EXT2LANG / BASE2LANG)
}
EXT2LANG = {'.py':'python', '.c':'c', '.h':'c', '.cpp':'c++', '.java':'java', '.js':'js', '.go':'go',
            '.rs':'rust', '.rb':'ruby', '.sh':'shell', '.lua':'lua', '.sql':'sql', '.php':'php',
            '.cs':'c#', '.css':'css', '.html':'html', '.xml':'html', '.json':'json', '.yaml':'yaml',
            '.ts':'typescript', '.kt':'kotlin', '.swift':'swift', '.m':'ambig_m', '.mm':'objective-c',
            '.r':'r', '.pl':'perl', '.scala':'scala', '.hs':'haskell', '.jl':'julia', '.zig':'zig',
            '.groovy':'groovy', '.ps1':'powershell', '.bat':'batch', '.vb':'vb', '.ino':'arduino',
            '.mk':'makefile', '.ini':'ini', '.toml':'ini', '.md':'markdown', '.tex':'tex',
            '.cmake':'cmake', '.sol':'solidity', '.erl':'erlang', '.ex':'elixir', '.clj':'clojure',
            '.fs':'fsharp', '.pas':'pascal', '.f90':'fortran', '.asm':'assembly'}
BASE2LANG = {'makefile':'makefile','dockerfile':'dockerfile','cmakelists.txt':'cmake'}

# [Full script with LANG/EXT2LANG tables at: gen_code_pdfs.py (local, v2.2.0)]