Update .gitignore

Co-authored-by: ilike2burnthing <59480337+ilike2burnthing@users.noreply.github.com>

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -8,6 +8,13 @@ body:
     options:
       - label: I have checked the README
         required: true
+- type: checkboxes
+  attributes:
+    label: Have you followed our Troubleshooting?
+    description: Please follow our <a href="https://github.com/FlareSolverr/FlareSolverr/wiki/Troubleshooting">Troubleshooting</a>.
+    options:
+      - label: I have followed your Troubleshooting
+        required: true
 - type: checkboxes
   attributes:
     label: Is there already an issue for your problem?
@@ -32,7 +39,8 @@ body:
       - Operating system:
       - Are you using Docker: [yes/no]
       - FlareSolverr User-Agent (see log traces or / endpoint):
-      - Are you using a proxy or VPN: [yes/no]
+      - Are you using a VPN: [yes/no]
+      - Are you using a Proxy: [yes/no]
       - Are you using Captcha Solver: [yes/no]
       - If using captcha solver, which one:
       - URL to test this issue:

.github/workflows/autotag.yml

@@ -6,12 +6,12 @@ on:
       - "master"
 
 jobs:
-  build:
+  tag-release:
     runs-on: ubuntu-latest
     steps:
       -
         name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
       -
         name: Auto Tag
         uses: Klemensas/action-autotag@stable

.github/workflows/release-docker.yml

@@ -6,44 +6,44 @@ on:
       - 'v*.*.*'
 
 jobs:
-  build:
+  build-docker-images:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     steps:
       -
         name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
       -
         name: Downcase repo
         run: echo REPOSITORY=$(echo ${{ github.repository }} | tr '[:upper:]' '[:lower:]') >> $GITHUB_ENV
       -
         name: Docker meta
         id: docker_meta
-        uses: crazy-max/ghaction-docker-meta@v1
+        uses: crazy-max/ghaction-docker-meta@v3
         with:
           images: ${{ env.REPOSITORY }},ghcr.io/${{ env.REPOSITORY }}
           tag-sha: false
       -
         name: Set up QEMU
-        uses: docker/setup-qemu-action@v1.0.1
+        uses: docker/setup-qemu-action@v2
       -
         name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v1
+        uses: docker/setup-buildx-action@v2
       -
         name: Login to DockerHub
-        uses: docker/login-action@v1
+        uses: docker/login-action@v2
         with:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
       -
         name: Login to GitHub Container Registry
-        uses: docker/login-action@v1
+        uses: docker/login-action@v2
         with:
           registry: ghcr.io
           username: ${{ github.repository_owner }}
           password: ${{ secrets.GH_PAT }}
       -
         name: Build and push
-        uses: docker/build-push-action@v2
+        uses: docker/build-push-action@v3
         with:
           context: .
           file: ./Dockerfile

.github/workflows/release.yml

@@ -6,26 +6,15 @@ on:
       - 'v*.*.*'
 
 jobs:
-  build:
+  create-release:
     name: Create release
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # get all commits, branches and tags (required for the changelog)
 
-      - name: Setup Node
-        uses: actions/setup-node@v2
-        with:
-          node-version: '16'
-
-      - name: Build artifacts
-        run: |
-          npm install
-          npm run build
-          npm run package
-
       - name: Build changelog
         id: github_changelog
         run: |
@@ -47,9 +36,60 @@ jobs:
           draft: false
           prerelease: false
 
+  build-linux-package:
+    name: Build Linux binary
+    needs: create-release
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0 # get all commits, branches and tags (required for the changelog)
+
+      - name: Setup Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.11'
+
+      - name: Build artifacts
+        run: |
+          python -m pip install -r requirements.txt
+          python -m pip install pyinstaller==5.9.0
+          cd src
+          python build_package.py
+
       - name: Upload release artifacts
-        uses: alexellis/upload-assets@0.2.2
+        uses: alexellis/upload-assets@0.4.0
         env:
           GITHUB_TOKEN: ${{ secrets.GH_PAT }}
         with:
-          asset_paths: '["./bin/*.zip"]'
+          asset_paths: '["./dist/flaresolverr_*"]'
+
+  build-windows-package:
+    name: Build Windows binary
+    needs: create-release
+    runs-on: windows-2022
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0 # get all commits, branches and tags (required for the changelog)
+
+      - name: Setup Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.11'
+
+      - name: Build artifacts
+        run: |
+          python -m pip install -r requirements.txt
+          python -m pip install pyinstaller==5.9.0
+          cd src
+          python build_package.py
+
+      - name: Upload release artifacts
+        uses: alexellis/upload-assets@0.4.0
+        env:
+          GITHUB_TOKEN: ${{ secrets.GH_PAT }}
+        with:
+          asset_paths: '["./dist/flaresolverr_*"]'

.gitignore

@@ -25,6 +25,7 @@ __pycache__/
 build/
 develop-eggs/
 dist/
+dist_chrome/
 downloads/
 eggs/
 .eggs/
@@ -123,3 +124,6 @@ venv.bak/
 .mypy_cache/
 .dmypy.json
 dmypy.json
+
+# node
+node_modules/

CHANGELOG.md

@@ -1,5 +1,67 @@
 # Changelog
 
+## v3.2.2 (2023/07/16)
+
+* Workaround for updated 'verify your are human' check
+
+## v3.2.1 (2023/06/10)
+
+* Kill dead Chrome processes in Windows
+* Fix Chrome GL erros in ASUSTOR NAS
+
+## v3.2.0 (2023/05/23)
+
+* Support "proxy" param in requests and sessions
+* Support "cookies" param in requests
+* Fix Chromium exec permissions in Linux package
+* Update Python dependencies
+
+## v3.1.2 (2023/04/02)
+
+* Fix headless mode in macOS
+* Remove redundant artifact from Windows binary package
+* Bump Selenium dependency
+
+## v3.1.1 (2023/03/25)
+
+* Distribute binary executables in compressed package
+* Add icon for binary executable
+* Include information about supported architectures in the readme
+* Check Python version on start
+
+## v3.1.0 (2023/03/20)
+
+* Build binaries for Linux x64 and Windows x64
+* Sessions with auto-creation on fetch request and TTL
+* Fix error trace: Crash Reports/pending No such file or directory
+* Fix Waitress server error with asyncore_use_poll=true
+* Attempt to fix Docker ARM32 build
+* Print platform information on start up
+* Add Fairlane challenge selector
+* Update DDOS-GUARD title
+* Update dependencies
+
+## v3.0.4 (2023/03/07)
+
+* Click on the Cloudflare's 'Verify you are human' button if necessary
+
+## v3.0.3 (2023/03/06)
+
+* Update undetected_chromedriver version to 3.4.6
+
+## v3.0.2 (2023/01/08)
+
+* Detect Cloudflare blocked access
+* Check Chrome / Chromium web browser is installed correctly
+
+## v3.0.1 (2023/01/06)
+
+* Kill Chromium processes properly to avoid defunct/zombie processes
+* Update undetected-chromedriver
+* Disable Zygote sandbox in Chromium browser
+* Add more selectors to detect blocked access
+* Include procps (ps), curl and vim packages in the Docker image
+
 ## v3.0.0 (2023/01/04)
 
 * This is the first release of FlareSolverr v3. There are some breaking changes

Dockerfile

@@ -29,7 +29,8 @@ RUN dpkg -i /libgl1-mesa-dri.deb \
     && dpkg -i /adwaita-icon-theme.deb \
     # Install dependencies
     && apt-get update \
-    && apt-get install -y --no-install-recommends chromium chromium-common chromium-driver xvfb \
+    && apt-get install -y --no-install-recommends chromium chromium-common chromium-driver xvfb dumb-init \
+        procps curl vim xauth \
     # Remove temporary files and hardware decoding libraries
     && rm -rf /var/lib/apt/lists/* \
     && rm -f /usr/lib/x86_64-linux-gnu/libmfxhw* \
@@ -47,19 +48,30 @@ RUN pip install -r requirements.txt \
 
 USER flaresolverr
 
+RUN mkdir -p "/app/.config/chromium/Crash Reports/pending"
+
 COPY src .
 COPY package.json ../
 
 EXPOSE 8191
 
+# dumb-init avoids zombie chromium processes
+ENTRYPOINT ["/usr/bin/dumb-init", "--"]
+
 CMD ["/usr/local/bin/python", "-u", "/app/flaresolverr.py"]
 
 # Local build
-# docker build -t ngosang/flaresolverr:3.0.0 .
+# docker build -t ngosang/flaresolverr:3.2.0 .
-# docker run -p 8191:8191 ngosang/flaresolverr:3.0.0
+# docker run -p 8191:8191 ngosang/flaresolverr:3.2.0
 
 # Multi-arch build
 # docker run --rm --privileged multiarch/qemu-user-static --reset -p yes
 # docker buildx create --use
-# docker buildx build -t ngosang/flaresolverr:3.0.0 --platform linux/386,linux/amd64,linux/arm/v7,linux/arm64/v8 .
+# docker buildx build -t ngosang/flaresolverr:3.2.0 --platform linux/386,linux/amd64,linux/arm/v7,linux/arm64/v8 .
 #   add --push to publish in DockerHub
+
+# Test multi-arch build
+# docker run --rm --privileged multiarch/qemu-user-static --reset -p yes
+# docker buildx create --use
+# docker buildx build -t ngosang/flaresolverr:3.2.0 --platform linux/arm/v7 --load .
+# docker run -p 8191:8191 --platform linux/arm/v7 ngosang/flaresolverr:3.2.0

README.md

@@ -64,16 +64,20 @@ Remember to restart the Docker daemon and the container after the update.
 
 ### Precompiled binaries
 
+> **Warning**
+> Precompiled binaries are only available for x64 architecture. For other architectures see Docker images.
+
 This is the recommended way for Windows users.
-* Download the [FlareSolverr zip](https://github.com/FlareSolverr/FlareSolverr/releases) from the release's assets. It is available for Windows and Linux.
+* Download the [FlareSolverr executable](https://github.com/FlareSolverr/FlareSolverr/releases) from the release's page. It is available for Windows x64 and Linux x64.
-* Extract the zip file. FlareSolverr executable and firefox folder must be in the same directory.
 * Execute FlareSolverr binary. In the environment variables section you can find how to change the configuration.
 
 ### From source code
 
-This is the recommended way for macOS users and for developers.
+> **Warning**
-* Install [Python 3.10](https://www.python.org/downloads/).
+> Installing from source code only works for x64 architecture. For other architectures see Docker images.
-* Install [Chrome](https://www.google.com/intl/en_us/chrome/) or [Chromium](https://www.chromium.org/getting-involved/download-chromium/) web browser.
+
+* Install [Python 3.11](https://www.python.org/downloads/).
+* Install [Chrome](https://www.google.com/intl/en_us/chrome/) (all OS) or [Chromium](https://www.chromium.org/getting-involved/download-chromium/) (just Linux, it doesn't work in Windows) web browser.
 * (Only in Linux / macOS) Install [Xvfb](https://en.wikipedia.org/wiki/Xvfb) package.
 * Clone this repository and open a shell in that path.
 * Run `pip install -r requirements.txt` command to install FlareSolverr dependencies.
@@ -140,16 +144,18 @@ session. When you no longer need to use a session you should make sure to close
 
 #### + `request.get`
 
-| Parameter         | Notes                                                                                                                                                                                                                                                                                                                                        |
+| Parameter           | Notes                                                                                                                                                                                                                                                                                                                                        |
-|-------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+|---------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| url               | Mandatory                                                                                                                                                                                                                                                                                                                                    |
+| url                 | Mandatory                                                                                                                                                                                                                                                                                                                                    |
-| session           | Optional. Will send the request from and existing browser instance. If one is not sent it will create a temporary instance that will be destroyed immediately after the request is completed.                                                                                                                                                |
+| session             | Optional. Will send the request from and existing browser instance. If one is not sent it will create a temporary instance that will be destroyed immediately after the request is completed.                                                                                                                                                |
-| maxTimeout        | Optional, default value 60000. Max timeout to solve the challenge in milliseconds.                                                                                                                                                                                                                                                           |
+| session_ttl_minutes | Optional. FlareSolverr will automatically rotate expired sessions based on the TTL provided in minutes.                                                                                                                                                                                                                                      |
-| cookies           | Optional. Will be used by the headless browser. Follow [this](https://github.com/puppeteer/puppeteer/blob/v3.3.0/docs/api.md#pagesetcookiecookies) format.                                                                                                                                                                                   |
+| maxTimeout          | Optional, default value 60000. Max timeout to solve the challenge in milliseconds.                                                                                                                                                                                                                                                           |
-| returnOnlyCookies | Optional, default false. Only returns the cookies. Response data, headers and other parts of the response are removed.                                                                                                                                                                                                                       |
+| cookies             | Optional. Will be used by the headless browser. Eg: `"cookies": [{"name": "cookie1", "value": "value1"}, {"name": "cookie2", "value": "value2"}]`.                                                                                                                                                                                           |
-| proxy             | Optional, default disabled. Eg: `"proxy": {"url": "http://127.0.0.1:8888"}`. You must include the proxy schema in the URL: `http://`, `socks4://` or `socks5://`. Authorization (username/password) is not supported. (When the `session` parameter is set, the proxy is ignored; a session specific proxy can be set in `sessions.create`.) |
+| returnOnlyCookies   | Optional, default false. Only returns the cookies. Response data, headers and other parts of the response are removed.                                                                                                                                                                                                                       |
+| proxy               | Optional, default disabled. Eg: `"proxy": {"url": "http://127.0.0.1:8888"}`. You must include the proxy schema in the URL: `http://`, `socks4://` or `socks5://`. Authorization (username/password) is not supported. (When the `session` parameter is set, the proxy is ignored; a session specific proxy can be set in `sessions.create`.) |
 
-:warning: If you want to use Cloudflare clearance cookie in your scripts, make sure you use the FlareSolverr User-Agent too. If they don't match you will see the challenge.
+> **Warning**
+> If you want to use Cloudflare clearance cookie in your scripts, make sure you use the FlareSolverr User-Agent too. If they don't match you will see the challenge.
 
 Example response from running the `curl` above:
 
@@ -239,13 +245,14 @@ Environment variables are set differently depending on the operating system. Som
 
 ## Captcha Solvers
 
-:warning: At this time none of the captcha solvers work. You can check the status in the open issues. Any help is welcome.
+> **Warning**
+> At this time none of the captcha solvers work. You can check the status in the open issues. Any help is welcome.
 
 Sometimes CloudFlare not only gives mathematical computations and browser tests, sometimes they also require the user to
 solve a captcha.
 If this is the case, FlareSolverr will return the error `Captcha detected but no automatic solver is configured.`
 
-FlareSolverr can be customized to solve the captchas automatically by setting the environment variable `CAPTCHA_SOLVER`
+FlareSolverr can be customized to solve the CAPTCHA automatically by setting the environment variable `CAPTCHA_SOLVER`
 to the file name of one of the adapters inside the [/captcha](src/captcha) directory.
 
