moshai-cli/commands/Download/FileAssets.js

A modern, fast Node.js CLI powered by arasadrahman

const fs = require('fs');
const path = require('path');
const os = require('os');
const axios = require('axios');
const cheerio = require('cheerio');
const mkdirp = require('mkdirp');
const chalk = require('chalk').default;
const urlLib = require('url');

module.exports = class DownloadFileAssetsCommand {
    static signature = 'download:assets {htmlFilePath} {outputDir?}';
    static description = 'Download assets (img, css, js, fonts) from main site only and rewrite paths';

    async handle({ htmlFilePath, outputDir }) {
        if (!fs.existsSync(htmlFilePath)) {
            console.error(chalk.red(`❌ File not found: ${htmlFilePath}`));
            return;
        }

        const html = fs.readFileSync(htmlFilePath, 'utf8');
        const $ = cheerio.load(html);
        const baseUrl = this.getBaseUrl($, htmlFilePath);
        const hostname = new URL(baseUrl).hostname;
        const defaultOut = path.join(os.homedir(), 'Desktop', hostname);
        outputDir = outputDir ? path.resolve(outputDir) : defaultOut;
        mkdirp.sync(outputDir);

        const assets = [];
        $('img[src],script[src],link[href],style').each((_, el) => {
            const tag = $(el);
            let attr = tag.is('link') || tag.is('style') ? 'href' : 'src';

            if (tag.is('style')) {
                const cssText = tag.html();
                const matches = [...cssText.matchAll(/url\(["']?(https?:\/\/[^"')]+)["']?\)/g)];
                matches.forEach(match => assets.push(match[1]));
                return;
            }

            const url = tag.attr(attr);
            if (url && url.startsWith('http') && new URL(url).origin === new URL(baseUrl).origin) {
                assets.push(url);
            }
        });

        const uniqueAssets = [...new Set(assets)];
        for (const assetUrl of uniqueAssets) {
            try {
                const parsed = urlLib.parse(assetUrl);
                const relativePath = parsed.pathname.replace(/^\//, '');
                const assetPath = path.join(outputDir, relativePath);
                mkdirp.sync(path.dirname(assetPath));
                const res = await axios.get(assetUrl, { responseType: 'arraybuffer' });
                fs.writeFileSync(assetPath, res.data);
                console.log(chalk.green(`✔ Downloaded: ${relativePath}`));

                // Rewrite HTML
                $('*').each((_, el) => {
                    const tag = $(el);
                    ['src', 'href'].forEach(attr => {
                        if (tag.attr(attr) === assetUrl) {
                            tag.attr(attr, relativePath);
                        }
                    });
                    if (tag.is('style')) {
                        let newCss = tag.html().replace(
                            new RegExp(assetUrl, 'g'),
                            relativePath
                        );
                        tag.html(newCss);
                    }
                });
            } catch (e) {
                console.error(chalk.red(`✖ Failed to download ${assetUrl}: ${e.message}`));
            }
        }

        const newHtml = $.html();
        fs.writeFileSync(path.join(outputDir, 'index.html'), newHtml);
        console.log(chalk.green('✅ HTML updated and saved.'));
    }

    getBaseUrl($, filePath) {
        const baseTag = $('base').attr('href');
        if (baseTag && baseTag.startsWith('http')) return baseTag;
        const html = fs.readFileSync(filePath, 'utf8');
        const matches = html.match(/https?:\/\/[^\s"'<>]+/g);
        if (matches && matches.length) return matches[0];
        throw new Error('Could not determine base URL from HTML');
    }
};