apify · nikitachapovskii-dev · Jun 11, 2026 · Jun 11, 2026 · Jun 15, 2026 · Jun 15, 2026
diff --git a/.dockerignore b/.dockerignore
@@ -0,0 +1,14 @@
+# Actor images build from the monorepo root (dockerContextDir in .actor/actor.json).
+# Keep the context lean: ship sources + manifests + the pnpm lockfile only.
+node_modules
+**/node_modules
+.git
+.turbo
+**/dist
+**/storage
+apify_storage
+crawlee_storage
+*.log
+.DS_Store
+.idea
+.vscode
diff --git a/.github/workflows/on-pull-request.yml b/.github/workflows/on-pull-request.yml
@@ -13,6 +13,7 @@ jobs:
         name: Build & Test
         if: (!contains(github.event.head_commit.message, '[skip ci]') && !contains(github.event.head_commit.message, 'docs:'))
         runs-on: ubuntu-22.04
+        timeout-minutes: 30
 
         steps:
             - name: Checkout repository
@@ -38,6 +39,7 @@ jobs:
     lint:
         name: Lint
         runs-on: ubuntu-22.04
+        timeout-minutes: 30
         steps:
             - name: Checkout repository
               uses: actions/checkout@v6

diff --git a/.github/workflows/release-generic-actors.yaml b/.github/workflows/release-generic-actors.yaml
@@ -34,6 +34,11 @@ on:
                 type: boolean
                 required: false
                 default: true
+            sitemap-scraper:
+                description: apify/sitemap-extractor
+                type: boolean
+                required: false
+                default: true
             build-channel:
                 description: Build channel
                 type: choice
@@ -95,6 +100,13 @@ jobs:
                       development-version: '0.0'
                       development-build-tag: development
                       should-build: ${{ github.event.inputs.camoufox-scraper }}
+                    # Platform actor is apify/sitemap-extractor (name in .actor/actor.json)
+                    - actor: sitemap-scraper
+                      stable-version: '0.1'
+                      stable-build-tag: latest
+                      development-version: '0.0'
+                      development-build-tag: development
+                      should-build: ${{ github.event.inputs.sitemap-scraper }}
         steps:
             - uses: actions/checkout@v6
 

diff --git a/.github/workflows/test-e2e.yaml b/.github/workflows/test-e2e.yaml
@@ -13,6 +13,7 @@ jobs:
     build_and_test:
         name: Build & Test
         runs-on: ubuntu-22.04
+        timeout-minutes: 60
 
         steps:
             - name: Cancel Workflow Action
@@ -40,6 +41,11 @@ jobs:
             - name: Install pnpm and dependencies
               uses: apify/actions/pnpm-install@v1.1.2
 
+            - name: Install browsers
+              run: |
+                  pnpm exec puppeteer browsers install chrome
+                  pnpm exec playwright install chromium
+
             - name: Build
               run: pnpm ci:build
 

diff --git a/.gitignore b/.gitignore
@@ -18,6 +18,7 @@ pids
 .vscode
 yarn.lock
 .yarn
+# npm locks are generated locally in actor dirs (the monorepo uses pnpm-lock.yaml)
 tmp
 jsconfig.json
 types

diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
@@ -14,14 +14,14 @@ Please provide steps to reproduce if you found a bug or ideally fork the reposit
 
 Before you submit your pull request, consider the following guidelines:
 
-- Search [GitHub](https://github.com/apify/apify-sdk-js/pulls) for an open or closed PR that relates to your submission. You don't want to duplicate effort.
+- Search [GitHub](https://github.com/apify/actor-scraper/pulls) for an open or closed PR that relates to your submission. You don't want to duplicate effort.
 
-- Fork the project and install NPM dependencies.
+- Fork the project and install dependencies with pnpm.
 
 - Run tests before you start working, to be sure they all pass and your setup is working correctly:
 
     ```sh
-    npm run test
+    pnpm test
     ```
 
 - Be sure to **include appropriate test cases**. Tests help make it clear what the PR is fixing and also make sure the changes won't break over time.