 ## Related projects

package.json

@@ -1,7 +1,7 @@
 {
   "name": "flaresolverr",
-  "version": "3.0.0",
+  "version": "3.2.2",
   "description": "Proxy server to bypass Cloudflare protection",
   "author": "Diego Heras (ngosang / ngosang@hotmail.es)",
   "license": "MIT"
-}
+}

requirements.txt

@@ -1,9 +1,12 @@
-bottle==0.12.23
+bottle==0.12.25
 waitress==2.1.2
-selenium==4.4.3
+selenium==4.9.1
 func-timeout==4.3.5
 # required by undetected_chromedriver
-requests==2.28.1
+requests==2.31.0
-websockets==10.3
+certifi==2023.5.7
+websockets==11.0.3
 # only required for linux
 xvfbwrapper==0.2.9
+# only required for windows
+pefile==2023.2.7

src/build_package.py

@@ -0,0 +1,108 @@
+import os
+import platform
+import shutil
+import subprocess
+import sys
+import zipfile
+
+import requests
+
+
+def clean_files():
+    try:
+        shutil.rmtree(os.path.join(os.path.dirname(os.path.abspath(__file__)), os.pardir, 'build'))
+    except Exception:
+        pass
+    try:
+        shutil.rmtree(os.path.join(os.path.dirname(os.path.abspath(__file__)), os.pardir, 'dist'))
+    except Exception:
+        pass
+    try:
+        shutil.rmtree(os.path.join(os.path.dirname(os.path.abspath(__file__)), os.pardir, 'dist_chrome'))
+    except Exception:
+        pass
+
+
+def download_chromium():
+    # https://commondatastorage.googleapis.com/chromium-browser-snapshots/index.html?prefix=Linux_x64/
+    revision = "1090006" if os.name == 'nt' else '1090007'
+    arch = 'Win' if os.name == 'nt' else 'Linux_x64'
+    dl_file = 'chrome-win' if os.name == 'nt' else 'chrome-linux'
+    dl_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), os.pardir, 'dist_chrome')
+    dl_path_folder = os.path.join(dl_path, dl_file)
+    dl_path_zip = dl_path_folder + '.zip'
+
+    # response = requests.get(
+    #     f'https://commondatastorage.googleapis.com/chromium-browser-snapshots/{arch}/LAST_CHANGE',
+    #     timeout=30)
+    # revision = response.text.strip()
+    print("Downloading revision: " + revision)
+
+    os.mkdir(dl_path)
+    with requests.get(
+            f'https://commondatastorage.googleapis.com/chromium-browser-snapshots/{arch}/{revision}/{dl_file}.zip',
+            stream=True) as r:
+        r.raise_for_status()
+        with open(dl_path_zip, 'wb') as f:
+            for chunk in r.iter_content(chunk_size=8192):
+                f.write(chunk)
+    print("File downloaded: " + dl_path_zip)
+    with zipfile.ZipFile(dl_path_zip, 'r') as zip_ref:
+        zip_ref.extractall(dl_path)
+    os.remove(dl_path_zip)
+
+    chrome_path = os.path.join(dl_path, "chrome")
+    shutil.move(dl_path_folder, chrome_path)
+    print("Extracted in: " + chrome_path)
+
+    if os.name != 'nt':
+        # Give executable permissions for *nix
+        # file * | grep executable | cut -d: -f1
+        print("Giving executable permissions...")
+        execs = ['chrome', 'chrome_crashpad_handler', 'chrome_sandbox', 'chrome-wrapper', 'nacl_helper',
+                 'nacl_helper_bootstrap', 'nacl_irt_x86_64.nexe', 'xdg-mime', 'xdg-settings']
+        for exec_file in execs:
+            exec_path = os.path.join(chrome_path, exec_file)
+            os.chmod(exec_path, 0o755)
+
+
+def run_pyinstaller():
+    sep = ';' if os.name == 'nt' else ':'
+    subprocess.check_call([sys.executable, "-m", "PyInstaller",
+                           "--icon", "resources/flaresolverr_logo.ico",
+                           "--add-data", f"package.json{sep}.",
+                           "--add-data", f"{os.path.join('dist_chrome', 'chrome')}{sep}chrome",
+                           os.path.join("src", "flaresolverr.py")],
+                          cwd=os.pardir)
+
+
+def compress_package():
+    dist_folder = os.path.join(os.path.dirname(os.path.abspath(__file__)), os.pardir, 'dist')
+    package_folder = os.path.join(dist_folder, 'package')
+    shutil.move(os.path.join(dist_folder, 'flaresolverr'), os.path.join(package_folder, 'flaresolverr'))
+    print("Package folder: " + package_folder)
+
+    compr_format = 'zip' if os.name == 'nt' else 'gztar'
+    compr_file_name = 'flaresolverr_windows_x64' if os.name == 'nt' else 'flaresolverr_linux_x64'
+    compr_file_path = os.path.join(dist_folder, compr_file_name)
+    shutil.make_archive(compr_file_path, compr_format, package_folder)
+    print("Compressed file path: " + compr_file_path)
+
+
+if __name__ == "__main__":
+    print("Building package...")
+    print("Platform: " + platform.platform())
+
+    print("Cleaning previous build...")
+    clean_files()
+
+    print("Downloading Chromium...")
+    download_chromium()
+
+    print("Building pyinstaller executable... ")
+    run_pyinstaller()
+
+    print("Compressing package... ")
+    compress_package()
+
+# NOTE: python -m pip install pyinstaller

src/dtos.py

@@ -33,6 +33,7 @@ class V1RequestBase(object):
     maxTimeout: int = None
     proxy: dict = None
     session: str = None
+    session_ttl_minutes: int = None
     headers: list = None  # deprecated v2.0.0, not used
     userAgent: str = None  # deprecated v2.0.0, not used
 
@@ -51,6 +52,8 @@ class V1ResponseBase(object):
     # V1ResponseBase
     status: str = None
     message: str = None
+    session: str = None
+    sessions: list[str] = None
     startTimestamp: int = None
     endTimestamp: int = None
     version: str = None

src/flaresolverr.py

@@ -3,11 +3,12 @@ import logging
 import os
 import sys
 
-from bottle import run, response, Bottle, request
+import certifi
+from bottle import run, response, Bottle, request, ServerAdapter
 
 from bottle_plugins.error_plugin import error_plugin
 from bottle_plugins.logger_plugin import logger_plugin
-from dtos import IndexResponse, V1RequestBase
+from dtos import V1RequestBase
 import flaresolverr_service
 import utils
 
@@ -60,6 +61,16 @@ def controller_v1():
 
 
 if __name__ == "__main__":
+    # check python version
+    if sys.version_info < (3, 9):
+        raise Exception("The Python version is less than 3.9, a version equal to or higher is required.")
+
+    # fix ssl certificates for compiled binaries
+    # https://github.com/pyinstaller/pyinstaller/issues/7229
+    # https://stackoverflow.com/questions/55736855/how-to-change-the-cafile-argument-in-the-ssl-module-in-python3
+    os.environ["REQUESTS_CA_BUNDLE"] = certifi.where()
+    os.environ["SSL_CERT_FILE"] = certifi.where()
+
     # validate configuration
     log_level = os.environ.get('LOG_LEVEL', 'info').upper()
     log_html = utils.get_config_log_html()
@@ -92,4 +103,10 @@ if __name__ == "__main__":
 
     # start webserver
     # default server 'wsgiref' does not support concurrent requests
-    run(app, host=server_host, port=server_port, quiet=True, server='waitress')
+    # https://github.com/FlareSolverr/FlareSolverr/issues/680
+    # https://github.com/Pylons/waitress/issues/31
+    class WaitressServerPoll(ServerAdapter):
+        def run(self, handler):
+            from waitress import serve
+            serve(handler, host=self.host, port=self.port, asyncore_use_poll=True)
+    run(app, host=server_host, port=server_port, quiet=True, server=WaitressServerPoll)

src/flaresolverr_service.py

@@ -1,42 +1,77 @@
 import logging
+import platform
+import sys
 import time
+from datetime import timedelta
 from urllib.parse import unquote
 
-from func_timeout import func_timeout, FunctionTimedOut
+from func_timeout import FunctionTimedOut, func_timeout
 from selenium.common import TimeoutException
 from selenium.webdriver.chrome.webdriver import WebDriver
 from selenium.webdriver.common.by import By
+from selenium.webdriver.support.expected_conditions import (
+    presence_of_element_located, staleness_of, title_is)
+from selenium.webdriver.common.action_chains import ActionChains
 from selenium.webdriver.support.wait import WebDriverWait
-from selenium.webdriver.support.expected_conditions import presence_of_element_located, staleness_of, title_is
 
-from dtos import V1RequestBase, V1ResponseBase, ChallengeResolutionT, ChallengeResolutionResultT, IndexResponse, \
-    HealthResponse, STATUS_OK, STATUS_ERROR
 import utils
+from dtos import (STATUS_ERROR, STATUS_OK, ChallengeResolutionResultT,
+                  ChallengeResolutionT, HealthResponse, IndexResponse,
+                  V1RequestBase, V1ResponseBase)
+from sessions import SessionsStorage
 
+ACCESS_DENIED_TITLES = [
+    # Cloudflare
+    'Access denied',
+    # Cloudflare http://bitturk.net/ Firefox
+    'Attention Required! | Cloudflare'
+]
 ACCESS_DENIED_SELECTORS = [
     # Cloudflare
-    'div.cf-error-title span.cf-code-label span'
+    'div.cf-error-title span.cf-code-label span',
+    # Cloudflare http://bitturk.net/ Firefox
+    '#cf-error-details div.cf-error-overview h1'
 ]
-CHALLENGE_TITLE = [
+CHALLENGE_TITLES = [
     # Cloudflare
     'Just a moment...',
     # DDoS-GUARD
-    'DDOS-GUARD',
+    'DDoS-Guard'
 ]
 CHALLENGE_SELECTORS = [
     # Cloudflare
     '#cf-challenge-running', '.ray_id', '.attack-box', '#cf-please-wait', '#challenge-spinner', '#trk_jschal_js',
     # Custom CloudFlare for EbookParadijs, Film-Paleis, MuziekFabriek and Puur-Hollands
-    'td.info #js_info'
+    'td.info #js_info',
+    # Fairlane / pararius.com
+    'div.vc div.text-box h2'
 ]
 SHORT_TIMEOUT = 10
+SESSIONS_STORAGE = SessionsStorage()
 
 
 def test_browser_installation():
     logging.info("Testing web browser installation...")
+    logging.info("Platform: " + platform.platform())
+
+    chrome_exe_path = utils.get_chrome_exe_path()
+    if chrome_exe_path is None:
+        logging.error("Chrome / Chromium web browser not installed!")
+        sys.exit(1)
+    else:
+        logging.info("Chrome / Chromium path: " + chrome_exe_path)
+
+    chrome_major_version = utils.get_chrome_major_version()
+    if chrome_major_version == '':
+        logging.error("Chrome / Chromium version not detected!")
+        sys.exit(1)
+    else:
+        logging.info("Chrome / Chromium major version: " + chrome_major_version)
+
+    logging.info("Launching web browser...")
     user_agent = utils.get_user_agent()
     logging.info("FlareSolverr User-Agent: " + user_agent)
-    logging.info("Test successful")
+    logging.info("Test successful!")
 
 
 def index_endpoint() -> IndexResponse:
@@ -90,11 +125,11 @@ def _controller_v1_handler(req: V1RequestBase) -> V1ResponseBase:
     # execute the command
     res: V1ResponseBase
     if req.cmd == 'sessions.create':
-        raise Exception("Not implemented yet.")
+        res = _cmd_sessions_create(req)
     elif req.cmd == 'sessions.list':
-        raise Exception("Not implemented yet.")
+        res = _cmd_sessions_list(req)
     elif req.cmd == 'sessions.destroy':
-        raise Exception("Not implemented yet.")
+        res = _cmd_sessions_destroy(req)
     elif req.cmd == 'request.get':
         res = _cmd_request_get(req)
     elif req.cmd == 'request.post':
@@ -141,19 +176,116 @@ def _cmd_request_post(req: V1RequestBase) -> V1ResponseBase:
     return res
 
 
+def _cmd_sessions_create(req: V1RequestBase) -> V1ResponseBase:
+    logging.debug("Creating new session...")
+
+    session, fresh = SESSIONS_STORAGE.create(session_id=req.session, proxy=req.proxy)
+    session_id = session.session_id
+
+    if not fresh:
+        return V1ResponseBase({
+            "status": STATUS_OK,
+            "message": "Session already exists.",
+            "session": session_id
+        })
+
+    return V1ResponseBase({
+        "status": STATUS_OK,
+        "message": "Session created successfully.",
+        "session": session_id
+    })
+
+
+def _cmd_sessions_list(req: V1RequestBase) -> V1ResponseBase:
+    session_ids = SESSIONS_STORAGE.session_ids()
+
+    return V1ResponseBase({
+        "status": STATUS_OK,
+        "message": "",
+        "sessions": session_ids
+    })
+
+
+def _cmd_sessions_destroy(req: V1RequestBase) -> V1ResponseBase:
+    session_id = req.session
+    existed = SESSIONS_STORAGE.destroy(session_id)
+
+    if not existed:
+        raise Exception("The session doesn't exist.")
+
+    return V1ResponseBase({
+        "status": STATUS_OK,
+        "message": "The session has been removed."
+    })
+
+
 def _resolve_challenge(req: V1RequestBase, method: str) -> ChallengeResolutionT:
     timeout = req.maxTimeout / 1000
     driver = None
     try:
-        driver = utils.get_webdriver()
+        if req.session:
+            session_id = req.session
+            ttl = timedelta(minutes=req.session_ttl_minutes) if req.session_ttl_minutes else None
+            session, fresh = SESSIONS_STORAGE.get(session_id, ttl)
+
+            if fresh:
+                logging.debug(f"new session created to perform the request (session_id={session_id})")
+            else:
+                logging.debug(f"existing session is used to perform the request (session_id={session_id}, "
+                              f"lifetime={str(session.lifetime())}, ttl={str(ttl)})")
+
+            driver = session.driver
+        else:
+            driver = utils.get_webdriver(req.proxy)
+            logging.debug('New instance of webdriver has been created to perform the request')
         return func_timeout(timeout, _evil_logic, (req, driver, method))
     except FunctionTimedOut:
         raise Exception(f'Error solving the challenge. Timeout after {timeout} seconds.')
     except Exception as e:
-        raise Exception('Error solving the challenge. ' + str(e))
+        raise Exception('Error solving the challenge. ' + str(e).replace('\n', '\\n'))
     finally:
-        if driver is not None:
+        if not req.session and driver is not None:
             driver.quit()
+            logging.debug('A used instance of webdriver has been destroyed')
+
+
+def click_verify(driver: WebDriver):
+    try:
+        logging.debug("Try to find the Cloudflare verify checkbox")
+        iframe = driver.find_element(By.XPATH, "//iframe[@title='Widget containing a Cloudflare security challenge']")
+        driver.switch_to.frame(iframe)
+        checkbox = driver.find_element(
+            by=By.XPATH,
+            value='//*[@id="cf-stage"]//label[@class="ctp-checkbox-label"]/input',
+        )
+        if checkbox:
+            actions = ActionChains(driver)
+            actions.move_to_element_with_offset(checkbox, 5, 7)
+            actions.click(checkbox)
+            actions.perform()
+            logging.debug("Cloudflare verify checkbox found and clicked")
+    except Exception:
+        logging.debug("Cloudflare verify checkbox not found on the page")
+    finally:
+        driver.switch_to.default_content()
+
+    try:
+        logging.debug("Try to find the Cloudflare 'Verify you are human' button")
+        button = driver.find_element(
+            by=By.XPATH,
+            value="//input[@type='button' and @value='Verify you are human']",
+        )
+        if button:
+            actions = ActionChains(driver)
+            actions.move_to_element_with_offset(button, 5, 7)
+            actions.click(button)
+            actions.perform()
+            logging.debug("The Cloudflare 'Verify you are human' button found and clicked")
+    except Exception as e:
+        logging.debug("The Cloudflare 'Verify you are human' button not found on the page")
+        # print(e)
+
+    time.sleep(2)
 
 
 def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> ChallengeResolutionT:
@@ -167,12 +299,30 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
         _post_request(req, driver)
     else:
         driver.get(req.url)
-    if utils.get_config_log_html():
+
-        logging.debug(f"Response HTML:\n{driver.page_source}")
+    # set cookies if required
+    if req.cookies is not None and len(req.cookies) > 0:
+        logging.debug(f'Setting cookies...')
+        for cookie in req.cookies:
+            driver.delete_cookie(cookie['name'])
+            driver.add_cookie(cookie)
+        # reload the page
+        if method == 'POST':
+            _post_request(req, driver)
+        else:
+            driver.get(req.url)
 
