refactor: use CloakBrowser

.github/workflows/docker-test.yml

@@ -43,7 +43,10 @@ jobs:
               uses: docker/build-push-action@bcafcacb16a39f128d818304e6c9c0c18556b85f # v7.1.0
               with:
                   context: .
-                  build-args: PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD=0 # also test bundling Chromium
+                  # also test bundling Chromium
+                  build-args: |
+                      PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD=0
+                      CLOAKBROWSER_SKIP_DOWNLOAD=0
                   load: true
                   tags: ${{ steps.meta.outputs.tags }}
                   labels: ${{ steps.meta.outputs.labels }}

.github/workflows/issue-command.yml

@@ -120,7 +120,7 @@ jobs:
                   cache: 'pnpm'
 
             - name: Install dependencies (pnpm)
-              run: pnpm i && pnpm rb && pnpm exec playwright install chromium
+              run: pnpm i && pnpm rb && pnpm exec cloakbrowser install
 
             - name: Fetch affected routes
               id: fetch-route

.github/workflows/test.yml

@@ -40,7 +40,7 @@ jobs:
             - name: Install dependencies (pnpm)
               run: pnpm i
             - name: Run postinstall script for dependencies
-              run: pnpm rb && pnpm exec playwright install chromium
+              run: pnpm rb && pnpm exec cloakbrowser install
             - name: Build routes
               run: pnpm build
             - name: Build worker routes
@@ -91,7 +91,7 @@ jobs:
               run: pnpm build
             - name: Install bundled Chromium
               if: ${{ matrix.chromium.dependency == '' }}
-              run: pnpm exec playwright install chromium
+              run: pnpm exec cloakbrowser install
             - name: Install Chromium
               if: ${{ matrix.chromium.dependency != '' }}
               # 'chromium-browser' from Ubuntu APT repo is a dummy package. Its version (85.0.4183.83) means

Dockerfile

@@ -40,7 +40,7 @@ WORKDIR /ver
 COPY ./package.json /app/
 RUN \
     set -ex && \
-    grep -Po '(?<="playwright": ")[^\s"]*(?=")' /app/package.json | tee /ver/.playwright_version && \
+    grep -Po '(?<="cloakbrowser": ")[^\s"]*(?=")' /app/package.json | tee /ver/.cloakbrowser_version && \
     grep -Po '(?<="@vercel/nft": ")[^\s"]*(?=")' /app/package.json | tee /ver/.nft_version && \
     grep -Po '(?<="fs-extra": ")[^\s"]*(?=")' /app/package.json | tee /ver/.fs_extra_version
 
@@ -88,29 +88,32 @@ FROM node:24-bookworm-slim AS chromium-downloader
 # Yeah, downloading Chromium never needs those dependencies below.
 
 WORKDIR /app
-COPY --from=dep-version-parser /ver/.playwright_version /app/.playwright_version
+COPY --from=dep-version-parser /ver/.cloakbrowser_version /app/.cloakbrowser_version
 
 ARG TARGETPLATFORM
 ARG USE_CHINA_NPM_REGISTRY=0
 ARG PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD=1
-# The official recommended way to use Playwright on x86(_64) is to use the bundled browser.
+ARG CLOAKBROWSER_SKIP_DOWNLOAD=1
+# CloakBrowser publishes prebuilt patched Chromium for both linux/amd64 and linux/arm64,
+ENV CLOAKBROWSER_CACHE_DIR=/app/node_modules/.cache/cloakbrowser
 RUN \
     set -ex ; \
-    if [ "$PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD" = 0 ] && [ "$TARGETPLATFORM" = 'linux/amd64' ]; then \
+    if [ "$PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD" = 0 ]; then \
         if [ "$USE_CHINA_NPM_REGISTRY" = 1 ]; then \
             npm config set registry https://registry.npmmirror.com && \
             yarn config set registry https://registry.npmmirror.com && \
             pnpm config set registry https://registry.npmmirror.com ; \
         fi; \
-        echo 'Downloading Chromium...' && \
+        echo 'Downloading CloakBrowser...' && \
         unset PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD && \
-        export PLAYWRIGHT_BROWSERS_PATH=/app/node_modules/.cache/ms-playwright && \
         corepack enable pnpm && \
-        pnpm --allow-build=playwright add playwright@$(cat /app/.playwright_version) --save-prod && \
-        pnpm rb && \
-        pnpm exec playwright install chromium ; \
+        pnpm add cloakbrowser@$(cat /app/.cloakbrowser_version) --save-prod && \
+        pnpm rb ; \
+        if [ "$CLOAKBROWSER_SKIP_DOWNLOAD" = 0 ]; then \
+            pnpm exec cloakbrowser install ; \
+        fi; \
     else \
-        mkdir -p /app/node_modules/.cache/ms-playwright ; \
+        mkdir -p "$CLOAKBROWSER_CACHE_DIR" ; \
     fi;
 