diff --git a/package.json b/package.json
@@ -1,7 +1,7 @@
 {
-    "name": "apify-sdk-js",
+    "name": "actor-scraper",
     "private": true,
-    "description": "Apify SDK monorepo",
+    "description": "Apify generic scrapers monorepo",
     "keywords": [
         "apify",
         "headless",
@@ -23,12 +23,12 @@
     "license": "Apache-2.0",
     "repository": {
         "type": "git",
-        "url": "git+https://github.com/apify/apify-ts"
+        "url": "git+https://github.com/apify/actor-scraper.git"
     },
     "bugs": {
-        "url": "https://github.com/apify/apify-ts/issues"
+        "url": "https://github.com/apify/actor-scraper/issues"
     },
-    "homepage": "https://sdk.apify.com",
+    "homepage": "https://github.com/apify/actor-scraper",
     "scripts": {
         "prepare": "husky",
         "prepublishOnly": "turbo run copy",
@@ -61,7 +61,7 @@
         "@apify/tsconfig": "^0.1.2",
         "@commitlint/config-conventional": "^20.0.0",
         "@isaacs/brace-expansion": "^5.0.1",
-        "@playwright/browser-chromium": "^1.46.0",
+        "@playwright/browser-chromium": "^1.61.0",
         "@types/content-type": "^1.1.8",
         "@types/fs-extra": "^11.0.4",
         "@types/node": "^24.0.0",
@@ -81,8 +81,8 @@
         "oxfmt": "0.46.0",
         "oxlint": "1.62.0",
         "oxlint-tsgolint": "0.22.0",
-        "playwright": "^1.46.0",
-        "puppeteer": "^24.0.0",
+        "playwright": "^1.61.0",
+        "puppeteer": "^25.1.0",
         "rimraf": "^6.0.1",
         "tsx": "^4.16.5",
         "turbo": "2.9.1",

diff --git a/packages/actor-scraper/camoufox-scraper/.actor/actor.json b/packages/actor-scraper/camoufox-scraper/.actor/actor.json
@@ -3,6 +3,8 @@
     "name": "camoufox-scraper",
     "version": "0.1",
     "buildTag": "latest",
+    "dockerContextDir": "../../../..",
+    "dockerfile": "../Dockerfile",
     "storages": {
         "dataset": {
             "actorSpecification": 1,

diff --git a/packages/actor-scraper/camoufox-scraper/.dockerignore b/packages/actor-scraper/camoufox-scraper/.dockerignore
@@ -10,3 +10,5 @@ dist
 
 # installed files
 node_modules
+
+# local npm lock is a workspace-link lock; it breaks the isolated build (the platform builds without it)
diff --git a/packages/actor-scraper/camoufox-scraper/Dockerfile b/packages/actor-scraper/camoufox-scraper/Dockerfile
@@ -1,31 +1,40 @@
+# Build context is the monorepo root (set via dockerContextDir in .actor/actor.json),
+# so the build can use the pnpm workspace lockfile for a deterministic install.
 FROM apify/actor-node-playwright-camoufox:22 AS builder
 
-COPY --chown=myuser package*.json ./
+# The browser base runs as `myuser`; the build stage needs root for corepack + a writable workdir.
+USER root
+WORKDIR /app
 
-RUN npm install --include=dev --audit=false
+RUN corepack enable
 
-COPY --chown=myuser . ./
+# Browsers ship with the base image; never let an npm postinstall download them.
+# puppeteer is pulled transitively via @crawlee/browser-pool even for non-puppeteer actors.
+ENV PUPPETEER_SKIP_DOWNLOAD=true \
+    PLAYWRIGHT_SKIP_BROWSER_DOWNLOAD=1
 
-RUN npm run build
+# Whole workspace (root .dockerignore keeps node_modules/.git/dist out of the context).
+COPY . ./
 
-FROM apify/actor-node-playwright-camoufox:22
+# Deterministic install (frozen lockfile, honors minimumReleaseAge) of the actor + its workspace deps.
+RUN pnpm install --frozen-lockfile --filter actor-camoufox-scraper...
 
-COPY --from=builder --chown=myuser /home/myuser/dist ./dist
+# Build the actor and its workspace dependency @apify/scraper-tools.
+RUN pnpm --filter actor-camoufox-scraper... build
 
-COPY --chown=myuser package*.json ./
+# Self-contained production bundle; inject-workspace-packages copies the built
+# @apify/scraper-tools (a runtime workspace dep) into node_modules instead of symlinking it.
+RUN pnpm config set inject-workspace-packages true \
+    && pnpm --filter actor-camoufox-scraper deploy --prod /deploy
 
-RUN npm --quiet set progress=false \
-    && npm install --omit=dev \
-    && echo "Installed NPM packages:" \
-    && (npm list --omit=dev --all || true) \
-    && echo "Node.js version:" \
-    && node --version \
-    && echo "NPM version:" \
-    && npm --version \
-    && rm -r ~/.npm
+FROM apify/actor-node-playwright-camoufox:22
 