     # wait for the page
+    if utils.get_config_log_html():
+        logging.debug(f"Response HTML:\n{driver.page_source}")
     html_element = driver.find_element(By.TAG_NAME, "html")
+    page_title = driver.title
 
+    # find access denied titles
+    for title in ACCESS_DENIED_TITLES:
+        if title == page_title:
+            raise Exception('Cloudflare has blocked this request. '
+                            'Probably your IP is banned for this site, check in your web browser.')
     # find access denied selectors
     for selector in ACCESS_DENIED_SELECTORS:
         found_elements = driver.find_elements(By.CSS_SELECTOR, selector)
@@ -182,11 +332,10 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
 
     # find challenge by title
     challenge_found = False
-    page_title = driver.title
+    for title in CHALLENGE_TITLES:
-    for title in CHALLENGE_TITLE:
+        if title.lower() == page_title.lower():
-        if title == page_title:
             challenge_found = True
-            logging.info("Challenge detected. Title found: " + title)
+            logging.info("Challenge detected. Title found: " + page_title)
             break
     if not challenge_found:
         # find challenge by selectors
@@ -197,17 +346,19 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
                 logging.info("Challenge detected. Selector found: " + selector)
                 break
 
+    attempt = 0
     if challenge_found:
         while True:
             try:
-                # wait until the title change
+                attempt = attempt + 1
-                for title in CHALLENGE_TITLE:
+                # wait until the title changes
-                    logging.debug("Waiting for title: " + title)
+                for title in CHALLENGE_TITLES:
+                    logging.debug("Waiting for title (attempt " + str(attempt) + "): " + title)
                     WebDriverWait(driver, SHORT_TIMEOUT).until_not(title_is(title))
 
                 # then wait until all the selectors disappear
                 for selector in CHALLENGE_SELECTORS:
-                    logging.debug("Waiting for selector: " + selector)
+                    logging.debug("Waiting for selector (attempt " + str(attempt) + "): " + selector)
                     WebDriverWait(driver, SHORT_TIMEOUT).until_not(
                         presence_of_element_located((By.CSS_SELECTOR, selector)))
 
@@ -216,6 +367,9 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
 
             except TimeoutException:
                 logging.debug("Timeout waiting for selector")
+
+                click_verify(driver)
+
                 # update the html (cloudflare reloads the page every 5 s)
                 html_element = driver.find_element(By.TAG_NAME, "html")
 

src/sessions.py

@@ -0,0 +1,82 @@
+import logging
+from dataclasses import dataclass
+from datetime import datetime, timedelta
+from typing import Optional, Tuple
+from uuid import uuid1
+
+from selenium.webdriver.chrome.webdriver import WebDriver
+
+import utils
+
+
+@dataclass
+class Session:
+    session_id: str
+    driver: WebDriver
+    created_at: datetime
+
+    def lifetime(self) -> timedelta:
+        return datetime.now() - self.created_at
+
+
+class SessionsStorage:
+    """SessionsStorage creates, stores and process all the sessions"""
+
+    def __init__(self):
+        self.sessions = {}
+
+    def create(self, session_id: Optional[str] = None, proxy: Optional[dict] = None,
+               force_new: Optional[bool] = False) -> Tuple[Session, bool]:
+        """create creates new instance of WebDriver if necessary,
+        assign defined (or newly generated) session_id to the instance
+        and returns the session object. If a new session has been created
+        second argument is set to True.
+
+        Note: The function is idempotent, so in case if session_id
+        already exists in the storage a new instance of WebDriver won't be created
+        and existing session will be returned. Second argument defines if 
+        new session has been created (True) or an existing one was used (False).
+        """
+        session_id = session_id or str(uuid1())
+
+        if force_new:
+            self.destroy(session_id)
+
+        if self.exists(session_id):
+            return self.sessions[session_id], False
+
+        driver = utils.get_webdriver(proxy)
+        created_at = datetime.now()
+        session = Session(session_id, driver, created_at)
+
+        self.sessions[session_id] = session
+
+        return session, True
+
+    def exists(self, session_id: str) -> bool:
+        return session_id in self.sessions
+
+    def destroy(self, session_id: str) -> bool:
+        """destroy closes the driver instance and removes session from the storage.
+        The function is noop if session_id doesn't exist.
+        The function returns True if session was found and destroyed,
+        and False if session_id wasn't found.
+        """
+        if not self.exists(session_id):
+            return False
+
+        session = self.sessions.pop(session_id)
+        session.driver.quit()
+        return True
+
+    def get(self, session_id: str, ttl: Optional[timedelta] = None) -> Tuple[Session, bool]:
+        session, fresh = self.create(session_id)
+
+        if ttl is not None and not fresh and session.lifetime() > ttl:
+            logging.debug(f'session\'s lifetime has expired, so the session is recreated (session_id={session_id})')
+            session, fresh = self.create(session_id, force_new=True)
+
+        return session, fresh
+
+    def session_ids(self) -> list[str]:
+        return list(self.sessions.keys())

src/tests.py

@@ -1,4 +1,5 @@
 import unittest
+from typing import Optional
 
 from webtest import TestApp
 
@@ -7,7 +8,7 @@ import flaresolverr
 import utils
 
 
-def _find_obj_by_key(key: str, value: str, _list: list) -> dict | None:
+def _find_obj_by_key(key: str, value: str, _list: list) -> Optional[dict]:
     for obj in _list:
         if obj[key] == value:
             return obj
@@ -23,10 +24,13 @@ class TestFlareSolverr(unittest.TestCase):
     cloudflare_url = "https://nowsecure.nl"
     cloudflare_url_2 = "https://idope.se/torrent-list/harry/"
     ddos_guard_url = "https://anidex.info/"
+    fairlane_url = "https://www.pararius.com/apartments/amsterdam"
     custom_cloudflare_url = "https://www.muziekfabriek.org"
     cloudflare_blocked_url = "https://cpasbiens3.fr/index.php?do=search&subaction=search"
 
     app = TestApp(flaresolverr.app)
+    # wait until the server is ready
+    app.get('/')
 
     def test_wrong_endpoint(self):
         res = self.app.get('/wrong', status=404)
@@ -64,7 +68,7 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertEqual("Error: Request parameter 'cmd' = 'request.bad' is invalid.", body.message)
         self.assertGreater(body.startTimestamp, 10000)
         self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
-        self.assertEqual(utils.get_flaresolverr_version(), body.version) 
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
 
     def test_v1_endpoint_request_get_no_cloudflare(self):
         res = self.app.post_json('/v1', {
@@ -78,7 +82,7 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertEqual("Challenge not detected!", body.message)
         self.assertGreater(body.startTimestamp, 10000)
         self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
-        self.assertEqual(utils.get_flaresolverr_version(), body.version) 
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
 
         solution = body.solution
         self.assertIn(self.google_url, solution.url)
@@ -100,7 +104,7 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertEqual("Challenge solved!", body.message)
         self.assertGreater(body.startTimestamp, 10000)
         self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
-        self.assertEqual(utils.get_flaresolverr_version(), body.version) 
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
 
         solution = body.solution
         self.assertIn(self.cloudflare_url, solution.url)
@@ -126,7 +130,7 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertEqual("Challenge solved!", body.message)
         self.assertGreater(body.startTimestamp, 10000)
         self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
-        self.assertEqual(utils.get_flaresolverr_version(), body.version) 
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
 
         solution = body.solution
         self.assertIn(self.cloudflare_url_2, solution.url)
@@ -152,7 +156,7 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertEqual("Challenge solved!", body.message)
         self.assertGreater(body.startTimestamp, 10000)
         self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
-        self.assertEqual(utils.get_flaresolverr_version(), body.version) 
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
 
         solution = body.solution
         self.assertIn(self.ddos_guard_url, solution.url)
@@ -166,6 +170,32 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertIsNotNone(cf_cookie, "DDOS-Guard cookie not found")
         self.assertGreater(len(cf_cookie["value"]), 10)
 
+    def test_v1_endpoint_request_get_fairlane_js(self):
+        res = self.app.post_json('/v1', {
+            "cmd": "request.get",
+            "url": self.fairlane_url
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("Challenge solved!", body.message)
+        self.assertGreater(body.startTimestamp, 10000)
+        self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
+
+        solution = body.solution
+        self.assertIn(self.fairlane_url, solution.url)
+        self.assertEqual(solution.status, 200)
+        self.assertIs(len(solution.headers), 0)
+        self.assertIn("<title>Rental Apartments Amsterdam</title>", solution.response)
+        self.assertGreater(len(solution.cookies), 0)
+        self.assertIn("Chrome/", solution.userAgent)
+
+        cf_cookie = _find_obj_by_key("name", "fl_pass_v2_b", solution.cookies)
+        self.assertIsNotNone(cf_cookie, "Fairlane cookie not found")
+        self.assertGreater(len(cf_cookie["value"]), 50)
+
     def test_v1_endpoint_request_get_custom_cloudflare_js(self):
         res = self.app.post_json('/v1', {
             "cmd": "request.get",
@@ -178,7 +208,7 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertEqual("Challenge solved!", body.message)
         self.assertGreater(body.startTimestamp, 10000)
         self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
-        self.assertEqual(utils.get_flaresolverr_version(), body.version) 
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
 
         solution = body.solution
         self.assertIn(self.custom_cloudflare_url, solution.url)
@@ -209,7 +239,45 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
         self.assertEqual(utils.get_flaresolverr_version(), body.version)
 
-    # todo: test Cmd 'request.get' should return OK with 'cookies' param
+    def test_v1_endpoint_request_get_cookies_param(self):
+        res = self.app.post_json('/v1', {
+            "cmd": "request.get",
+            "url": self.google_url,
+            "cookies": [
+                {
+                    "name": "testcookie1",
+                    "value": "testvalue1"
+                },
+                {
+                    "name": "testcookie2",
+                    "value": "testvalue2"
+                }
+            ]
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("Challenge not detected!", body.message)
+        self.assertGreater(body.startTimestamp, 10000)
+        self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
+
+        solution = body.solution
+        self.assertIn(self.google_url, solution.url)
+        self.assertEqual(solution.status, 200)
+        self.assertIs(len(solution.headers), 0)
+        self.assertIn("<title>Google</title>", solution.response)
+        self.assertGreater(len(solution.cookies), 1)
+        self.assertIn("Chrome/", solution.userAgent)
+
+        user_cookie1 = _find_obj_by_key("name", "testcookie1", solution.cookies)
+        self.assertIsNotNone(user_cookie1, "User cookie 1 not found")
+        self.assertEqual("testvalue1", user_cookie1["value"])
+
+        user_cookie2 = _find_obj_by_key("name", "testcookie2", solution.cookies)
+        self.assertIsNotNone(user_cookie2, "User cookie 2 not found")
+        self.assertEqual("testvalue2", user_cookie2["value"])
 
     def test_v1_endpoint_request_get_returnOnlyCookies_param(self):
         res = self.app.post_json('/v1', {
@@ -234,10 +302,88 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertGreater(len(solution.cookies), 0)
         self.assertIn("Chrome/", solution.userAgent)
 
