Compare commits

..

No commits in common. "master" and "v3.0.0" have entirely different histories.

32 changed files with 2015 additions and 3485 deletions

View File

@ -8,13 +8,6 @@ body:
options: options:
- label: I have checked the README - label: I have checked the README
required: true 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 - type: checkboxes
attributes: attributes:
label: Is there already an issue for your problem? label: Is there already an issue for your problem?
@ -29,13 +22,6 @@ body:
options: options:
- label: I have read the Discussions - label: I have read the Discussions
required: true required: true
- type: input
attributes:
label: Have you ACTUALLY checked all these?
description: Please do not waste our time and yours; these checks are there for a reason, it is not just so you can tick boxes for fun. If you type <b>YES</b> and it is clear you did not or have put in no effort, your issue will be closed and locked without comment. If you type <b>NO</b> but still open this issue, you will be permanently blocked for timewasting.
placeholder: YES or NO
validations:
required: true
- type: textarea - type: textarea
attributes: attributes:
label: Environment label: Environment
@ -46,8 +32,7 @@ body:
- Operating system: - Operating system:
- Are you using Docker: [yes/no] - Are you using Docker: [yes/no]
- FlareSolverr User-Agent (see log traces or / endpoint): - FlareSolverr User-Agent (see log traces or / endpoint):
- Are you using a VPN: [yes/no] - Are you using a proxy or VPN: [yes/no]
- Are you using a Proxy: [yes/no]
- Are you using Captcha Solver: [yes/no] - Are you using Captcha Solver: [yes/no]
- If using captcha solver, which one: - If using captcha solver, which one:
- URL to test this issue: - URL to test this issue:

View File

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

View File

@ -4,64 +4,50 @@ on:
push: push:
tags: tags:
- 'v*.*.*' - 'v*.*.*'
pull_request:
branches:
- master
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: true
jobs: jobs:
build-docker-images: build:
if: ${{ !github.event.pull_request.head.repo.fork }} runs-on: ubuntu-latest
runs-on: ubuntu-22.04
steps: steps:
- name: Checkout -
uses: actions/checkout@v4 name: Checkout
uses: actions/checkout@v2
- name: Downcase repo -
name: Downcase repo
run: echo REPOSITORY=$(echo ${{ github.repository }} | tr '[:upper:]' '[:lower:]') >> $GITHUB_ENV run: echo REPOSITORY=$(echo ${{ github.repository }} | tr '[:upper:]' '[:lower:]') >> $GITHUB_ENV
-
- name: Docker meta name: Docker meta
id: docker_meta id: docker_meta
uses: docker/metadata-action@v5 uses: crazy-max/ghaction-docker-meta@v1
with: with:
images: | images: ${{ env.REPOSITORY }},ghcr.io/${{ env.REPOSITORY }}
${{ env.REPOSITORY }},enable=${{ github.event_name != 'pull_request' }} tag-sha: false
ghcr.io/${{ env.REPOSITORY }} -
tags: | name: Set up QEMU
type=semver,pattern={{version}},prefix=v uses: docker/setup-qemu-action@v1.0.1
type=ref,event=pr -
flavor: | name: Set up Docker Buildx
latest=auto uses: docker/setup-buildx-action@v1
-
- name: Set up QEMU name: Login to DockerHub
uses: docker/setup-qemu-action@v3 uses: docker/login-action@v1
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
- name: Login to DockerHub
uses: docker/login-action@v3
if: github.event_name != 'pull_request'
with: with:
username: ${{ secrets.DOCKERHUB_USERNAME }} username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_TOKEN }} password: ${{ secrets.DOCKERHUB_TOKEN }}
-
- name: Login to GitHub Container Registry name: Login to GitHub Container Registry
uses: docker/login-action@v3 uses: docker/login-action@v1
with: with:
registry: ghcr.io registry: ghcr.io
username: ${{ github.repository_owner }} username: ${{ github.repository_owner }}
password: ${{ secrets.GH_PAT }} password: ${{ secrets.GH_PAT }}
-
- name: Build and push name: Build and push
uses: docker/build-push-action@v6 uses: docker/build-push-action@v2
with: with:
context: . context: .
file: ./Dockerfile file: ./Dockerfile
platforms: linux/386,linux/amd64,linux/arm/v7,linux/arm64/v8 platforms: linux/386,linux/amd64,linux/arm/v7,linux/arm64/v8
push: true push: ${{ github.event_name != 'pull_request' }}
tags: ${{ steps.docker_meta.outputs.tags }} tags: ${{ steps.docker_meta.outputs.tags }}
labels: ${{ steps.docker_meta.outputs.labels }} labels: ${{ steps.docker_meta.outputs.labels }}

View File

@ -6,15 +6,26 @@ on:
- 'v*.*.*' - 'v*.*.*'
jobs: jobs:
create-release: build:
name: Create release name: Create release
runs-on: ubuntu-22.04 runs-on: ubuntu-latest
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v3 uses: actions/checkout@v2
with: with:
fetch-depth: 0 # get all commits, branches and tags (required for the changelog) 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 - name: Build changelog
id: github_changelog id: github_changelog
run: | run: |
@ -36,60 +47,9 @@ jobs:
draft: false draft: false
prerelease: 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==5.13.0
cd src
python build_package.py
- name: Upload release artifacts - name: Upload release artifacts
uses: alexellis/upload-assets@0.4.0 uses: alexellis/upload-assets@0.2.2
env: env:
GITHUB_TOKEN: ${{ secrets.GH_PAT }} GITHUB_TOKEN: ${{ secrets.GH_PAT }}
with: with:
asset_paths: '["./dist/flaresolverr_*"]' asset_paths: '["./bin/*.zip"]'
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==5.13.0
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_*"]'