-COPY --chown=myuser . ./
+# Base WORKDIR is /home/myuser and ships a template node_modules plus the Xvfb entrypoint.
+# Drop the template's node_modules and overlay only the lean production bundle; the inherited
+# ENTRYPOINT (xvfb-entrypoint.sh) still wraps CMD with Xvfb, and the bundled browser is reused.
+RUN rm -rf node_modules
+COPY --from=builder --chown=myuser /deploy ./
 
 ENV APIFY_DISABLE_OUTDATED_WARNING=1
 
-CMD ./start_xvfb_and_run_cmd.sh && npm run start:prod --silent
+CMD ["node", "dist/main.js"]
diff --git a/packages/actor-scraper/camoufox-scraper/package.json b/packages/actor-scraper/camoufox-scraper/package.json
@@ -28,13 +28,13 @@
     },
     "repository": {
         "type": "git",
-        "url": "https://github.com/apify/apify-sdk-js"
+        "url": "https://github.com/apify/actor-scraper"
     },
     "author": {
         "name": "Apify Technologies",
         "email": "support@apify.com",
         "url": "https://apify.com"
     },
     "license": "Apache-2.0",
-    "homepage": "https://github.com/apify/apify-sdk-js"
+    "homepage": "https://github.com/apify/actor-scraper"
 }
diff --git a/packages/actor-scraper/cheerio-scraper/.actor/actor.json b/packages/actor-scraper/cheerio-scraper/.actor/actor.json
@@ -3,6 +3,8 @@
     "name": "cheerio-scraper",
     "version": "0.1",
     "buildTag": "latest",
+    "dockerContextDir": "../../../..",
+    "dockerfile": "../Dockerfile",
     "storages": {
         "dataset": {
             "actorSpecification": 1,

diff --git a/packages/actor-scraper/cheerio-scraper/.dockerignore b/packages/actor-scraper/cheerio-scraper/.dockerignore
@@ -8,3 +8,5 @@ storage
 
 # installed files
 node_modules
+
+# local npm lock is a workspace-link lock; it breaks the isolated build (the platform builds without it)
diff --git a/packages/actor-scraper/cheerio-scraper/Dockerfile b/packages/actor-scraper/cheerio-scraper/Dockerfile
@@ -1,32 +1,33 @@
+# Build context is the monorepo root (set via dockerContextDir in .actor/actor.json),
+# so the build can use the pnpm workspace lockfile for a deterministic install.
 FROM apify/actor-node:22 AS builder
 
-COPY package*.json ./
+WORKDIR /app
 
-RUN npm install --include=dev --audit=false
+RUN corepack enable
 
+# Whole workspace (root .dockerignore keeps node_modules/.git/dist out of the context).
 COPY . ./
 
-RUN npm run build
+# Deterministic install (frozen lockfile, honors minimumReleaseAge) of the actor + its workspace deps.
+RUN pnpm install --frozen-lockfile --filter actor-cheerio-scraper...
 
-FROM apify/actor-node:22
+# Build the actor and its workspace dependency @apify/scraper-tools.
+RUN pnpm --filter actor-cheerio-scraper... build
 
-COPY --from=builder /usr/src/app/dist ./dist
+# Self-contained production bundle. inject-workspace-packages copies the built
+# @apify/scraper-tools (a runtime workspace dep) into node_modules instead of symlinking it.
+RUN pnpm config set inject-workspace-packages true \
+    && pnpm --filter actor-cheerio-scraper deploy --prod /deploy
 
-COPY package*.json ./
+FROM apify/actor-node:22
 
-RUN rm -rf node_modules \
-    && npm --quiet set progress=false \
-    && npm install --omit=dev --omit=optional \
-    && echo "Installed NPM packages:" \
-    && (npm list --omit=dev --all || true) \
-    && echo "Node.js version:" \
-    && node --version \
-    && echo "NPM version:" \
-    && npm --version \
-    && rm -r ~/.npm
+WORKDIR /usr/src/app
 
-COPY . ./
+# Replace the base image's template with only the lean production bundle.
+RUN find /usr/src/app -mindepth 1 -delete
+COPY --from=builder /deploy ./
 