-    # todo: test Cmd 'request.get' should return OK with HTTP 'proxy' param
+    def test_v1_endpoint_request_get_proxy_http_param(self):
-    # todo: test Cmd 'request.get' should return OK with HTTP 'proxy' param with credentials
+        """
-    # todo: test Cmd 'request.get' should return OK with SOCKSv5 'proxy' param
+        To configure TinyProxy in local:
-    # todo: test Cmd 'request.get' should fail with wrong 'proxy' param
+           * sudo vim /etc/tinyproxy/tinyproxy.conf
+              * edit => LogFile "/tmp/tinyproxy.log"
+              * edit => Syslog Off
+           * sudo tinyproxy -d
+           * sudo tail -f /tmp/tinyproxy.log
+        """
+        res = self.app.post_json('/v1', {
+            "cmd": "request.get",
+            "url": self.google_url,
+            "proxy": {
+                "url": self.proxy_url
+            }
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("Challenge not detected!", body.message)
+        self.assertGreater(body.startTimestamp, 10000)
+        self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
+
+        solution = body.solution
+        self.assertIn(self.google_url, solution.url)
+        self.assertEqual(solution.status, 200)
+        self.assertIs(len(solution.headers), 0)
+        self.assertIn("<title>Google</title>", solution.response)
+        self.assertGreater(len(solution.cookies), 0)
+        self.assertIn("Chrome/", solution.userAgent)
+
+    def test_v1_endpoint_request_get_proxy_socks_param(self):
+        """
+        To configure Dante in local:
+           * https://linuxhint.com/set-up-a-socks5-proxy-on-ubuntu-with-dante/
+           * sudo vim /etc/sockd.conf
+           * sudo systemctl restart sockd.service
+           * curl --socks5 socks5://127.0.0.1:1080 https://www.google.com
+        """
+        res = self.app.post_json('/v1', {
+            "cmd": "request.get",
+            "url": self.google_url,
+            "proxy": {
+                "url": self.proxy_socks_url
+            }
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("Challenge not detected!", body.message)
+        self.assertGreater(body.startTimestamp, 10000)
+        self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
+
+        solution = body.solution
+        self.assertIn(self.google_url, solution.url)
+        self.assertEqual(solution.status, 200)
+        self.assertIs(len(solution.headers), 0)
+        self.assertIn("<title>Google</title>", solution.response)
+        self.assertGreater(len(solution.cookies), 0)
+        self.assertIn("Chrome/", solution.userAgent)
+
+    def test_v1_endpoint_request_get_proxy_wrong_param(self):
+        res = self.app.post_json('/v1', {
+            "cmd": "request.get",
+            "url": self.google_url,
+            "proxy": {
+                "url": "http://127.0.0.1:43210"
+            }
+        }, status=500)
+        self.assertEqual(res.status_code, 500)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_ERROR, body.status)
+        self.assertIn("Error: Error solving the challenge. Message: unknown error: net::ERR_PROXY_CONNECTION_FAILED",
+                      body.message)
+        self.assertGreater(body.startTimestamp, 10000)
+        self.assertGreaterEqual(body.endTimestamp, body.startTimestamp)
+        self.assertEqual(utils.get_flaresolverr_version(), body.version)
 
     def test_v1_endpoint_request_get_fail_timeout(self):
         res = self.app.post_json('/v1', {
@@ -351,12 +497,99 @@ class TestFlareSolverr(unittest.TestCase):
         self.assertEqual(STATUS_OK, body.status)
         self.assertEqual("Challenge not detected!", body.message)
 
-    # todo: test Cmd 'sessions.create' should return OK
+    def test_v1_endpoint_sessions_create_without_session(self):
-    # todo: test Cmd 'sessions.create' should return OK with session
+        res = self.app.post_json('/v1', {
-    # todo: test Cmd 'sessions.list' should return OK
+            "cmd": "sessions.create"
-    # todo: test Cmd 'sessions.destroy' should return OK
+        })
-    # todo: test Cmd 'sessions.destroy' should fail
+        self.assertEqual(res.status_code, 200)
-    # todo: test Cmd 'request.get' should use session
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("Session created successfully.", body.message)
+        self.assertIsNotNone(body.session)
+
+    def test_v1_endpoint_sessions_create_with_session(self):
+        res = self.app.post_json('/v1', {
+            "cmd": "sessions.create",
+            "session": "test_create_session"
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("Session created successfully.", body.message)
+        self.assertEqual(body.session, "test_create_session")
+
+    def test_v1_endpoint_sessions_create_with_proxy(self):
+        res = self.app.post_json('/v1', {
+            "cmd": "sessions.create",
+            "proxy": {
+                "url": self.proxy_url
+            }
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("Session created successfully.", body.message)
+        self.assertIsNotNone(body.session)
+
+    def test_v1_endpoint_sessions_list(self):
+        self.app.post_json('/v1', {
+            "cmd": "sessions.create",
+            "session": "test_list_sessions"
+        })
+        res = self.app.post_json('/v1', {
+            "cmd": "sessions.list"
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("", body.message)
+        self.assertGreaterEqual(len(body.sessions), 1)
+        self.assertIn("test_list_sessions", body.sessions)
+
+    def test_v1_endpoint_sessions_destroy_existing_session(self):
+        self.app.post_json('/v1', {
+            "cmd": "sessions.create",
+            "session": "test_destroy_sessions"
+        })
+        res = self.app.post_json('/v1', {
+            "cmd": "sessions.destroy",
+            "session": "test_destroy_sessions"
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
+        self.assertEqual("The session has been removed.", body.message)
+
+    def test_v1_endpoint_sessions_destroy_non_existing_session(self):
+        res = self.app.post_json('/v1', {
+            "cmd": "sessions.destroy",
+            "session": "non_existing_session_name"
+        }, status=500)
+        self.assertEqual(res.status_code, 500)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_ERROR, body.status)
+        self.assertEqual("Error: The session doesn't exist.", body.message)
+
+    def test_v1_endpoint_request_get_with_session(self):
+        self.app.post_json('/v1', {
+            "cmd": "sessions.create",
+            "session": "test_request_sessions"
+        })
+        res = self.app.post_json('/v1', {
+            "cmd": "request.get",
+            "session": "test_request_sessions",
+            "url": self.google_url
+        })
+        self.assertEqual(res.status_code, 200)
+
+        body = V1ResponseBase(res.json)
+        self.assertEqual(STATUS_OK, body.status)
 
 
 if __name__ == '__main__':

src/tests_sites.py

@@ -39,6 +39,8 @@ def asset_cloudflare_solution(self, res, site_url, site_text):
 
 class TestFlareSolverr(unittest.TestCase):
     app = TestApp(flaresolverr.app)
+    # wait until the server is ready
+    app.get('/')
 
     def test_v1_endpoint_request_get_cloudflare(self):
         sites_get = [

src/undetected_chromedriver/_compat.py

@@ -1,259 +0,0 @@
-#!/usr/bin/env python3
-# this module is part of undetected_chromedriver
-
-
-"""
-
-         888                                                  888         d8b
-         888                                                  888         Y8P
-         888                                                  888
- .d8888b 88888b.  888d888 .d88b.  88888b.d88b.   .d88b.   .d88888 888d888 888 888  888  .d88b.  888d888
-d88P"    888 "88b 888P"  d88""88b 888 "888 "88b d8P  Y8b d88" 888 888P"   888 888  888 d8P  Y8b 888P"
-888      888  888 888    888  888 888  888  888 88888888 888  888 888     888 Y88  88P 88888888 888
-Y88b.    888  888 888    Y88..88P 888  888  888 Y8b.     Y88b 888 888     888  Y8bd8P  Y8b.     888
- "Y8888P 888  888 888     "Y88P"  888  888  888  "Y8888   "Y88888 888     888   Y88P    "Y8888  888   88888888
-
-by UltrafunkAmsterdam (https://github.com/ultrafunkamsterdam)
-
-"""
-
-import io
-import logging
-import os
-import random
-import re
-import string
-import sys
-import zipfile
-from distutils.version import LooseVersion
-from urllib.request import urlopen, urlretrieve
-
-from selenium.webdriver import Chrome as _Chrome, ChromeOptions as _ChromeOptions
-
-TARGET_VERSION = 0
-logger = logging.getLogger("uc")
-
-
-class Chrome:
-    def __new__(cls, *args, emulate_touch=False, **kwargs):
-
-        if not ChromeDriverManager.installed:
-            ChromeDriverManager(*args, **kwargs).install()
-        if not ChromeDriverManager.selenium_patched:
-            ChromeDriverManager(*args, **kwargs).patch_selenium_webdriver()
-        if not kwargs.get("executable_path"):
-            kwargs["executable_path"] = "./{}".format(
-                ChromeDriverManager(*args, **kwargs).executable_path
-            )
-        if not kwargs.get("options"):
-            kwargs["options"] = ChromeOptions()
-        instance = object.__new__(_Chrome)
-        instance.__init__(*args, **kwargs)
-
-        instance._orig_get = instance.get
-
-        def _get_wrapped(*args, **kwargs):
-            if instance.execute_script("return navigator.webdriver"):
-                instance.execute_cdp_cmd(
-                    "Page.addScriptToEvaluateOnNewDocument",
-                    {
-                        "source": """
-
-                                   Object.defineProperty(window, 'navigator', {
-                                       value: new Proxy(navigator, {
-                                       has: (target, key) => (key === 'webdriver' ? false : key in target),
-                                       get: (target, key) =>
-                                           key === 'webdriver'
-                                           ? undefined
-                                           : typeof target[key] === 'function'
-                                           ? target[key].bind(target)
-                                           : target[key]
-                                       })
-                                   });
-                                    
-                                                            
-                        """
-                    },
-                )
-            return instance._orig_get(*args, **kwargs)
-
-        instance.get = _get_wrapped
-        instance.get = _get_wrapped
-        instance.get = _get_wrapped
-
-        original_user_agent_string = instance.execute_script(
-            "return navigator.userAgent"
-        )
-        instance.execute_cdp_cmd(
-            "Network.setUserAgentOverride",
-            {
-                "userAgent": original_user_agent_string.replace("Headless", ""),
-            },
-        )
-        if emulate_touch:
-            instance.execute_cdp_cmd(
-                "Page.addScriptToEvaluateOnNewDocument",
-                {
-                    "source": """
-                                   Object.defineProperty(navigator, 'maxTouchPoints', {
-                                       get: () => 1
-                               })"""
-                },
-            )
-        logger.info(f"starting undetected_chromedriver.Chrome({args}, {kwargs})")
-        return instance
-
-
-class ChromeOptions:
-    def __new__(cls, *args, **kwargs):
-        if not ChromeDriverManager.installed:
-            ChromeDriverManager(*args, **kwargs).install()
-        if not ChromeDriverManager.selenium_patched:
-            ChromeDriverManager(*args, **kwargs).patch_selenium_webdriver()
-
-        instance = object.__new__(_ChromeOptions)
-        instance.__init__()
-        instance.add_argument("start-maximized")
-        instance.add_experimental_option("excludeSwitches", ["enable-automation"])
-        instance.add_argument("--disable-blink-features=AutomationControlled")
-        return instance
-
-
-class ChromeDriverManager(object):
-    installed = False
-    selenium_patched = False
-    target_version = None
-
-    DL_BASE = "https://chromedriver.storage.googleapis.com/"
-
-    def __init__(self, executable_path=None, target_version=None, *args, **kwargs):
-
-        _platform = sys.platform
-
-        if TARGET_VERSION:
-            # use global if set
-            self.target_version = TARGET_VERSION
-
-        if target_version:
-            # use explicitly passed target
-            self.target_version = target_version  # user override
-
-        if not self.target_version:
-            # none of the above (default) and just get current version
-            self.target_version = self.get_release_version_number().version[
-                0
-            ]  # only major version int
-
-        self._base = base_ = "chromedriver{}"
-
-        exe_name = self._base
-        if _platform in ("win32",):
-            exe_name = base_.format(".exe")
-        if _platform in ("linux",):
-            _platform += "64"
-            exe_name = exe_name.format("")
-        if _platform in ("darwin",):
-            _platform = "mac64"
-            exe_name = exe_name.format("")
-        self.platform = _platform
-        self.executable_path = executable_path or exe_name
-        self._exe_name = exe_name
-
-    def patch_selenium_webdriver(self_):
-        """
-        Patches selenium package Chrome, ChromeOptions classes for current session
-
-        :return:
-        """
-        import selenium.webdriver.chrome.service
-        import selenium.webdriver
-
-        selenium.webdriver.Chrome = Chrome
-        selenium.webdriver.ChromeOptions = ChromeOptions
-        logger.info("Selenium patched. Safe to import Chrome / ChromeOptions")
-        self_.__class__.selenium_patched = True
-
-    def install(self, patch_selenium=True):
-        """
-        Initialize the patch
-
-        This will:
-         download chromedriver if not present
-         patch the downloaded chromedriver
-         patch selenium package if <patch_selenium> is True (default)
-
-        :param patch_selenium: patch selenium webdriver classes for Chrome and ChromeDriver (for current python session)
-        :return:
-        """
-        if not os.path.exists(self.executable_path):
-            self.fetch_chromedriver()
-            if not self.__class__.installed:
-                if self.patch_binary():
-                    self.__class__.installed = True
-
-        if patch_selenium:
-            self.patch_selenium_webdriver()
-
-    def get_release_version_number(self):
-        """
-        Gets the latest major version available, or the latest major version of self.target_version if set explicitly.
-
-        :return: version string
-        """
-        path = (
-            "LATEST_RELEASE"
-            if not self.target_version
-            else f"LATEST_RELEASE_{self.target_version}"
-        )
-        return LooseVersion(urlopen(self.__class__.DL_BASE + path).read().decode())
-
-    def fetch_chromedriver(self):
-        """
-        Downloads ChromeDriver from source and unpacks the executable
-
-        :return: on success, name of the unpacked executable
-        """
-        base_ = self._base
-        zip_name = base_.format(".zip")
-        ver = self.get_release_version_number().vstring
-        if os.path.exists(self.executable_path):
-            return self.executable_path
-        urlretrieve(
-            f"{self.__class__.DL_BASE}{ver}/{base_.format(f'_{self.platform}')}.zip",
-            filename=zip_name,
-        )
-        with zipfile.ZipFile(zip_name) as zf:
-            zf.extract(self._exe_name)
-        os.remove(zip_name)
-        if sys.platform != "win32":
-            os.chmod(self._exe_name, 0o755)
-        return self._exe_name
-
-    @staticmethod
-    def random_cdc():
-        cdc = random.choices(string.ascii_lowercase, k=26)
-        cdc[-6:-4] = map(str.upper, cdc[-6:-4])
-        cdc[2] = cdc[0]
-        cdc[3] = "_"
-        return "".join(cdc).encode()
-
-    def patch_binary(self):
-        """
-        Patches the ChromeDriver binary
-
-        :return: False on failure, binary name on success
-        """
-        linect = 0
-        replacement = self.random_cdc()
-        with io.open(self.executable_path, "r+b") as fh:
-            for line in iter(lambda: fh.readline(), b""):
-                if b"cdc_" in line:
-                    fh.seek(-len(line), 1)
-                    newline = re.sub(b"cdc_.{22}", replacement, line)
-                    fh.write(newline)
-                    linect += 1
-            return linect
-
-
-def install(executable_path=None, target_version=None, *args, **kwargs):
-    ChromeDriverManager(executable_path, target_version, *args, **kwargs).install()

