Files
Orchestrator/privat/CV/andreknie.de/scripts/dedupe-kniepunkt.js
T
ankn b46de6dc7d
Deploy CV Site (andreknie.de) / deploy (push) Canceled after 0s
ci: migrate GitHub actions to Gitea actions
2026-07-22 15:42:06 +02:00

92 lines
2.9 KiB
JavaScript

import fs from 'fs'
import path from 'path'
import matter from 'gray-matter'
const CONTENT_DIR = path.join(process.cwd(), 'content')
const KNIEPUNKT_DIR = path.join(CONTENT_DIR, 'kniepunkt')
const POSTS_DIR = path.join(CONTENT_DIR, 'posts')
function getFiles(dir) {
return fs.readdirSync(dir).map(f => path.join(dir, f)).filter(f => f.endsWith('.md'))
}
// 1. Clean up stubs (less than 150 chars of body)
const allPosts = getFiles(POSTS_DIR)
allPosts.forEach(file => {
const content = fs.readFileSync(file, 'utf-8')
const { content: body } = matter(content)
if (body.trim().length < 150) {
console.log(`🗑️ Deleting stub: ${path.basename(file)} (length: ${body.trim().length})`)
fs.unlinkSync(file)
}
})
// 2. Deduplicate Kniepunkte
const allKniepunktFiles = getFiles(KNIEPUNKT_DIR)
const issues = {}
allKniepunktFiles.forEach(file => {
const content = fs.readFileSync(file, 'utf-8')
const { data, content: body } = matter(content)
if (body.trim().length < 150) {
console.log(`🗑️ Deleting stub: ${path.basename(file)}`)
fs.unlinkSync(file)
return
}
const filename = path.basename(file)
// Extract issue number (1 to 3 digits)
const match = filename.match(/(?:kniepunkt|kp|issue|ausgabe)[^\d]*(\d{1,3})/i) || data.title.match(/(?:kniepunkt|kp|issue|ausgabe)[^\d]*(\d{1,3})/i)
let num = -1
if (match) {
num = parseInt(match[1], 10)
} else {
// try to find any standalone 2-3 digit number
const m2 = filename.match(/\b(\d{2,3})\b/)
if (m2) num = parseInt(m2[1], 10)
}
if (num !== -1) {
if (!issues[num]) issues[num] = []
issues[num].push({ file, filename, data, body, length: body.length })
} else {
console.log(`⚠️ Could not parse issue number for: ${filename}`)
}
})
Object.keys(issues).forEach(num => {
const files = issues[num]
if (files.length > 1) {
console.log(`\n🔍 Found ${files.length} duplicates for Issue ${num}:`)
// Pick the best one
// Priority: 1. Real date (not 2026-07-18) 2. Longest content 3. Starts with YYYY-MM-DD
files.sort((a, b) => {
const aRealDate = a.data.date !== '2026-07-18'
const bRealDate = b.data.date !== '2026-07-18'
if (aRealDate && !bRealDate) return -1
if (!aRealDate && bRealDate) return 1
const aDatePrefix = /^\d{4}-\d{2}-\d{2}/.test(a.filename)
const bDatePrefix = /^\d{4}-\d{2}-\d{2}/.test(b.filename)
if (aDatePrefix && !bDatePrefix) return -1
if (!aDatePrefix && bDatePrefix) return 1
return b.length - a.length // Longer is better
})
const best = files[0]
console.log(`✅ Keeping best: ${best.filename} (Date: ${best.data.date}, Len: ${best.length})`)
// Delete the rest
for (let i = 1; i < files.length; i++) {
console.log(`🗑️ Deleting duplicate: ${files[i].filename}`)
fs.unlinkSync(files[i].file)
}
}
})
console.log('✅ Deduplication complete!')