 # ---------------------------------------------------------------------------------------------------------------------
@@ -121,48 +124,51 @@ LABEL org.opencontainers.image.authors="https://github.com/DIYgod/RSSHub"
 
 ENV NODE_ENV=production
 ENV TZ=Asia/Shanghai
+ENV CLOAKBROWSER_AUTO_UPDATE=false
+ENV CLOAKBROWSER_DOWNLOAD_URL=https://github.com/CloakHQ/cloakbrowser/releases/download
 
 WORKDIR /app
 
 # install deps first to avoid cache miss or disturbing buildkit to build concurrently
 ARG TARGETPLATFORM
 ARG PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD=1
+ARG CLOAKBROWSER_SKIP_DOWNLOAD=1
 # https://playwright.dev/docs/docker#introduction
 # https://www.debian.org/releases/bookworm/amd64/release-notes/ch-information.en.html#noteworthy-obsolete-packages
-# On arm/arm64, install Chromium from the distribution repositories.
+# CloakBrowser ships prebuilt patched Chromium for both linux/amd64 and linux/arm64
 RUN \
     set -ex && \
     apt-get update && \
     apt-get install -yq --no-install-recommends \
         dumb-init git curl \
     ; \
     if [ "$PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD" = 0 ]; then \
-        if [ "$TARGETPLATFORM" = 'linux/amd64' ]; then \
-            apt-get install -yq --no-install-recommends \
-                ca-certificates fonts-liberation wget xdg-utils \
-                libasound2 libatk-bridge2.0-0 libatk1.0-0 libatspi2.0-0 libcairo2 libcups2 libdbus-1-3 libdrm2 \
-                libexpat1 libgbm1 libglib2.0-0 libnspr4 libnss3 libpango-1.0-0 libx11-6 libxcb1 libxcomposite1 \
-                libxdamage1 libxext6 libxfixes3 libxkbcommon0 libxrandr2 \
-            ; \
-        else \
-            apt-get install -yq --no-install-recommends \
-                chromium \
-            && \
-            echo "CHROMIUM_EXECUTABLE_PATH=$(which chromium)" | tee /app/.env ; \
-        fi; \
+        apt-get install -yq --no-install-recommends \
+            ca-certificates fonts-liberation wget xdg-utils \
+            libasound2 libatk-bridge2.0-0 libatk1.0-0 libatspi2.0-0 libcairo2 libcups2 libdbus-1-3 libdrm2 \
+            libexpat1 libgbm1 libglib2.0-0 libnspr4 libnss3 libpango-1.0-0 libx11-6 libxcb1 libxcomposite1 \
+            libxdamage1 libxext6 libxfixes3 libxkbcommon0 libxrandr2 \
+            # https://github.com/CloakHQ/CloakBrowser/tree/main#font-setup-on-linux
+            fonts-noto-color-emoji fonts-freefont-ttf fonts-unifont \
+            fonts-ipafont-gothic fonts-wqy-zenhei fonts-tlwg-loma-otf \
+        ; \
     fi; \
     rm -rf /var/lib/apt/lists/*
 
-COPY --from=chromium-downloader /app/node_modules/.cache/ms-playwright /app/node_modules/.cache/ms-playwright
+ENV CLOAKBROWSER_CACHE_DIR=/app/node_modules/.cache/cloakbrowser
+COPY --from=chromium-downloader /app/node_modules/.cache/cloakbrowser /app/node_modules/.cache/cloakbrowser
 