src/undetected_chromedriver/cdp.py

@@ -1,112 +1,112 @@
-#!/usr/bin/env python3
+#!/usr/bin/env python3
-# this module is part of undetected_chromedriver
+# this module is part of undetected_chromedriver
-
+
-import json
+import json
-import logging
+import logging
-from collections.abc import Mapping, Sequence
+
-
+import requests
-import requests
+import websockets
-import websockets
+
-
+
-log = logging.getLogger(__name__)
+log = logging.getLogger(__name__)
-
+
-
+
-class CDPObject(dict):
+class CDPObject(dict):
-    def __init__(self, *a, **k):
+    def __init__(self, *a, **k):
-        super().__init__(*a, **k)
+        super().__init__(*a, **k)
-        self.__dict__ = self
+        self.__dict__ = self
-        for k in self.__dict__:
+        for k in self.__dict__:
-            if isinstance(self.__dict__[k], dict):
+            if isinstance(self.__dict__[k], dict):
-                self.__dict__[k] = CDPObject(self.__dict__[k])
+                self.__dict__[k] = CDPObject(self.__dict__[k])
-            elif isinstance(self.__dict__[k], list):
+            elif isinstance(self.__dict__[k], list):
-                for i in range(len(self.__dict__[k])):
+                for i in range(len(self.__dict__[k])):
-                    if isinstance(self.__dict__[k][i], dict):
+                    if isinstance(self.__dict__[k][i], dict):
-                        self.__dict__[k][i] = CDPObject(self)
+                        self.__dict__[k][i] = CDPObject(self)
-
+
-    def __repr__(self):
+    def __repr__(self):
-        tpl = f"{self.__class__.__name__}(\n\t{{}}\n\t)"
+        tpl = f"{self.__class__.__name__}(\n\t{{}}\n\t)"
-        return tpl.format("\n  ".join(f"{k} = {v}" for k, v in self.items()))
+        return tpl.format("\n  ".join(f"{k} = {v}" for k, v in self.items()))
-
+
-
+
-class PageElement(CDPObject):
+class PageElement(CDPObject):
-    pass
+    pass
-
+
-
+
-class CDP:
+class CDP:
-    log = logging.getLogger("CDP")
+    log = logging.getLogger("CDP")
-
+
-    endpoints = CDPObject(
+    endpoints = CDPObject(
-        {
+        {
-            "json": "/json",
+            "json": "/json",
-            "protocol": "/json/protocol",
+            "protocol": "/json/protocol",
-            "list": "/json/list",
+            "list": "/json/list",
-            "new": "/json/new?{url}",
+            "new": "/json/new?{url}",
-            "activate": "/json/activate/{id}",
+            "activate": "/json/activate/{id}",
-            "close": "/json/close/{id}",
+            "close": "/json/close/{id}",
-        }
+        }
-    )
+    )
-
+
-    def __init__(self, options: "ChromeOptions"):  # noqa
+    def __init__(self, options: "ChromeOptions"):  # noqa
-        self.server_addr = "http://{0}:{1}".format(*options.debugger_address.split(":"))
+        self.server_addr = "http://{0}:{1}".format(*options.debugger_address.split(":"))
-
+
-        self._reqid = 0
+        self._reqid = 0
-        self._session = requests.Session()
+        self._session = requests.Session()
-        self._last_resp = None
+        self._last_resp = None
-        self._last_json = None
+        self._last_json = None
-
+
-        resp = self.get(self.endpoints.json)  # noqa
+        resp = self.get(self.endpoints.json)  # noqa
-        self.sessionId = resp[0]["id"]
+        self.sessionId = resp[0]["id"]
-        self.wsurl = resp[0]["webSocketDebuggerUrl"]
+        self.wsurl = resp[0]["webSocketDebuggerUrl"]
-
+
-    def tab_activate(self, id=None):
+    def tab_activate(self, id=None):
-        if not id:
+        if not id:
-            active_tab = self.tab_list()[0]
+            active_tab = self.tab_list()[0]
-            id = active_tab.id  # noqa
+            id = active_tab.id  # noqa
-            self.wsurl = active_tab.webSocketDebuggerUrl  # noqa
+            self.wsurl = active_tab.webSocketDebuggerUrl  # noqa
-        return self.post(self.endpoints["activate"].format(id=id))
+        return self.post(self.endpoints["activate"].format(id=id))
-
+
-    def tab_list(self):
+    def tab_list(self):
-        retval = self.get(self.endpoints["list"])
+        retval = self.get(self.endpoints["list"])
-        return [PageElement(o) for o in retval]
+        return [PageElement(o) for o in retval]
-
+
-    def tab_new(self, url):
+    def tab_new(self, url):
-        return self.post(self.endpoints["new"].format(url=url))
+        return self.post(self.endpoints["new"].format(url=url))
-
+
-    def tab_close_last_opened(self):
+    def tab_close_last_opened(self):
-        sessions = self.tab_list()
+        sessions = self.tab_list()
-        opentabs = [s for s in sessions if s["type"] == "page"]
+        opentabs = [s for s in sessions if s["type"] == "page"]
-        return self.post(self.endpoints["close"].format(id=opentabs[-1]["id"]))
+        return self.post(self.endpoints["close"].format(id=opentabs[-1]["id"]))
-
+
-    async def send(self, method: str, params: dict):
+    async def send(self, method: str, params: dict):
-        self._reqid += 1
+        self._reqid += 1
-        async with websockets.connect(self.wsurl) as ws:
+        async with websockets.connect(self.wsurl) as ws:
-            await ws.send(
+            await ws.send(
-                json.dumps({"method": method, "params": params, "id": self._reqid})
+                json.dumps({"method": method, "params": params, "id": self._reqid})
-            )
+            )
-            self._last_resp = await ws.recv()
+            self._last_resp = await ws.recv()
-            self._last_json = json.loads(self._last_resp)
+            self._last_json = json.loads(self._last_resp)
-            self.log.info(self._last_json)
+            self.log.info(self._last_json)
-
+
-    def get(self, uri):
+    def get(self, uri):
-        resp = self._session.get(self.server_addr + uri)
+        resp = self._session.get(self.server_addr + uri)
-        try:
+        try:
-            self._last_resp = resp
+            self._last_resp = resp
-            self._last_json = resp.json()
+            self._last_json = resp.json()
-        except Exception:
+        except Exception:
-            return
+            return
-        else:
+        else:
-            return self._last_json
+            return self._last_json
-
+
-    def post(self, uri, data: dict = None):
+    def post(self, uri, data: dict = None):
-        if not data:
+        if not data:
-            data = {}
+            data = {}
-        resp = self._session.post(self.server_addr + uri, json=data)
+        resp = self._session.post(self.server_addr + uri, json=data)
-        try:
+        try:
-            self._last_resp = resp
+            self._last_resp = resp
-            self._last_json = resp.json()
+            self._last_json = resp.json()
-        except Exception:
+        except Exception:
-            return self._last_resp
+            return self._last_resp
-
+
-    @property
+    @property
-    def last_json(self):
+    def last_json(self):
-        return self._last_json
+        return self._last_json

src/undetected_chromedriver/devtool.py

@@ -1,191 +1,190 @@
-import asyncio
+import asyncio
-import logging
+from collections.abc import Mapping
-import time
+from collections.abc import Sequence
-import traceback
+from functools import wraps
-from collections.abc import Mapping
+import logging
-from collections.abc import Sequence
+import threading
-from typing import Any
+import time
-from typing import Awaitable
+import traceback
-from typing import Callable
+from typing import Any
-from typing import List
+from typing import Awaitable
-from typing import Optional
+from typing import Callable
-from contextlib import ExitStack
+from typing import List
-import threading
+from typing import Optional
-from functools import wraps, partial
+
-
+
-
+class Structure(dict):
-class Structure(dict):
+    """
-    """
+    This is a dict-like object structure, which you should subclass
-    This is a dict-like object structure, which you should subclass
+    Only properties defined in the class context are used on initialization.
-    Only properties defined in the class context are used on initialization.
+
-
+    See example
-    See example
+    """
-    """
+
-
+    _store = {}
-    _store = {}
+
-
+    def __init__(self, *a, **kw):
-    def __init__(self, *a, **kw):
+        """
-        """
+        Instantiate a new instance.
-        Instantiate a new instance.
+
-
+        :param a:
-        :param a:
+        :param kw:
-        :param kw:
+        """
-        """
+
-
+        super().__init__()
-        super().__init__()
+
-
+        # auxiliar dict
-        # auxiliar dict
+        d = dict(*a, **kw)
-        d = dict(*a, **kw)
+        for k, v in d.items():
-        for k, v in d.items():
+            if isinstance(v, Mapping):
-            if isinstance(v, Mapping):
+                self[k] = self.__class__(v)
-                self[k] = self.__class__(v)
+            elif isinstance(v, Sequence) and not isinstance(v, (str, bytes)):
-            elif isinstance(v, Sequence) and not isinstance(v, (str, bytes)):
+                self[k] = [self.__class__(i) for i in v]
-                self[k] = [self.__class__(i) for i in v]
+            else:
-            else:
+                self[k] = v
-                self[k] = v
+        super().__setattr__("__dict__", self)
-        super().__setattr__("__dict__", self)
+
-
+    def __getattr__(self, item):
-    def __getattr__(self, item):
+        return getattr(super(), item)
-        return getattr(super(), item)
+
-
+    def __getitem__(self, item):
-    def __getitem__(self, item):
+        return super().__getitem__(item)
-        return super().__getitem__(item)
+
-
+    def __setattr__(self, key, value):
-    def __setattr__(self, key, value):
+        self.__setitem__(key, value)
-        self.__setitem__(key, value)
+
-
+    def __setitem__(self, key, value):
-    def __setitem__(self, key, value):
+        super().__setitem__(key, value)
-        super().__setitem__(key, value)
+
-
+    def update(self, *a, **kw):
-    def update(self, *a, **kw):
+        super().update(*a, **kw)
-        super().update(*a, **kw)
+
-
+    def __eq__(self, other):
-    def __eq__(self, other):
+        return frozenset(other.items()) == frozenset(self.items())
-        return frozenset(other.items()) == frozenset(self.items())
+
-
+    def __hash__(self):
-    def __hash__(self):
+        return hash(frozenset(self.items()))
-        return hash(frozenset(self.items()))
+
-
+    @classmethod
-    @classmethod
+    def __init_subclass__(cls, **kwargs):
-    def __init_subclass__(cls, **kwargs):
+        cls._store = {}
-        cls._store = {}
+
-
+    def _normalize_strings(self):
-    def _normalize_strings(self):
+        for k, v in self.copy().items():
-        for k, v in self.copy().items():
+            if isinstance(v, (str)):
-            if isinstance(v, (str)):
+                self[k] = v.strip()
-                self[k] = v.strip()
+
-
+
-
+def timeout(seconds=3, on_timeout: Optional[Callable[[callable], Any]] = None):
-def timeout(seconds=3, on_timeout: Optional[Callable[[callable], Any]] = None):
+    def wrapper(func):
-    def wrapper(func):
+        @wraps(func)
-        @wraps(func)
+        def wrapped(*args, **kwargs):
-        def wrapped(*args, **kwargs):
+            def function_reached_timeout():
-            def function_reached_timeout():
+                if on_timeout:
-                if on_timeout:
+                    on_timeout(func)
-                    on_timeout(func)
+                else:
-                else:
+                    raise TimeoutError("function call timed out")
-                    raise TimeoutError("function call timed out")
+
-
+            t = threading.Timer(interval=seconds, function=function_reached_timeout)
-            t = threading.Timer(interval=seconds, function=function_reached_timeout)
+            t.start()
-            t.start()
+            try:
-            try:
+                return func(*args, **kwargs)
-                return func(*args, **kwargs)
+            except:
-            except:
+                t.cancel()
-                t.cancel()
+                raise
-                raise
+            finally:
-            finally:
+                t.cancel()
-                t.cancel()
+
-
+        return wrapped
-        return wrapped
+
-
+    return wrapper
-    return wrapper
+
-
+
-
+def test():
-def test():
+    import sys, os
-    import sys, os
+
-
+    sys.path.insert(0, os.path.abspath(os.path.dirname(__file__)))
-    sys.path.insert(0, os.path.abspath(os.path.dirname(__file__)))
+    import undetected_chromedriver as uc
-    import undetected_chromedriver as uc
+    import threading
-    import threading
+
-
+    def collector(
-    def collector(
+        driver: uc.Chrome,
-        driver: uc.Chrome,
+        stop_event: threading.Event,
-        stop_event: threading.Event,
+        on_event_coro: Optional[Callable[[List[str]], Awaitable[Any]]] = None,
-        on_event_coro: Optional[Callable[[List[str]], Awaitable[Any]]] = None,
+        listen_events: Sequence = ("browser", "network", "performance"),
-        listen_events: Sequence = ("browser", "network", "performance"),
+    ):
-    ):
+        def threaded(driver, stop_event, on_event_coro):
-        def threaded(driver, stop_event, on_event_coro):
+            async def _ensure_service_started():
-            async def _ensure_service_started():
+                while (
-                while (
+                    getattr(driver, "service", False)
-                    getattr(driver, "service", False)
+                    and getattr(driver.service, "process", False)
-                    and getattr(driver.service, "process", False)
+                    and driver.service.process.poll()
-                    and driver.service.process.poll()
+                ):
-                ):
+                    print("waiting for driver service to come back on")
-                    print("waiting for driver service to come back on")
+                    await asyncio.sleep(0.05)
-                    await asyncio.sleep(0.05)
+                    # await asyncio.sleep(driver._delay or .25)
-                    # await asyncio.sleep(driver._delay or .25)
+
-
+            async def get_log_lines(typ):
-            async def get_log_lines(typ):
+                await _ensure_service_started()
-                await _ensure_service_started()
+                return driver.get_log(typ)
-                return driver.get_log(typ)
+
-
+            async def looper():
-            async def looper():
+                while not stop_event.is_set():
-                while not stop_event.is_set():
+                    log_lines = []
-                    log_lines = []
+                    try:
-                    try:
+                        for _ in listen_events:
-                        for _ in listen_events:
+                            try:
-                            try:
+                                log_lines += await get_log_lines(_)
-                                log_lines += await get_log_lines(_)
+                            except:
-                            except:
+                                if logging.getLogger().getEffectiveLevel() <= 10:
-                                if logging.getLogger().getEffectiveLevel() <= 10:
+                                    traceback.print_exc()
-                                    traceback.print_exc()
+                                continue
-                                continue
+                        if log_lines and on_event_coro:
-                        if log_lines and on_event_coro:
+                            await on_event_coro(log_lines)
-                            await on_event_coro(log_lines)
+                    except Exception as e:
-                    except Exception as e:
+                        if logging.getLogger().getEffectiveLevel() <= 10:
-                        if logging.getLogger().getEffectiveLevel() <= 10:
+                            traceback.print_exc()
-                            traceback.print_exc()
+
-
+            loop = asyncio.new_event_loop()
-            loop = asyncio.new_event_loop()
+            asyncio.set_event_loop(loop)
-            asyncio.set_event_loop(loop)
+            loop.run_until_complete(looper())
-            loop.run_until_complete(looper())
+
-
+        t = threading.Thread(target=threaded, args=(driver, stop_event, on_event_coro))
-        t = threading.Thread(target=threaded, args=(driver, stop_event, on_event_coro))
+        t.start()
-        t.start()
+
-
+    async def on_event(data):
-    async def on_event(data):
+        print("on_event")
-        print("on_event")
+        print("data:", data)
-        print("data:", data)
+
-
+    def func_called(fn):
-    def func_called(fn):
+        def wrapped(*args, **kwargs):
-        def wrapped(*args, **kwargs):
+            print(
-            print(
+                "func called! %s  (args: %s, kwargs: %s)" % (fn.__name__, args, kwargs)
-                "func called! %s  (args: %s, kwargs: %s)" % (fn.__name__, args, kwargs)
+            )
-            )
+            while driver.service.process and driver.service.process.poll() is not None:
-            while driver.service.process and driver.service.process.poll() is not None:
+                time.sleep(0.1)
-                time.sleep(0.1)
+            res = fn(*args, **kwargs)
-            res = fn(*args, **kwargs)
+            print("func completed! (result: %s)" % res)
-            print("func completed! (result: %s)" % res)
+            return res
-            return res
+
-
+        return wrapped
-        return wrapped
+
-
+    logging.basicConfig(level=10)
-    logging.basicConfig(level=10)
+
-
+    options = uc.ChromeOptions()
-    options = uc.ChromeOptions()
+    options.set_capability(
-    options.set_capability(
+        "goog:loggingPrefs", {"performance": "ALL", "browser": "ALL", "network": "ALL"}
-        "goog:loggingPrefs", {"performance": "ALL", "browser": "ALL", "network": "ALL"}
+    )
-    )
+
-
+    driver = uc.Chrome(version_main=96, options=options)
-    driver = uc.Chrome(version_main=96, options=options)
+
-
+    # driver.command_executor._request = timeout(seconds=1)(driver.command_executor._request)
-    # driver.command_executor._request = timeout(seconds=1)(driver.command_executor._request)
+    driver.command_executor._request = func_called(driver.command_executor._request)
-    driver.command_executor._request = func_called(driver.command_executor._request)
+    collector_stop = threading.Event()
-    collector_stop = threading.Event()
+    collector(driver, collector_stop, on_event)
-    collector(driver, collector_stop, on_event)
+
-
+    driver.get("https://nowsecure.nl")
-    driver.get("https://nowsecure.nl")
+
-
+    time.sleep(10)
-    time.sleep(10)
+
-
+    driver.quit()
-    driver.quit()

