Bilingual Awesome List Readme

Bilingual Awesome-List README

Overview

Produces a README_CN.md alongside an existing English README.md, with:

• Translated section headers, descriptions, TOC, overview table, footer
• Identical repo table rows (reused verbatim — no translation needed)
• Mutual language-switcher line near the top of each file

Critical Pitfall

execute_code blocks are isolated sandboxes — variables do NOT persist between calls.
Put the entire parse → build → write pipeline in ONE execute_code block, or use terminal("cat ...") to re-read the file at the top of each block.

Step-by-Step

1. Read the EN file with terminal("cat ...")

Use terminal (not read_file) to get raw content without line-number prefixes:

from hermes_tools import terminal, write_file
import re

r = terminal("cat '/path/to/README.md'")
en_lines = r["output"].split('\n')

2. Parse section headers → repo table rows

EN_SECTION_HEADERS = {
    "## 🏛️ Official & Core": "官方与核心",
    # ... map every EN header to its ZH key
}

sections = {}   # zh_cat -> [table row strings]
current_cat = None
for line in en_lines:
    s = line.strip()
    if s in EN_SECTION_HEADERS:
        current_cat = EN_SECTION_HEADERS[s]
        sections[current_cat] = []
    elif current_cat and s.startswith('| [') and 'github.com' in s:
        sections[current_cat].append(s)

3. Define translation tables (in same block)

ORDER = ["官方与核心", "精选列表与资源合集", ...]  # display order

CAT_EMOJI   = {"官方与核心": "🏛️", ...}
CAT_DESC_ZH = {"官方与核心": "Claude Code 官方仓库与 Anthropic 维护的核心资源。", ...}

4. Build CN README lines

L = []

# Header block
L += [
    "# ✨ Awesome Claude Code Skills",
    "",
    "> GitHub 上 Star 数量最多的 ...",
    "",
    "[![Awesome](https://awesome.re/badge.svg)](https://awesome.re)",
    "...",                                           # translated badges
    "",
    "**语言切换 / Language:**   **🇨🇳 中文**  |  [🇺🇸 English](README.md)",
    "",
    "---",
    "",
]

# TOC
L += ["## 📋 目录", ""]
for cat in ORDER:
    emoji  = CAT_EMOJI[cat]
    count  = len(sections.get(cat, []))
    anchor = re.sub(r'[^\w\u4e00-\u9fff-]', '', cat.lower().replace(' ', '-'))
    L.append(f"- [{emoji} {cat}](#{anchor}) ({count})")
L += ["", "---", ""]

# Overview table
L += ["## 📊 概览", "", "...", "", "| 分类 | 项目数 |", "|------|:------:|"]
for cat in ORDER:
    L.append(f"| {CAT_EMOJI[cat]} {cat} | {len(sections.get(cat, []))} |")
L += ["", "---", ""]

# Sections (reuse EN table rows verbatim)
for cat in ORDER:
    rows = sections.get(cat, [])
    L += [
        f"## {CAT_EMOJI[cat]} {cat}",
        "",
        CAT_DESC_ZH[cat],
        "",
        "| 仓库 | Star | 简介 |",
        "|-----|-----:|------|",
    ]
    L += rows
    L.append("")

# Footer (Contributing + License)
L += ["---", "", "## 🤝 贡献", "", "...", "", "## 📄 许可证", "", "...", ""]

5. Write CN file

write_file("/path/to/README_CN.md", '\n'.join(L))

6. Patch EN README to add language switcher

Use patch tool to insert one line after the badge block:

**Language / 语言切换:**   **🇺🇸 English**  |  [🇨🇳 中文](README_CN.md)

Verification

# Check row counts match
grep -c '| \[' README.md
grep -c '| \[' README_CN.md
# Both should be equal (e.g. 200)

Notes

• TOC anchors for Chinese text: GitHub lowercases and strips non-alphanumeric/CJK chars. Use re.sub(r'[^\w\u4e00-\u9fff-]', '', cat.lower().replace(' ', '-')) as anchor.
• Repo table rows: never translate — reuse verbatim from EN so repo names/links/stars stay correct.
• Badge URLs: URL-encode Chinese characters or use ASCII-only badge labels to avoid rendering issues.
• Stars badge: keep Combined%20Stars in ASCII; year badge can use Chinese if URL-encoded.