compare-layouts
Skill135 starsChanged 2 days agoApache-2.0View on GitHub
比较 Original (ltc-guji.cls) 和 Digital (ltc-guji-digital.cls) TeX 文件的 layout 输出
In this repository17 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- clocSkill
- convert-to-digitalSkill
- fix-github-issueSkill
- prepare-next-versionSkill
- refactorSkill
- regression-testSkill
- release_processSkill
- save_conversationSkill
- startupSkill
- summarize-experienceSkill
- test-texSkill
- track-workSkill
- update_changelogSkill
- update_wikiSkill
- 数字化一致性检查Skill
The file
Files
Guji Layout Compare
比较原始和数字化 TeX 文件的排版一致性。
使用方法
Python 脚本(推荐)
# 基本用法:自动检测编译、比较、生成报告
python3 scripts/digitalize/compare_guji_layouts.py original.tex digital.tex
# 强制重新编译
python3 scripts/digitalize/compare_guji_layouts.py original.tex digital.tex --force
# 指定报告输出路径
python3 scripts/digitalize/compare_guji_layouts.py original.tex digital.tex -o report.md
# 只比较已有的 JSON(跳过编译)
python3 scripts/digitalize/compare_guji_layouts.py original.tex digital.tex --no-compile功能特性
- 智能编译检测
- 自动检查 layout JSON 是否存在
- 验证 JSON 中的
source_mtime字段与 TeX 文件修改时间 - 仅在需要时编译,避免重复编译浪费时间
- page_summary 比较
- 优先使用简化的
page_summary字段(无坐标,仅文字内容) - 支持夹注双列(
[右小列|左小列]格式显示) - 自动统计总字数
- 兼容旧版 JSON(fallback 到
pages字段逐字符比较)
- 优先使用简化的
- 详细 Markdown 报告
- 概要统计(页数、字数、匹配率)
- 差异总结表格
- 第一处差异的详细逐列对比
- PDF 页面截图对比(需要
pdftoppm)
输出示例
## 步骤三:比较排版
原始 48 页 / 数字化 41 页
匹配页数:14
总字数:13224(一致)
前 10 页完全一致
第一处差异在 PDF第11页(对开第5页·右)
## 步骤四:生成报告
详细报告已保存至:排版一致性比较报告.md手动分步操作
如果需要手动控制编译流程:
步骤 1: 导出 Layout JSON
cd path/to/tex && ENABLE_EXPORT=1 lualatex yourfile.tex
# → yourfile-layout.json
ENABLE_EXPORT=1 lualatex yourfile-digital.tex
# → yourfile-digital-layout.json步骤 2: 只比较(跳过编译)
python3 scripts/compare_guji_layouts.py yourfile.tex yourfile-digital.tex --no-compileLayout JSON 结构
从 v0.2.9 开始,导出的 layout JSON 包含:
{
"version": "1.0",
"generator": "luatex-cn",
"source_file": "yourfile.tex",
"source_mtime": 1771885547,
"document": { ... },
"page_summary": [
{"page": 0, "type": "spread_right", "cols": ["欽定四庫全書", "", ...]},
...
],
"pages": [ ... ]
}page_summary: 每页每列的简化文字内容(用于快速比较)pages: 完整的字符位置/坐标信息(用于精确分析)
相关文件
- 比较脚本:
scripts/digitalize/compare_guji_layouts.py - Export 模块:
tex/core/luatex-cn-core-export.lua - 转换工具:
scripts/digitalize/semantic_to_digital.py— 语义→数字化转换scripts/digitalize/digital_to_semantic.py— 数字化→语义逆向转换scripts/digitalize/plugins/— 转换插件(如四库全书简明目录)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