 RUN \
     set -ex && \
-    if [ "$PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD" = 0 ] && [ "$TARGETPLATFORM" = 'linux/amd64' ]; then \
+    if [ "$PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD" = 0 ] && [ "$CLOAKBROWSER_SKIP_DOWNLOAD" = 0 ]; then \
         echo 'Verifying Chromium installation...' && \
-        _chrome_path=$(find /app/node_modules/.cache/ms-playwright/ -name chrome -xtype f -executable | head -n1) && \
-        echo "CHROMIUM_EXECUTABLE_PATH=$_chrome_path" | tee /app/.env && \
+        _chrome_path=$(find /app/node_modules/.cache/cloakbrowser/ -name chrome -xtype f -executable | head -n1) && \
+        if [ -z "$_chrome_path" ]; then \
+            echo "!!! CloakBrowser binary not found !!!" && \
+            exit 1 ; \
+        fi; \
         if ldd "$_chrome_path" | grep "not found"; then \
-            echo "!!! Chromium has unmet shared libs !!!" && \
+            echo "!!! CloakBrowser has unmet shared libs !!!" && \
             exit 1 ; \
         else \
             echo "Awesome! All shared libs are met!" ; \

lib/bilibili-video-route.test.ts

@@ -11,15 +11,13 @@ const destroy = vi.fn();
 const getPlaywrightPage = vi.fn();
 const goto = vi.fn();
 const on = vi.fn();
-const setCookie = vi.fn();
-const setRequestInterception = vi.fn();
+const pageRoute = vi.fn();
 const waitForResponse = vi.fn();
 
 const page = {
     goto,
     on,
-    setCookie,
-    setRequestInterception,
+    route: pageRoute,
     waitForResponse,
 };
 
@@ -66,8 +64,7 @@ describe('/bilibili/user/video/:uid', () => {
         getPlaywrightPage.mockReset();
         goto.mockReset();
         on.mockReset();
-        setCookie.mockReset();
-        setRequestInterception.mockReset();
+        pageRoute.mockReset();
         waitForResponse.mockReset();
     });
 
@@ -109,6 +106,7 @@ describe('/bilibili/user/video/:uid', () => {
         getPlaywrightPage.mockImplementation(async (_url, options) => {
             await options.onBeforeLoad?.(page);
             return {
+                context: {},
                 destroy,
                 page,
             };
@@ -171,6 +169,7 @@ describe('/bilibili/user/video/:uid', () => {
         getPlaywrightPage.mockImplementation(async (_url, options) => {
             await options.onBeforeLoad?.(page);
             return {
+                context: {},
                 destroy,
                 page,
             };

lib/index.ts

@@ -8,6 +8,7 @@ import app from '@/app';
 import { config } from '@/config';
 import { getLocalhostAddress } from '@/utils/common-utils';
 import logger from '@/utils/logger';
+import { checkChromium } from '@/utils/playwright-utils';
 
 const port = config.connect.port;
 const hostIPList = getLocalhostAddress();
@@ -30,6 +31,8 @@ if (config.enableCluster) {
         for (let i = 0; i < numCPUs; i++) {
             cluster.fork();
         }
+
+        void checkChromium();
     } else {
         logger.info(`Worker ${process.pid} is running`);
         serve({
@@ -58,6 +61,8 @@ if (config.enableCluster) {
             maxHeaderSize: 1024 * 32,
         },
     });
+
+    void checkChromium();
 }
 
 export default server;

lib/routes/acs/journal.tsx

@@ -28,14 +28,14 @@ async function handler(ctx) {
 
     let title = '';
 
-    const browser = await playwright();
+    const context = await playwright();
     const items = await cache.tryGet(
         currentUrl,
         async () => {
-            const page = await browser.newPage();
-            await page.setRequestInterception(true);
-            page.on('request', (request) => {
-                request.resourceType() === 'document' || request.resourceType() === 'script' ? request.continue() : request.abort();
+            const page = await context.newPage();
+            await page.route('**/*', (route) => {
+                const request = route.request();
+                request.resourceType() === 'document' || request.resourceType() === 'script' ? route.continue() : route.abort();
             });
             await page.goto(currentUrl, {
                 waitUntil: 'domcontentloaded',
@@ -76,7 +76,7 @@ async function handler(ctx) {
         false
     );
 
-    await browser.close();
+    await context.close();
 
     return {
         title,

lib/routes/aip/journal-pupp.ts

@@ -18,12 +18,12 @@ const handler = async (ctx) => {
     }
 
     // use Playwright due to the obstacle by cloudflare challenge
-    const browser = await playwright();
+    const context = await playwright();
 
     const { jrnlName, list } = await cache.tryGet(
         jrnlUrl,
         async () => {
-            const response = await playwrightGet(jrnlUrl, browser);
+            const response = await playwrightGet(jrnlUrl, context);
             const $ = load(response);
             const jrnlName = $('.header-journal-title').text();
             const list = $('.card')
@@ -52,7 +52,7 @@ const handler = async (ctx) => {
         false
     );
 
