AgentSkillsCN

Text Merger

文本合并器

SKILL.md

Classical Text Merger Skill

A skill to merge classical Chinese texts (like the Daodejing) with their modern translations, handling inconsistent numbering and chapter matching.

Overview

This skill provides a robust way to take two files—one with original classical Chinese text and another with modern translations—and merge them into a single, well-formatted document. It intelligently handles:

  • Mixed numbering formats (Chinese vs. Arabic numerals).
  • Inconsistent chapter headers.
  • Potential mislabeling or shifts in chapter numbering in source files.

Files

  • merge_daodejing.js: The core Node.js script that performs the parsing and merging logic.
  • 道德经原文.txt: Source file for original text.
  • 道德经翻译.txt: Source file for translations.

Usage

To merge the files, run the following command in the terminal:

powershell
node merge_daodejing.js

The script will:

  1. Parse the original text, extracting chapters based on Chinese numeral headers (e.g., 第一章).
  2. Parse the translation text, extracting chapters based on bracketed or bolded headers (e.g., 第1章).
  3. Convert all chapter numbers to a common Arabic format for matching.
  4. Apply correction logic for any known numbering shifts in the source translation.
  5. Generate 道德经-原文与翻译.txt.

Matching Logic Details

The skill uses a chineseToNumber utility to normalize chapter identifiers:

  • Handles single digits: 一 -> 1
  • Handles tens: 十一 -> 11, 二十 -> 20, 二十一 -> 21
  • Handles custom shifts (e.g., if Chapter 67 in the source is actually Chapter 80).

Output Format

The merged file follows this structure:

text
第[编号]章
------------------------------------------------------------
【原文】
[原文内容]

【翻译】
[翻译内容]

============================================================