src/undetected_chromedriver/dprocess.py

@@ -1,75 +1,76 @@
-import multiprocessing
+import atexit
-import os
+import logging
-import platform
+import multiprocessing
-import sys
+import os
-from subprocess import PIPE
+import platform
-from subprocess import Popen
+import signal
-import atexit
+from subprocess import PIPE
-import traceback
+from subprocess import Popen
-import logging
+import sys
-import signal
+
-
+
-CREATE_NEW_PROCESS_GROUP = 0x00000200
+CREATE_NEW_PROCESS_GROUP = 0x00000200
-DETACHED_PROCESS = 0x00000008
+DETACHED_PROCESS = 0x00000008
-
+
-REGISTERED = []
+REGISTERED = []
-
+
-
+
-def start_detached(executable, *args):
+def start_detached(executable, *args):
-    """
+    """
-    Starts a fully independent subprocess (with no parent)
+    Starts a fully independent subprocess (with no parent)
-    :param executable: executable
+    :param executable: executable
-    :param args: arguments to the executable, eg: ['--param1_key=param1_val', '-vvv' ...]
+    :param args: arguments to the executable, eg: ['--param1_key=param1_val', '-vvv' ...]
-    :return: pid of the grandchild process
+    :return: pid of the grandchild process
-    """
+    """
-
+
-    # create pipe
+    # create pipe
-    reader, writer = multiprocessing.Pipe(False)
+    reader, writer = multiprocessing.Pipe(False)
-
+
-    # do not keep reference
+    # do not keep reference
-    multiprocessing.Process(
+    process = multiprocessing.Process(
-        target=_start_detached,
+        target=_start_detached,
-        args=(executable, *args),
+        args=(executable, *args),
-        kwargs={"writer": writer},
+        kwargs={"writer": writer},
-        daemon=True,
+        daemon=True,
-    ).start()
+    )
-    # receive pid from pipe
+    process.start()
-    pid = reader.recv()
+    process.join()
-    REGISTERED.append(pid)
+    # receive pid from pipe
-    # close pipes
+    pid = reader.recv()
-    writer.close()
+    REGISTERED.append(pid)
-    reader.close()
+    # close pipes
-
+    writer.close()
-    return pid
+    reader.close()
-
+
-
+    return pid
-def _start_detached(executable, *args, writer: multiprocessing.Pipe = None):
+
-
+
-    # configure launch
+def _start_detached(executable, *args, writer: multiprocessing.Pipe = None):
-    kwargs = {}
+    # configure launch
-    if platform.system() == "Windows":
+    kwargs = {}
-        kwargs.update(creationflags=DETACHED_PROCESS | CREATE_NEW_PROCESS_GROUP)
+    if platform.system() == "Windows":
-    elif sys.version_info < (3, 2):
+        kwargs.update(creationflags=DETACHED_PROCESS | CREATE_NEW_PROCESS_GROUP)
-        # assume posix
+    elif sys.version_info < (3, 2):
-        kwargs.update(preexec_fn=os.setsid)
+        # assume posix
-    else:  # Python 3.2+ and Unix
+        kwargs.update(preexec_fn=os.setsid)
-        kwargs.update(start_new_session=True)
+    else:  # Python 3.2+ and Unix
-
+        kwargs.update(start_new_session=True)
-    # run
+
-    p = Popen([executable, *args], stdin=PIPE, stdout=PIPE, stderr=PIPE, **kwargs)
+    # run
-
+    p = Popen([executable, *args], stdin=PIPE, stdout=PIPE, stderr=PIPE, **kwargs)
-    # send pid to pipe
+
-    writer.send(p.pid)
+    # send pid to pipe
-    sys.exit()
+    writer.send(p.pid)
-
+    sys.exit()
-
+
-def _cleanup():
+
-    for pid in REGISTERED:
+def _cleanup():
-        try:
+    for pid in REGISTERED:
-            logging.getLogger(__name__).debug("cleaning up pid %d " % pid)
+        try:
-            os.kill(pid, signal.SIGTERM)
+            logging.getLogger(__name__).debug("cleaning up pid %d " % pid)
-        except:  # noqa
+            os.kill(pid, signal.SIGTERM)
-            pass
+        except:  # noqa
-
+            pass
-
+
-atexit.register(_cleanup)
+
+atexit.register(_cleanup)

src/undetected_chromedriver/options.py

@@ -1,70 +1,85 @@
-#!/usr/bin/env python3
+#!/usr/bin/env python3
-# this module is part of undetected_chromedriver
+# this module is part of undetected_chromedriver
-
+
-
+
-import json
+import json
-import os
+import os
-
+
-from selenium.webdriver.chromium.options import ChromiumOptions as _ChromiumOptions
+from selenium.webdriver.chromium.options import ChromiumOptions as _ChromiumOptions
-
+
-
+
-class ChromeOptions(_ChromiumOptions):
+class ChromeOptions(_ChromiumOptions):
-    _session = None
+    _session = None
-    _user_data_dir = None
+    _user_data_dir = None
-
+
-    @property
+    @property
-    def user_data_dir(self):
+    def user_data_dir(self):
-        return self._user_data_dir
+        return self._user_data_dir
-
+
-    @user_data_dir.setter
+    @user_data_dir.setter
-    def user_data_dir(self, path: str):
+    def user_data_dir(self, path: str):
-        """
+        """
-        Sets the browser profile folder to use, or creates a new profile
+        Sets the browser profile folder to use, or creates a new profile
-        at given <path>.
+        at given <path>.
-
+
-        Parameters
+        Parameters
-        ----------
+        ----------
-        path: str
+        path: str
-            the path to a chrome profile folder
+            the path to a chrome profile folder
-            if it does not exist, a new profile will be created at given location
+            if it does not exist, a new profile will be created at given location
-        """
+        """
-        apath = os.path.abspath(path)
+        apath = os.path.abspath(path)
-        self._user_data_dir = os.path.normpath(apath)
+        self._user_data_dir = os.path.normpath(apath)
-
+
-    @staticmethod
+    @staticmethod
-    def _undot_key(key, value):
+    def _undot_key(key, value):
-        """turn a (dotted key, value) into a proper nested dict"""
+        """turn a (dotted key, value) into a proper nested dict"""
-        if "." in key:
+        if "." in key:
-            key, rest = key.split(".", 1)
+            key, rest = key.split(".", 1)
-            value = ChromeOptions._undot_key(rest, value)
+            value = ChromeOptions._undot_key(rest, value)
-        return {key: value}
+        return {key: value}
-
+
-    def handle_prefs(self, user_data_dir):
+    @staticmethod
-        prefs = self.experimental_options.get("prefs")
+    def _merge_nested(a, b):
-        if prefs:
+        """
-
+        merges b into a
-            user_data_dir = user_data_dir or self._user_data_dir
+        leaf values in a are overwritten with values from b
-            default_path = os.path.join(user_data_dir, "Default")
+        """
-            os.makedirs(default_path, exist_ok=True)
+        for key in b:
-
+            if key in a:
-            # undot prefs dict keys
+                if isinstance(a[key], dict) and isinstance(b[key], dict):
-            undot_prefs = {}
+                    ChromeOptions._merge_nested(a[key], b[key])
-            for key, value in prefs.items():
+                    continue
-                undot_prefs.update(self._undot_key(key, value))
+            a[key] = b[key]
-
+        return a
-            prefs_file = os.path.join(default_path, "Preferences")
+
-            if os.path.exists(prefs_file):
+    def handle_prefs(self, user_data_dir):
-                with open(prefs_file, encoding="latin1", mode="r") as f:
+        prefs = self.experimental_options.get("prefs")
-                    undot_prefs.update(json.load(f))
+        if prefs:
-
+            user_data_dir = user_data_dir or self._user_data_dir
-            with open(prefs_file, encoding="latin1", mode="w") as f:
+            default_path = os.path.join(user_data_dir, "Default")
-                json.dump(undot_prefs, f)
+            os.makedirs(default_path, exist_ok=True)
-
+
-            # remove the experimental_options to avoid an error
+            # undot prefs dict keys
-            del self._experimental_options["prefs"]
+            undot_prefs = {}
-
+            for key, value in prefs.items():
-    @classmethod
+                undot_prefs = self._merge_nested(
-    def from_options(cls, options):
+                    undot_prefs, self._undot_key(key, value)
-        o = cls()
+                )
-        o.__dict__.update(options.__dict__)
+
-        return o
+            prefs_file = os.path.join(default_path, "Preferences")
+            if os.path.exists(prefs_file):
+                with open(prefs_file, encoding="latin1", mode="r") as f:
+                    undot_prefs = self._merge_nested(json.load(f), undot_prefs)
+
+            with open(prefs_file, encoding="latin1", mode="w") as f:
+                json.dump(undot_prefs, f)
+
+            # remove the experimental_options to avoid an error
+            del self._experimental_options["prefs"]
+
+    @classmethod
+    def from_options(cls, options):
+        o = cls()
+        o.__dict__.update(options.__dict__)
+        return o