-    await browser.close();
+    await context.close();
 
     return {
         title: jrnlName,

lib/routes/aip/utils.tsx

@@ -1,11 +1,11 @@
 import { renderToString } from 'hono/jsx/dom/server';
 
-const playwrightGet = async (url, browser) => {
-    const page = await browser.newPage();
+const playwrightGet = async (url, context) => {
+    const page = await context.newPage();
     // await page.setExtraHTTPHeaders({ referer: host });
-    await page.setRequestInterception(true);
-    page.on('request', (request) => {
-        request.resourceType() === 'document' ? request.continue() : request.abort();
+    await page.route('**/*', (route) => {
+        const request = route.request();
+        request.resourceType() === 'document' ? route.continue() : route.abort();
     });
     await page.goto(url, {
         waitUntil: 'domcontentloaded',

lib/routes/alternativeto/utils.ts

@@ -4,17 +4,17 @@ const baseURL = 'https://alternativeto.net';
 
 const playwrightGet = (url, cache) =>
     cache.tryGet(url, async () => {
-        const browser = await playwright();
-        const page = await browser.newPage();
-        await page.setRequestInterception(true);
-        page.on('request', (request) => {
-            request.resourceType() === 'document' ? request.continue() : request.abort();
+        const context = await playwright();
+        const page = await context.newPage();
+        await page.route('**/*', (route) => {
+            const request = route.request();
+            request.resourceType() === 'document' ? route.continue() : route.abort();
         });
         await page.goto(url, {
             waitUntil: 'domcontentloaded',
         });
         const html = await page.evaluate(() => document.documentElement.innerHTML);
-        await browser.close();
+        await context.close();
         return html;
     });
 

lib/routes/apkpure/versions.ts

@@ -28,19 +28,19 @@ async function handler(ctx) {
     const baseUrl = 'https://apkpure.com';
     const link = `${baseUrl}/${region}/${pkg}/versions`;
 
-    const browser = await playwright();
-    const page = await browser.newPage();
-    await page.setRequestInterception(true);
-    page.on('request', (request) => {
-        request.resourceType() === 'document' ? request.continue() : request.abort();
+    const context = await playwright();
+    const page = await context.newPage();
+    await page.route('**/*', (route) => {
+        const request = route.request();
+        request.resourceType() === 'document' ? route.continue() : route.abort();
     });
     logger.http(`Requesting ${link}`);
     await page.goto(link, {
         waitUntil: 'domcontentloaded',
     });
 
     const r = await page.evaluate(() => document.documentElement.innerHTML);
-    await browser.close();
+    await context.close();
 
     const $ = load(r);
     const img = new URL($('.ver-top img').attr('src'));

lib/routes/bilibili/cache.ts

@@ -51,7 +51,7 @@ const getCookie = (disableConfig = false) => {
                 waitForRequest = new Promise<string>((resolve) => {
                     page.on('requestfinished', async (request) => {
                         if (request.url() === 'https://api.bilibili.com/x/web-interface/nav') {
-                            const cookies = await page.cookies();
+                            const cookies = await page.context().cookies();
                             let cookieString = cookies.map((cookie) => `${cookie.name}=${cookie.value}`).join('; ');
                             cookieString = cookieString.replace(/b_lsid=[0-9A-F]+_[0-9A-F]+/, `b_lsid=${utils.lsid()}`);
                             resolve(cookieString);

lib/routes/bilibili/video.ts

@@ -140,7 +140,7 @@ async function applyCookie(page: Page, cookie: string) {
         .filter((item) => item !== undefined);
 
     if (cookies.length > 0) {
-        await page.setCookie(...cookies);
+        await page.context().addCookies(cookies);
     }
 }
 
@@ -184,9 +184,9 @@ async function fetchVideoListFromBrowser(uid: string): Promise<VideoListData> {
                 await applyCookie(page, cookie);
             }
 
-            await page.setRequestInterception(true);
-            page.on('request', (request) => {
-                allowedBrowserRequestTypes.has(request.resourceType()) ? request.continue() : request.abort();
+            await page.route('**/*', (route) => {
+                const request = route.request();
+                allowedBrowserRequestTypes.has(request.resourceType()) ? route.continue() : route.abort();
             });
         },
         gotoConfig: { waitUntil: 'domcontentloaded' },