feat: add Playwright long image downloads

2026-07-02 18:10:33 +08:00
parent 8260ff1e0e
commit f9eb3d1440
16 changed files with 993 additions and 64 deletions
@@ -0,0 +1,106 @@
+import fs from 'node:fs';
+import fsPromises from 'node:fs/promises';
+import os from 'node:os';
+import path from 'node:path';
+import { pathToFileURL } from 'node:url';
+
+const DEFAULT_VIEWPORT_WIDTH = 1280;
+const DEFAULT_VIEWPORT_HEIGHT = 720;
+const MAX_LONG_IMAGE_HEIGHT = 20000;
+
+export function isLongImageDownloadRequest(query) {
+  const value = String(query?.download ?? query?.export ?? '').trim().toLowerCase();
+  return value === 'long-image' || value === 'long_image' || value === 'png';
+}
+
+export function longImagePathForHtml(htmlPath, outputPath = null) {
+  return outputPath || String(htmlPath).replace(/\.html$/i, '.long.png');
+}
+
+function clampDimension(value, fallback, max) {
+  const number = Math.ceil(Number(value) || fallback);
+  return Math.max(320, Math.min(number, max));
+}
+
+async function launchChromium(chromium) {
+  const base = {
+    headless: true,
+    args: ['--disable-dev-shm-usage', '--hide-scrollbars'],
+  };
+  if (process.env.PLAYWRIGHT_CHROMIUM_EXECUTABLE_PATH) {
+    return chromium.launch({
+      ...base,
+      executablePath: process.env.PLAYWRIGHT_CHROMIUM_EXECUTABLE_PATH,
+    });
+  }
+  if (process.env.PLAYWRIGHT_CHROMIUM_CHANNEL) {
+    return chromium.launch({ ...base, channel: process.env.PLAYWRIGHT_CHROMIUM_CHANNEL });
+  }
+  try {
+    return await chromium.launch(base);
+  } catch (error) {
+    if (process.platform === 'darwin') {
+      return chromium.launch({ ...base, channel: 'chrome' });
+    }
+    throw error;
+  }
+}
+
+export async function renderLongImage({
+  htmlPath = null,
+  url = null,
+  outputPath = null,
+  viewportWidth = DEFAULT_VIEWPORT_WIDTH,
+  viewportHeight = DEFAULT_VIEWPORT_HEIGHT,
+} = {}) {
+  if (!htmlPath && !url) throw new Error('缺少 htmlPath 或 url');
+  const targetUrl = url || pathToFileURL(path.resolve(htmlPath)).toString();
+  const destination = outputPath ? path.resolve(outputPath) : longImagePathForHtml(path.resolve(htmlPath));
+  const { chromium } = await import('playwright');
+  let browser = null;
+  try {
+    browser = await launchChromium(chromium);
+    const page = await browser.newPage({
+      viewport: {
+        width: clampDimension(viewportWidth, DEFAULT_VIEWPORT_WIDTH, 2400),
+        height: clampDimension(viewportHeight, DEFAULT_VIEWPORT_HEIGHT, MAX_LONG_IMAGE_HEIGHT),
+      },
+      deviceScaleFactor: 2,
+    });
+    await page.goto(targetUrl, { waitUntil: 'networkidle', timeout: 30000 });
+    await page.evaluate(() => document.fonts?.ready).catch(() => null);
+    const size = await page.evaluate(() => ({
+      width: Math.max(document.documentElement.scrollWidth, document.body.scrollWidth, 320),
+      height: Math.max(document.documentElement.scrollHeight, document.body.scrollHeight, 320),
+    }));
+    await page.setViewportSize({
+      width: clampDimension(size.width, DEFAULT_VIEWPORT_WIDTH, 2400),
+      height: Math.min(clampDimension(size.height, DEFAULT_VIEWPORT_HEIGHT, MAX_LONG_IMAGE_HEIGHT), 2400),
+    });
+    await fsPromises.mkdir(path.dirname(destination), { recursive: true });
+    await page.screenshot({
+      path: destination,
+      fullPage: true,
+      type: 'png',
+      animations: 'disabled',
+      caret: 'hide',
+    });
+    return {
+      outputPath: destination,
+      bytes: fs.statSync(destination).size,
+    };
+  } finally {
+    await browser?.close().catch(() => {});
+  }
+}
+
+export async function renderLongImageBuffer({ url }) {
+  const tempDir = await fsPromises.mkdtemp(path.join(os.tmpdir(), 'memind-long-image-'));
+  const outputPath = path.join(tempDir, 'page.long.png');
+  try {
+    await renderLongImage({ url, outputPath });
+    return await fsPromises.readFile(outputPath);
+  } finally {
+    await fsPromises.rm(tempDir, { recursive: true, force: true }).catch(() => {});
+  }
+}