src/undetected_chromedriver/patcher.py

@@ -1,276 +1,275 @@
-#!/usr/bin/env python3
+#!/usr/bin/env python3
-# this module is part of undetected_chromedriver
+# this module is part of undetected_chromedriver
-
+
-import io
+from distutils.version import LooseVersion
-import logging
+import io
-import os
+import logging
-import random
+import os
-import re
+import random
-import string
+import re
-import sys
+import string
-import time
+import sys
-import zipfile
+import time
-from distutils.version import LooseVersion
+from urllib.request import urlopen
-from urllib.request import urlopen, urlretrieve
+from urllib.request import urlretrieve
-import secrets
+import zipfile
-
+
-
+
-logger = logging.getLogger(__name__)
+logger = logging.getLogger(__name__)
-
+
-IS_POSIX = sys.platform.startswith(("darwin", "cygwin", "linux"))
+IS_POSIX = sys.platform.startswith(("darwin", "cygwin", "linux", "linux2"))
-
+
-
+
-class Patcher(object):
+class Patcher(object):
-    url_repo = "https://chromedriver.storage.googleapis.com"
+    url_repo = "https://chromedriver.storage.googleapis.com"
-    zip_name = "chromedriver_%s.zip"
+    zip_name = "chromedriver_%s.zip"
-    exe_name = "chromedriver%s"
+    exe_name = "chromedriver%s"
-
+
-    platform = sys.platform
+    platform = sys.platform
-    if platform.endswith("win32"):
+    if platform.endswith("win32"):
-        zip_name %= "win32"
+        zip_name %= "win32"
-        exe_name %= ".exe"
+        exe_name %= ".exe"
-    if platform.endswith("linux"):
+    if platform.endswith(("linux", "linux2")):
-        zip_name %= "linux64"
+        zip_name %= "linux64"
-        exe_name %= ""
+        exe_name %= ""
-    if platform.endswith("darwin"):
+    if platform.endswith("darwin"):
-        zip_name %= "mac64"
+        zip_name %= "mac64"
-        exe_name %= ""
+        exe_name %= ""
-
+
-    if platform.endswith("win32"):
+    if platform.endswith("win32"):
-        d = "~/appdata/roaming/undetected_chromedriver"
+        d = "~/appdata/roaming/undetected_chromedriver"
-    elif platform.startswith("linux"):
+    elif "LAMBDA_TASK_ROOT" in os.environ:
-        d = "~/.local/share/undetected_chromedriver"
+        d = "/tmp/undetected_chromedriver"
-    elif platform.endswith("darwin"):
+    elif platform.startswith(("linux", "linux2")):
-        d = "~/Library/Application Support/undetected_chromedriver"
+        d = "~/.local/share/undetected_chromedriver"
-    else:
+    elif platform.endswith("darwin"):
-        d = "~/.undetected_chromedriver"
+        d = "~/Library/Application Support/undetected_chromedriver"
-    data_path = os.path.abspath(os.path.expanduser(d))
+    else:
-
+        d = "~/.undetected_chromedriver"
-    def __init__(self, executable_path=None, force=False, version_main: int = 0):
+    data_path = os.path.abspath(os.path.expanduser(d))
-        """
+
-
+    def __init__(self, executable_path=None, force=False, version_main: int = 0):
-        Args:
+        """
-            executable_path: None = automatic
+        Args:
-                             a full file path to the chromedriver executable
+            executable_path: None = automatic
-            force: False
+                             a full file path to the chromedriver executable
-                    terminate processes which are holding lock
+            force: False
-            version_main: 0 = auto
+                    terminate processes which are holding lock
-                specify main chrome version (rounded, ex: 82)
+            version_main: 0 = auto
-        """
+                specify main chrome version (rounded, ex: 82)
-
+        """
-        self.force = force
+        self.force = force
-        self.executable_path = None
+        self._custom_exe_path = False
-        prefix = secrets.token_hex(8)
+        prefix = "undetected"
-
+
-        if not os.path.exists(self.data_path):
+        if not os.path.exists(self.data_path):
-            os.makedirs(self.data_path, exist_ok=True)
+            os.makedirs(self.data_path, exist_ok=True)
-
+
-        if not executable_path:
+        if not executable_path:
-            self.executable_path = os.path.join(
+            self.executable_path = os.path.join(
-                self.data_path, "_".join([prefix, self.exe_name])
+                self.data_path, "_".join([prefix, self.exe_name])
-            )
+            )
-
+
-        if not IS_POSIX:
+        if not IS_POSIX:
-            if executable_path:
+            if executable_path:
-                if not executable_path[-4:] == ".exe":
+                if not executable_path[-4:] == ".exe":
-                    executable_path += ".exe"
+                    executable_path += ".exe"
-
+
-        self.zip_path = os.path.join(self.data_path, prefix)
+        self.zip_path = os.path.join(self.data_path, prefix)
-
+
-        if not executable_path:
+        if not executable_path:
-            self.executable_path = os.path.abspath(
+            self.executable_path = os.path.abspath(
-                os.path.join(".", self.executable_path)
+                os.path.join(".", self.executable_path)
-            )
+            )
-
+
-        self._custom_exe_path = False
+        if executable_path:
-
+            self._custom_exe_path = True
-        if executable_path:
+            self.executable_path = executable_path
-            self._custom_exe_path = True
+        self.version_main = version_main
-            self.executable_path = executable_path
+        self.version_full = None
-        self.version_main = version_main
+
-        self.version_full = None
+    def auto(self, executable_path=None, force=False, version_main=None):
-
+        if executable_path:
-    def auto(self, executable_path=None, force=False, version_main=None):
+            self.executable_path = executable_path
-        """"""
+            self._custom_exe_path = True
-        if executable_path:
+
-            self.executable_path = executable_path
+        if self._custom_exe_path:
-            self._custom_exe_path = True
+            ispatched = self.is_binary_patched(self.executable_path)
-
+            if not ispatched:
-        if self._custom_exe_path:
+                return self.patch_exe()
-            ispatched = self.is_binary_patched(self.executable_path)
+            else:
-            if not ispatched:
+                return
-                return self.patch_exe()
+
-            else:
+        if version_main:
-                return
+            self.version_main = version_main
-
+        if force is True:
-        if version_main:
+            self.force = force
-            self.version_main = version_main
+
-        if force is True:
+        try:
-            self.force = force
+            os.unlink(self.executable_path)
-
+        except PermissionError:
-        try:
+            if self.force:
-            os.unlink(self.executable_path)
+                self.force_kill_instances(self.executable_path)
-        except PermissionError:
+                return self.auto(force=not self.force)
-            if self.force:
+            try:
-                self.force_kill_instances(self.executable_path)
+                if self.is_binary_patched():
-                return self.auto(force=not self.force)
+                    # assumes already running AND patched
-            try:
+                    return True
-                if self.is_binary_patched():
+            except PermissionError:
-                    # assumes already running AND patched
+                pass
-                    return True
+            # return False
-            except PermissionError:
+        except FileNotFoundError:
-                pass
+            pass
-            # return False
+
-        except FileNotFoundError:
+        release = self.fetch_release_number()
-            pass
+        self.version_main = release.version[0]
-
+        self.version_full = release
-        release = self.fetch_release_number()
+        self.unzip_package(self.fetch_package())
-        self.version_main = release.version[0]
+        return self.patch()
-        self.version_full = release
+
-        self.unzip_package(self.fetch_package())
+    def patch(self):
-        return self.patch()
+        self.patch_exe()
-
+        return self.is_binary_patched()
-    def patch(self):
+
-        self.patch_exe()
+    def fetch_release_number(self):
-        return self.is_binary_patched()
+        """
-
+        Gets the latest major version available, or the latest major version of self.target_version if set explicitly.
-    def fetch_release_number(self):
+        :return: version string
-        """
+        :rtype: LooseVersion
-        Gets the latest major version available, or the latest major version of self.target_version if set explicitly.
+        """
-        :return: version string
+        path = "/latest_release"
-        :rtype: LooseVersion
+        if self.version_main:
-        """
+            path += f"_{self.version_main}"
-        path = "/latest_release"
+        path = path.upper()
-        if self.version_main:
+        logger.debug("getting release number from %s" % path)
-            path += f"_{self.version_main}"
+        return LooseVersion(urlopen(self.url_repo + path).read().decode())
-        path = path.upper()
+
-        logger.debug("getting release number from %s" % path)
+    def parse_exe_version(self):
-        return LooseVersion(urlopen(self.url_repo + path).read().decode())
+        with io.open(self.executable_path, "rb") as f:
-
+            for line in iter(lambda: f.readline(), b""):
-    def parse_exe_version(self):
+                match = re.search(rb"platform_handle\x00content\x00([0-9.]*)", line)
-        with io.open(self.executable_path, "rb") as f:
+                if match:
-            for line in iter(lambda: f.readline(), b""):
+                    return LooseVersion(match[1].decode())
-                match = re.search(rb"platform_handle\x00content\x00([0-9.]*)", line)
+
-                if match:
+    def fetch_package(self):
-                    return LooseVersion(match[1].decode())
+        """
-
+        Downloads ChromeDriver from source
-    def fetch_package(self):
+
-        """
+        :return: path to downloaded file
-        Downloads ChromeDriver from source
+        """
-
+        u = "%s/%s/%s" % (self.url_repo, self.version_full.vstring, self.zip_name)
-        :return: path to downloaded file
+        logger.debug("downloading from %s" % u)
-        """
+        # return urlretrieve(u, filename=self.data_path)[0]
-        u = "%s/%s/%s" % (self.url_repo, self.version_full.vstring, self.zip_name)
+        return urlretrieve(u)[0]
-        logger.debug("downloading from %s" % u)
+
-        # return urlretrieve(u, filename=self.data_path)[0]
+    def unzip_package(self, fp):
-        return urlretrieve(u)[0]
+        """
-
+        Does what it says
-    def unzip_package(self, fp):
+
-        """
+        :return: path to unpacked executable
-        Does what it says
+        """
-
+        logger.debug("unzipping %s" % fp)
-        :return: path to unpacked executable
+        try:
-        """
+            os.unlink(self.zip_path)
-        logger.debug("unzipping %s" % fp)
+        except (FileNotFoundError, OSError):
-        try:
+            pass
-            os.unlink(self.zip_path)
+
-        except (FileNotFoundError, OSError):
+        os.makedirs(self.zip_path, mode=0o755, exist_ok=True)
-            pass
+        with zipfile.ZipFile(fp, mode="r") as zf:
-
+            zf.extract(self.exe_name, self.zip_path)
-        os.makedirs(self.zip_path, mode=0o755, exist_ok=True)
+        os.rename(os.path.join(self.zip_path, self.exe_name), self.executable_path)
-        with zipfile.ZipFile(fp, mode="r") as zf:
+        os.remove(fp)
-            zf.extract(self.exe_name, self.zip_path)
+        os.rmdir(self.zip_path)
-        os.rename(os.path.join(self.zip_path, self.exe_name), self.executable_path)
+        os.chmod(self.executable_path, 0o755)
-        os.remove(fp)
+        return self.executable_path
-        os.rmdir(self.zip_path)
+
-        os.chmod(self.executable_path, 0o755)
+    @staticmethod
-        return self.executable_path
+    def force_kill_instances(exe_name):
-
+        """
-    @staticmethod
+        kills running instances.
-    def force_kill_instances(exe_name):
+        :param: executable name to kill, may be a path as well
-        """
+
-        kills running instances.
+        :return: True on success else False
-        :param: executable name to kill, may be a path as well
+        """
-
+        exe_name = os.path.basename(exe_name)
-        :return: True on success else False
+        if IS_POSIX:
-        """
+            r = os.system("kill -f -9 $(pidof %s)" % exe_name)
-        exe_name = os.path.basename(exe_name)
+        else:
-        if IS_POSIX:
+            r = os.system("taskkill /f /im %s" % exe_name)
-            r = os.system("kill -f -9 $(pidof %s)" % exe_name)
+        return not r
-        else:
+
-            r = os.system("taskkill /f /im %s" % exe_name)
+    @staticmethod
-        return not r
+    def gen_random_cdc():
-
+        cdc = random.choices(string.ascii_letters, k=27)
-    @staticmethod
+        return "".join(cdc).encode()
-    def gen_random_cdc():
+
-        cdc = random.choices(string.ascii_lowercase, k=26)
+    def is_binary_patched(self, executable_path=None):
-        cdc[-6:-4] = map(str.upper, cdc[-6:-4])
+        executable_path = executable_path or self.executable_path
-        cdc[2] = cdc[0]
+        try:
-        cdc[3] = "_"
+            with io.open(executable_path, "rb") as fh:
-        return "".join(cdc).encode()
+                return fh.read().find(b"undetected chromedriver") != -1
-
+        except FileNotFoundError:
-    def is_binary_patched(self, executable_path=None):
+            return False
-        """simple check if executable is patched.
+
-
+    def patch_exe(self):
-        :return: False if not patched, else True
+        start = time.perf_counter()
-        """
+        logger.info("patching driver executable %s" % self.executable_path)
-        executable_path = executable_path or self.executable_path
+        with io.open(self.executable_path, "r+b") as fh:
-        with io.open(executable_path, "rb") as fh:
+            content = fh.read()
-            for line in iter(lambda: fh.readline(), b""):
+            # match_injected_codeblock = re.search(rb"{window.*;}", content)
-                if b"cdc_" in line:
+            match_injected_codeblock = re.search(rb"\{window\.cdc.*?;\}", content)
-                    return False
+            if match_injected_codeblock:
-            else:
+                target_bytes = match_injected_codeblock[0]
-                return True
+                new_target_bytes = (
-
+                    b'{console.log("undetected chromedriver 1337!")}'.ljust(
-    def patch_exe(self):
+                        len(target_bytes), b" "
-        """
+                    )
-        Patches the ChromeDriver binary
+                )
-
+                new_content = content.replace(target_bytes, new_target_bytes)
-        :return: False on failure, binary name on success
+                if new_content == content:
-        """
+                    logger.warning(
-        logger.info("patching driver executable %s" % self.executable_path)
+                        "something went wrong patching the driver binary. could not find injection code block"
-
+                    )
-        linect = 0
+                else:
-        replacement = self.gen_random_cdc()
+                    logger.debug(
-        with io.open(self.executable_path, "r+b") as fh:
+                        "found block:\n%s\nreplacing with:\n%s"
-            for line in iter(lambda: fh.readline(), b""):
+                        % (target_bytes, new_target_bytes)
-                if b"cdc_" in line:
+                    )
-                    fh.seek(-len(line), 1)
+                fh.seek(0)
-                    newline = re.sub(b"cdc_.{22}", replacement, line)
+                fh.write(new_content)
-                    fh.write(newline)
+        logger.debug(
-                    linect += 1
+            "patching took us {:.2f} seconds".format(time.perf_counter() - start)
-            return linect
+        )
-
+
-    def __repr__(self):
+    def __repr__(self):
-        return "{0:s}({1:s})".format(
+        return "{0:s}({1:s})".format(
-            self.__class__.__name__,
+            self.__class__.__name__,
-            self.executable_path,
+            self.executable_path,
-        )
+        )
-
+
-    def __del__(self):
+    def __del__(self):
-
+        if self._custom_exe_path:
-        if self._custom_exe_path:
+            # if the driver binary is specified by user
-            # if the driver binary is specified by user
+            # we assume it is important enough to not delete it
-            # we assume it is important enough to not delete it
+            return
-            return
+        else:
-        else:
+            timeout = 3  # stop trying after this many seconds
-            timeout = 3  # stop trying after this many seconds
+            t = time.monotonic()
-            t = time.monotonic()
+            while True:
-            while True:
+                now = time.monotonic()
-                now = time.monotonic()
+                if now - t > timeout:
-                if now - t > timeout:
+                    # we don't want to wait until the end of time
-                    # we don't want to wait until the end of time
+                    logger.debug(
-                    logger.debug(
+                        "could not unlink %s in time (%d seconds)"
-                        "could not unlink %s in time (%d seconds)"
+                        % (self.executable_path, timeout)
-                        % (self.executable_path, timeout)
+                    )
-                    )
+                    break
-                    break
+                try:
-                try:
+                    os.unlink(self.executable_path)
-                    os.unlink(self.executable_path)
+                    logger.debug("successfully unlinked %s" % self.executable_path)
-                    logger.debug("successfully unlinked %s" % self.executable_path)
+                    break
-                    break
+                except (OSError, RuntimeError, PermissionError):
-                except (OSError, RuntimeError, PermissionError):
+                    time.sleep(0.1)
-                    time.sleep(0.1)
+                    continue
-                    continue
+                except FileNotFoundError:
-                except FileNotFoundError:
+                    break
-                    break

