#!/usr/bin/env node
/**
 * Deduplicate files in uploads/ by content hash.
 * Keeps the first file per hash, removes later duplicates, and deletes matching Photo docs.
 * Dry run by default; set APPLY=1 to actually delete.
 */
const fs = require('fs');
const path = require('path');
const crypto = require('crypto');
const { pipeline } = require('stream/promises');
const mongoose = require('mongoose');
const Photo = require('../models/photo');

const MONGO_URI = process.env.MONGO_URI || 'mongodb://localhost:27017/photogallery';
const APPLY = process.env.APPLY === '1';
const UPLOAD_DIR = path.join(__dirname, '..', 'uploads');

async function hashFile(filePath) {
    const hash = crypto.createHash('sha256');
    const stream = fs.createReadStream(filePath);
    await pipeline(stream, hash);
    return hash.digest('hex');
}

function walkFiles(dir) {
    const results = [];
    for (const entry of fs.readdirSync(dir)) {
        const full = path.join(dir, entry);
        const stat = fs.statSync(full);
        if (stat.isDirectory()) {
            results.push(...walkFiles(full));
        } else {
            results.push(full);
        }
    }
    return results;
}

async function main() {
    const files = walkFiles(UPLOAD_DIR).filter(f => f.toLowerCase().endsWith('.webp'));
    console.log(`Scanning ${files.length} webp files...`);

    const hashMap = new Map();
    const dupes = [];

    for (const file of files) {
        const h = await hashFile(file);
        if (!hashMap.has(h)) {
            hashMap.set(h, file);
        } else {
            dupes.push({ hash: h, keep: hashMap.get(h), remove: file });
        }
    }

    if (!dupes.length) {
        console.log('No duplicate content found.');
        return;
    }

    console.log(`Found ${dupes.length} duplicate files (by content).`);
    dupes.slice(0, 10).forEach(d =>
        console.log(`Hash ${d.hash.slice(0, 12)} keep=${path.basename(d.keep)} remove=${path.basename(d.remove)}`)
    );

    if (!APPLY) {
        console.log('Dry run. Set APPLY=1 to delete duplicates and matching Photo docs.');
        return;
    }

    await mongoose.connect(MONGO_URI);
    console.log(`Connected to Mongo: ${MONGO_URI}`);

    let removedFiles = 0;
    let removedDocs = 0;
    for (const d of dupes) {
        try {
            fs.unlinkSync(d.remove);
            removedFiles++;
        } catch (err) {
            console.error('Failed to delete file', d.remove, err.message);
        }
        const relPath = path.relative(path.join(__dirname, '..'), d.remove).replace(/\\/g, '/');
        const filename = path.basename(d.remove);
        const res = await Photo.deleteMany({
            $or: [
                { path: relPath },
                { filename }
            ]
        });
        removedDocs += res.deletedCount || 0;
    }

    console.log(`Deleted ${removedFiles} duplicate files.`);
    console.log(`Deleted ${removedDocs} Photo docs matching removed files.`);
    await mongoose.disconnect();
    console.log('Done.');
}

main().catch(err => {
    console.error(err);
    process.exit(1);
});