 ENV APIFY_DISABLE_OUTDATED_WARNING=1
 
-CMD npm run start:prod --silent
+CMD ["node", "dist/main.js"]
diff --git a/packages/actor-scraper/cheerio-scraper/README.md b/packages/actor-scraper/cheerio-scraper/README.md
@@ -183,7 +183,7 @@ async function pageFunction(context) {
 }
 ```
 
-The code runs in [Node.js 16](https://nodejs.org/) and the function accepts a single argument, the `context` object, whose properties are listed below.
+The code runs in [Node.js 22](https://nodejs.org/) and the function accepts a single argument, the `context` object, whose properties are listed below.
 
 The return value of the page function is an object (or an array of objects) representing the data extracted from the web page. The return value must be stringify-able to JSON, i.e. it can only contain basic types and no circular references. If you prefer not to extract any data from the page and skip it in the clean results, simply return `null` or `undefined`.
 

diff --git a/packages/actor-scraper/cheerio-scraper/package.json b/packages/actor-scraper/cheerio-scraper/package.json
@@ -26,7 +26,7 @@
     },
     "repository": {
         "type": "git",
-        "url": "https://github.com/apify/apify-sdk-js"
+        "url": "https://github.com/apify/actor-scraper"
     },
     "author": {
         "name": "Apify Technologies",
@@ -38,5 +38,5 @@
         "Ondra Urban <ondra@apify.com>"
     ],
     "license": "Apache-2.0",
-    "homepage": "https://github.com/apify/apify-sdk-js"
+    "homepage": "https://github.com/apify/actor-scraper"
 }
diff --git a/packages/actor-scraper/jsdom-scraper/.actor/actor.json b/packages/actor-scraper/jsdom-scraper/.actor/actor.json
@@ -3,6 +3,8 @@
     "name": "jsdom-scraper",
     "version": "0.1",
     "buildTag": "latest",
+    "dockerContextDir": "../../../..",
+    "dockerfile": "../Dockerfile",
     "storages": {
         "dataset": {
             "actorSpecification": 1,

diff --git a/packages/actor-scraper/jsdom-scraper/.dockerignore b/packages/actor-scraper/jsdom-scraper/.dockerignore
@@ -8,3 +8,5 @@ storage
 
 # installed files
 node_modules
+
+# local npm lock is a workspace-link lock; it breaks the isolated build (the platform builds without it)
diff --git a/packages/actor-scraper/jsdom-scraper/Dockerfile b/packages/actor-scraper/jsdom-scraper/Dockerfile
@@ -1,29 +1,33 @@
+# Build context is the monorepo root (set via dockerContextDir in .actor/actor.json),
+# so the build can use the pnpm workspace lockfile for a deterministic install.
 FROM apify/actor-node:22 AS builder
 
-COPY package*.json ./
+WORKDIR /app
 
-RUN npm install --include=dev --audit=false
+RUN corepack enable
 
+# Whole workspace (root .dockerignore keeps node_modules/.git/dist out of the context).
 COPY . ./
 
-RUN npm run build
+# Deterministic install (frozen lockfile, honors minimumReleaseAge) of the actor + its workspace deps.
+RUN pnpm install --frozen-lockfile --filter actor-jsdom-scraper...
 
-FROM apify/actor-node:22
+# Build the actor and its workspace dependency @apify/scraper-tools.
+RUN pnpm --filter actor-jsdom-scraper... build
+
+# Self-contained production bundle; inject-workspace-packages copies the built
+# @apify/scraper-tools (a runtime workspace dep) into node_modules instead of symlinking it.
+RUN pnpm config set inject-workspace-packages true \
+    && pnpm --filter actor-jsdom-scraper deploy --prod /deploy
 
-COPY --from=builder /usr/src/app/dist ./dist
+FROM apify/actor-node:22
 
-COPY package*.json ./
+WORKDIR /usr/src/app
 
-RUN npm --quiet set progress=false \
-    && npm install --omit=dev --omit=optional \
-    && echo "Installed NPM packages:" \
-    && (npm list --omit=dev --all || true) \
-    && echo "Node.js version:" \
-    && node --version \
-    && echo "NPM version:" \
-    && npm --version \
-    && rm -r ~/.npm
+# Replace the base image's template with only the lean production bundle.
+RUN find /usr/src/app -mindepth 1 -delete
+COPY --from=builder /deploy ./
 
-COPY . ./
+ENV APIFY_DISABLE_OUTDATED_WARNING=1
 
-CMD npm run start:prod --silent
+CMD ["node", "dist/main.js"]
-Original file line number
+Diff line change
@@ Expand Up / @@ -18,6 +18,7 @@ pids @@
     .vscode
     yarn.lock
     .yarn
+    # npm locks are generated locally in actor dirs (the monorepo uses pnpm-lock.yaml)
     tmp
     jsconfig.json
     types
@@ Expand Down @@
Original file line number	Diff line number	Diff line change
Expand Up		@@ -10,3 +10,5 @@ dist

		# installed files
		node_modules

		# local npm lock is a workspace-link lock; it breaks the isolated build (the platform builds without it)
Original file line number	Diff line number	Diff line change
Expand Up		@@ -8,3 +8,5 @@ storage

		# installed files
		node_modules

		# local npm lock is a workspace-link lock; it breaks the isolated build (the platform builds without it)