src/undetected_chromedriver/reactor.py

@@ -1,102 +1,99 @@
-#!/usr/bin/env python3
+#!/usr/bin/env python3
-# this module is part of undetected_chromedriver
+# this module is part of undetected_chromedriver
-
+
-import asyncio
+import asyncio
-import json
+import json
-import logging
+import logging
-import threading
+import threading
-
+
-logger = logging.getLogger(__name__)
+
-
+logger = logging.getLogger(__name__)
-
+
-class Reactor(threading.Thread):
+
-    def __init__(self, driver: "Chrome"):
+class Reactor(threading.Thread):
-        super().__init__()
+    def __init__(self, driver: "Chrome"):
-
+        super().__init__()
-        self.driver = driver
+
-        self.loop = asyncio.new_event_loop()
+        self.driver = driver
-
+        self.loop = asyncio.new_event_loop()
-        self.lock = threading.Lock()
+
-        self.event = threading.Event()
+        self.lock = threading.Lock()
-        self.daemon = True
+        self.event = threading.Event()
-        self.handlers = {}
+        self.daemon = True
-
+        self.handlers = {}
-    def add_event_handler(self, method_name, callback: callable):
+
-        """
+    def add_event_handler(self, method_name, callback: callable):
-
+        """
-        Parameters
+
-        ----------
+        Parameters
-        event_name: str
+        ----------
-            example "Network.responseReceived"
+        event_name: str
-
+            example "Network.responseReceived"
-        callback: callable
+
-            callable which accepts 1 parameter: the message object dictionary
+        callback: callable
-
+            callable which accepts 1 parameter: the message object dictionary
-        Returns
+
-        -------
+        Returns
-
+        -------
-        """
+
-        with self.lock:
+        """
-            self.handlers[method_name.lower()] = callback
+        with self.lock:
-
+            self.handlers[method_name.lower()] = callback
-    @property
+
-    def running(self):
+    @property
-        return not self.event.is_set()
+    def running(self):
-
+        return not self.event.is_set()
-    def run(self):
+
-        try:
+    def run(self):
-            asyncio.set_event_loop(self.loop)
+        try:
-            self.loop.run_until_complete(self.listen())
+            asyncio.set_event_loop(self.loop)
-        except Exception as e:
+            self.loop.run_until_complete(self.listen())
-            logger.warning("Reactor.run() => %s", e)
+        except Exception as e:
-
+            logger.warning("Reactor.run() => %s", e)
-    async def _wait_service_started(self):
+
-        while True:
+    async def _wait_service_started(self):
-            with self.lock:
+        while True:
-                if (
+            with self.lock:
-                    getattr(self.driver, "service", None)
+                if (
-                    and getattr(self.driver.service, "process", None)
+                    getattr(self.driver, "service", None)
-                    and self.driver.service.process.poll()
+                    and getattr(self.driver.service, "process", None)
-                ):
+                    and self.driver.service.process.poll()
-                    await asyncio.sleep(self.driver._delay or 0.25)
+                ):
-                else:
+                    await asyncio.sleep(self.driver._delay or 0.25)
-                    break
+                else:
-
+                    break
-    async def listen(self):
+
-
+    async def listen(self):
-        while self.running:
+        while self.running:
-
+            await self._wait_service_started()
-            await self._wait_service_started()
+            await asyncio.sleep(1)
-            await asyncio.sleep(1)
+
-
+            try:
-            try:
+                with self.lock:
-                with self.lock:
+                    log_entries = self.driver.get_log("performance")
-                    log_entries = self.driver.get_log("performance")
+
-
+                for entry in log_entries:
-                for entry in log_entries:
+                    try:
-
+                        obj_serialized: str = entry.get("message")
-                    try:
+                        obj = json.loads(obj_serialized)
-
+                        message = obj.get("message")
-                        obj_serialized: str = entry.get("message")
+                        method = message.get("method")
-                        obj = json.loads(obj_serialized)
+
-                        message = obj.get("message")
+                        if "*" in self.handlers:
-                        method = message.get("method")
+                            await self.loop.run_in_executor(
-
+                                None, self.handlers["*"], message
-                        if "*" in self.handlers:
+                            )
-                            await self.loop.run_in_executor(
+                        elif method.lower() in self.handlers:
-                                None, self.handlers["*"], message
+                            await self.loop.run_in_executor(
-                            )
+                                None, self.handlers[method.lower()], message
-                        elif method.lower() in self.handlers:
+                            )
-                            await self.loop.run_in_executor(
+
-                                None, self.handlers[method.lower()], message
+                        # print(type(message), message)
-                            )
+                    except Exception as e:
-
+                        raise e from None
-                        # print(type(message), message)
+
-                    except Exception as e:
+            except Exception as e:
-                        raise e from None
+                if "invalid session id" in str(e):
-
+                    pass
-            except Exception as e:
+                else:
-                if "invalid session id" in str(e):
+                    logging.debug("exception ignored :", e)
-                    pass
-                else:
-                    logging.debug("exception ignored :", e)

src/undetected_chromedriver/v2.py

@@ -1,4 +0,0 @@
-# for backward compatibility
-import sys
-
-sys.modules[__name__] = sys.modules[__package__]

src/undetected_chromedriver/webelement.py

@@ -1,37 +1,86 @@
-import selenium.webdriver.remote.webelement
+from typing import List
-
+
-
+from selenium.webdriver.common.by import By
-class WebElement(selenium.webdriver.remote.webelement.WebElement):
+import selenium.webdriver.remote.webelement
-    """
+
-    Custom WebElement class which makes it easier to view elements when
+
-    working in an interactive environment.
+class WebElement(selenium.webdriver.remote.webelement.WebElement):
-
+    def click_safe(self):
-    standard webelement repr:
+        super().click()
-    <selenium.webdriver.remote.webelement.WebElement (session="85ff0f671512fa535630e71ee951b1f2", element="6357cb55-92c3-4c0f-9416-b174f9c1b8c4")>
+        self._parent.reconnect(0.1)
-
+
-    using this WebElement class:
+    def children(
-    <WebElement(<a class="mobile-show-inline-block mc-update-infos init-ok" href="#" id="main-cat-switcher-mobile">)>
+        self, tag=None, recursive=False
-
+    ) -> List[selenium.webdriver.remote.webelement.WebElement]:
-    """
+        """
-
+        returns direct child elements of current element
-    @property
+        :param tag: str,  if supplied, returns <tag> nodes only
-    def attrs(self):
+        """
-        if not hasattr(self, "_attrs"):
+        script = "return [... arguments[0].children]"
-            self._attrs = self._parent.execute_script(
+        if tag:
-                """
+            script += ".filter( node => node.tagName === '%s')" % tag.upper()
-                var items = {}; 
+        if recursive:
-                for (index = 0; index < arguments[0].attributes.length; ++index) 
+            return list(_recursive_children(self, tag))
-                {
+        return list(self._parent.execute_script(script, self))
-                 items[arguments[0].attributes[index].name] = arguments[0].attributes[index].value 
+
-                }; 
+
-                return items;
+class UCWebElement(WebElement):
-                """,
+    """
-                self,
+    Custom WebElement class which makes it easier to view elements when
-            )
+    working in an interactive environment.
-        return self._attrs
+
-
+    standard webelement repr:
-    def __repr__(self):
+    <selenium.webdriver.remote.webelement.WebElement (session="85ff0f671512fa535630e71ee951b1f2", element="6357cb55-92c3-4c0f-9416-b174f9c1b8c4")>
-        strattrs = " ".join([f'{k}="{v}"' for k, v in self.attrs.items()])
+
-        if strattrs:
+    using this WebElement class:
-            strattrs = " " + strattrs
+    <WebElement(<a class="mobile-show-inline-block mc-update-infos init-ok" href="#" id="main-cat-switcher-mobile">)>
-        return f"{self.__class__.__name__} <{self.tag_name}{strattrs}>"
+
+    """
+
+    def __init__(self, parent, id_):
+        super().__init__(parent, id_)
+        self._attrs = None
+
+    @property
+    def attrs(self):
+        if not self._attrs:
+            self._attrs = self._parent.execute_script(
+                """
+                var items = {}; 
+                for (index = 0; index < arguments[0].attributes.length; ++index) 
+                {
+                 items[arguments[0].attributes[index].name] = arguments[0].attributes[index].value 
+                }; 
+                return items;
+                """,
+                self,
+            )
+        return self._attrs
+
+    def __repr__(self):
+        strattrs = " ".join([f'{k}="{v}"' for k, v in self.attrs.items()])
+        if strattrs:
+            strattrs = " " + strattrs
+        return f"{self.__class__.__name__} <{self.tag_name}{strattrs}>"
+
+
+def _recursive_children(element, tag: str = None, _results=None):
+    """
+    returns all children of <element> recursively
+
+    :param element: `WebElement` object.
+            find children below this <element>
+
+    :param tag: str = None.
+            if provided, return only <tag> elements. example: 'a', or 'img'
+    :param _results: do not use!
+    """
+    results = _results or set()
+    for element in element.children():
+        if tag:
+            if element.tag_name == tag:
+                results.add(element)
+        else:
+            results.add(element)
+        results |= _recursive_children(element, tag, results)
+    return results

src/utils.py

@@ -8,6 +8,7 @@ from selenium.webdriver.chrome.webdriver import WebDriver
 import undetected_chromedriver as uc
 
 FLARESOLVERR_VERSION = None
+CHROME_EXE_PATH = None
 CHROME_MAJOR_VERSION = None
 USER_AGENT = None
 XVFB_DISPLAY = None
@@ -28,12 +29,14 @@ def get_flaresolverr_version() -> str:
         return FLARESOLVERR_VERSION
 
     package_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), os.pardir, 'package.json')
+    if not os.path.isfile(package_path):
+        package_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'package.json')
     with open(package_path) as f:
         FLARESOLVERR_VERSION = json.loads(f.read())['version']
         return FLARESOLVERR_VERSION
 
 
-def get_webdriver() -> WebDriver:
+def get_webdriver(proxy: dict = None) -> WebDriver:
     global PATCHED_DRIVER_PATH
     logging.debug('Launching web browser...')
 
@@ -44,6 +47,27 @@ def get_webdriver() -> WebDriver:
     # todo: this param shows a warning in chrome head-full
     options.add_argument('--disable-setuid-sandbox')
     options.add_argument('--disable-dev-shm-usage')
+    # this option removes the zygote sandbox (it seems that the resolution is a bit faster)
+    options.add_argument('--no-zygote')
+    # attempt to fix Docker ARM32 build
+    options.add_argument('--disable-gpu-sandbox')
+    options.add_argument('--disable-software-rasterizer')
+    options.add_argument('--ignore-certificate-errors')
+    options.add_argument('--ignore-ssl-errors')
+    # fix GL errors in ASUSTOR NAS
+    # https://github.com/FlareSolverr/FlareSolverr/issues/782
+    # https://github.com/microsoft/vscode/issues/127800#issuecomment-873342069
+    # https://peter.sh/experiments/chromium-command-line-switches/#use-gl
+    options.add_argument('--use-gl=swiftshader')
+    # workaround for updated 'verify your are human' check
+    # https://github.com/FlareSolverr/FlareSolverr/issues/811
+    options.add_argument('--auto-open-devtools-for-tabs')
+    options.add_argument('--headless=true')
+
+    if proxy and 'url' in proxy:
+        proxy_url = proxy['url']
+        logging.debug("Using webdriver proxy: %s", proxy_url)
+        options.add_argument('--proxy-server=%s' % proxy_url)
 
     # note: headless mode is detected (options.headless = True)
     # we launch the browser in head-full mode with the window hidden
@@ -65,15 +89,26 @@ def get_webdriver() -> WebDriver:
         if PATCHED_DRIVER_PATH is not None:
             driver_exe_path = PATCHED_DRIVER_PATH
 
+    # detect chrome path
+    browser_executable_path = get_chrome_exe_path()
+
     # downloads and patches the chromedriver
     # if we don't set driver_executable_path it downloads, patches, and deletes the driver each time
-    driver = uc.Chrome(options=options, driver_executable_path=driver_exe_path, version_main=version_main,
+    driver = uc.Chrome(options=options, browser_executable_path=browser_executable_path,
+                       driver_executable_path=driver_exe_path, version_main=version_main,
                        windows_headless=windows_headless)
 
     # save the patched driver to avoid re-downloads
     if driver_exe_path is None:
         PATCHED_DRIVER_PATH = os.path.join(driver.patcher.data_path, driver.patcher.exe_name)
         shutil.copy(driver.patcher.executable_path, PATCHED_DRIVER_PATH)
+        
+    # workaround for updated 'verify your are human' check
+    # https://github.com/FlareSolverr/FlareSolverr/issues/811
+    driver.execute_script('''window.open("","_blank");''')
+    driver.switch_to.window(window_name=driver.window_handles[0])
+    driver.close()
+    driver.switch_to.window(window_name=driver.window_handles[0])
 
     # selenium vanilla
     # options = webdriver.ChromeOptions()
@@ -86,23 +121,45 @@ def get_webdriver() -> WebDriver:
     return driver
 
 
+def get_chrome_exe_path() -> str:
+    global CHROME_EXE_PATH
+    if CHROME_EXE_PATH is not None:
+        return CHROME_EXE_PATH
+    # linux pyinstaller bundle
+    chrome_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'chrome', "chrome")
+    if os.path.exists(chrome_path):
+        if not os.access(chrome_path, os.X_OK):
+            raise Exception(f'Chrome binary "{chrome_path}" is not executable. '
+                            f'Please, extract the archive with "tar xzf <file.tar.gz>".')
+        CHROME_EXE_PATH = chrome_path
+        return CHROME_EXE_PATH
+    # windows pyinstaller bundle
+    chrome_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'chrome', "chrome.exe")
+    if os.path.exists(chrome_path):
+        CHROME_EXE_PATH = chrome_path
+        return CHROME_EXE_PATH
+    # system
+    CHROME_EXE_PATH = uc.find_chrome_executable()
+    return CHROME_EXE_PATH
+
+
 def get_chrome_major_version() -> str:
     global CHROME_MAJOR_VERSION
     if CHROME_MAJOR_VERSION is not None:
         return CHROME_MAJOR_VERSION
 
     if os.name == 'nt':
+        # Example: '104.0.5112.79'
         try:
-            stream = os.popen(
+            complete_version = extract_version_nt_executable(get_chrome_exe_path())
-                'reg query "HKLM\\SOFTWARE\\Wow6432Node\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\Google Chrome"')
-            output = stream.read()
-            # Example: '104.0.5112.79'
-            complete_version = extract_version_registry(output)
         except Exception:
-            # Example: '104.0.5112.79'
+            try:
-            complete_version = extract_version_folder()
+                complete_version = extract_version_nt_registry()
+            except Exception:
+                # Example: '104.0.5112.79'
+                complete_version = extract_version_nt_folder()
     else:
-        chrome_path = uc.find_chrome_executable()
+        chrome_path = get_chrome_exe_path()
         process = os.popen(f'"{chrome_path}" --version')
         # Example 1: 'Chromium 104.0.5112.79 Arch Linux\n'
         # Example 2: 'Google Chrome 104.0.5112.79 Arch Linux\n'
@@ -110,24 +167,32 @@ def get_chrome_major_version() -> str:
         process.close()
 
     CHROME_MAJOR_VERSION = complete_version.split('.')[0].split(' ')[-1]
-    logging.info(f"Chrome major version: {CHROME_MAJOR_VERSION}")
     return CHROME_MAJOR_VERSION
 
 
-def extract_version_registry(output) -> str:
+def extract_version_nt_executable(exe_path: str) -> str:
-    try:
+    import pefile
-        google_version = ''
+    pe = pefile.PE(exe_path, fast_load=True)
-        for letter in output[output.rindex('DisplayVersion    REG_SZ') + 24:]:
+    pe.parse_data_directories(
-            if letter != '\n':
+        directories=[pefile.DIRECTORY_ENTRY["IMAGE_DIRECTORY_ENTRY_RESOURCE"]]
-                google_version += letter
+    )
-            else:
+    return pe.FileInfo[0][0].StringTable[0].entries[b"FileVersion"].decode('utf-8')
-                break
-        return google_version.strip()
-    except TypeError:
-        return ''
 
 
-def extract_version_folder() -> str:
+def extract_version_nt_registry() -> str:
+    stream = os.popen(
+        'reg query "HKLM\\SOFTWARE\\Wow6432Node\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\Google Chrome"')
+    output = stream.read()
+    google_version = ''
+    for letter in output[output.rindex('DisplayVersion    REG_SZ') + 24:]:
+        if letter != '\n':
+            google_version += letter
+        else:
+            break
+    return google_version.strip()
+
+
+def extract_version_nt_folder() -> str:
     # Check if the Chrome folder exists in the x32 or x64 Program Files folders.
     for i in range(2):
         path = 'C:\\Program Files' + (' (x86)' if i else '') + '\\Google\\Chrome\\Application'