4
.gitignore vendored
View File

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

View File

@ -1,196 +1,5 @@
# Changelog # Changelog
## v3.3.24 (2025/06/04)
* Remove hidden character
## v3.3.23 (2025/06/04)
* Update base image to bookworm. Thanks @rwjack
## v3.3.22 (2025/06/03)
* Disable search engine choice screen
* Fix headless=false stalling. Thanks @MAKMED1337
* Change from click to keys. Thanks @sh4dowb
* Don't open devtools
* Bump Chromium to v137 for build
* Bump requirements
## v3.3.21 (2024/06/26)
* Add challenge selector to catch reloading page on non-English systems
* Escape values for generated form used in request.post. Thanks @mynameisbogdan
## v3.3.20 (2024/06/21)
* maxTimeout should always be int
* Check not running in Docker before logging version_main error
* Update Cloudflare challenge and checkbox selectors. Thanks @tenettow & @21hsmw
## v3.3.19 (2024/05/23)
* Fix occasional headless issue on Linux when set to "false". Thanks @21hsmw
## v3.3.18 (2024/05/20)
* Fix LANG ENV for Linux
* Fix Chrome v124+ not closing on Windows. Thanks @RileyXX
## v3.3.17 (2024/04/09)
* Fix file descriptor leak in service on quit(). Thanks @zkulis
## v3.3.16 (2024/02/28)
* Fix of the subprocess.STARTUPINFO() call. Thanks @ceconelo
* Add FreeBSD support. Thanks @Asthowen
* Use headless configuration properly. Thanks @hashworks
## v3.3.15 (2024/02/20)
* Fix looping challenges
## v3.3.14-hotfix2 (2024/02/17)
* Hotfix 2 - bad Chromium build, instances failed to terminate
## v3.3.14-hotfix (2024/02/17)
* Hotfix for Linux build - some Chrome files no longer exist
## v3.3.14 (2024/02/17)
* Update Chrome downloads. Thanks @opemvbs
## v3.3.13 (2024/01/07)
* Fix too many open files error
## v3.3.12 (2023/12/15)
* Fix looping challenges and invalid cookies
## v3.3.11 (2023/12/11)
* Update UC 3.5.4 & Selenium 4.15.2. Thanks @txtsd
## v3.3.10 (2023/11/14)
* Add LANG ENV - resolves issues with YGGtorrent
## v3.3.9 (2023/11/13)
* Fix for Docker build, capture TypeError
## v3.3.8 (2023/11/13)
* Fix headless=true for Chrome 117+. Thanks @NabiKAZ
* Support running Chrome 119 from source. Thanks @koleg and @Chris7X
* Fix "OSError: [WinError 6] The handle is invalid" on exit. Thanks @enesgorkemgenc
## v3.3.7 (2023/11/05)
* Bump to rebuild. Thanks @JoachimDorchies
## v3.3.6 (2023/09/15)
* Update checkbox selector, again
## v3.3.5 (2023/09/13)
* Change checkbox selector, support languages other than English
## v3.3.4 (2023/09/02)
* Update checkbox selector
## v3.3.3 (2023/08/31)
* Update undetected_chromedriver to v3.5.3
## v3.3.2 (2023/08/03)
* Fix URL domain in Prometheus exporter
## v3.3.1 (2023/08/03)
* Fix for Cloudflare verify checkbox
* Fix HEADLESS=false in Windows binary
* Fix Prometheus exporter for management and health endpoints
* Remove misleading stack trace when the verify checkbox is not found
* Revert "Update base Docker image to Debian Bookworm" #849
* Revert "Install Chromium 115 from Debian testing" #849
## v3.3.0 (2023/08/02)
* Fix for new Cloudflare detection. Thanks @cedric-bour for #845
* Add support for proxy authentication username/password. Thanks @jacobprice808 for #807
* Implement Prometheus metrics
* Fix Chromium Driver for Chrome / Chromium version > 114
* Use Chromium 115 in binary packages (Windows and Linux)
* Install Chromium 115 from Debian testing (Docker)
* Update base Docker image to Debian Bookworm
* Update Selenium 4.11.2
* Update pyinstaller 5.13.0
* Add more traces in build_package.py
## v3.2.2 (2023/07/16)
* Workaround for updated 'verify you 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) ## v3.0.0 (2023/01/04)
* This is the first release of FlareSolverr v3. There are some breaking changes * This is the first release of FlareSolverr v3. There are some breaking changes

View File

@ -1,4 +1,4 @@
FROM python:3.11-slim-bookworm as builder FROM python:3.11-slim-bullseye as builder
# Build dummy packages to skip installing them and their dependencies # Build dummy packages to skip installing them and their dependencies
RUN apt-get update \ RUN apt-get update \
@ -12,7 +12,7 @@ RUN apt-get update \
&& equivs-build adwaita-icon-theme \ && equivs-build adwaita-icon-theme \
&& mv adwaita-icon-theme_*.deb /adwaita-icon-theme.deb && mv adwaita-icon-theme_*.deb /adwaita-icon-theme.deb
FROM python:3.11-slim-bookworm FROM python:3.11-slim-bullseye
# Copy dummy packages # Copy dummy packages
COPY --from=builder /*.deb / COPY --from=builder /*.deb /
@ -29,8 +29,7 @@ RUN dpkg -i /libgl1-mesa-dri.deb \
&& dpkg -i /adwaita-icon-theme.deb \ && dpkg -i /adwaita-icon-theme.deb \
# Install dependencies # Install dependencies
&& apt-get update \ && apt-get update \
&& apt-get install -y --no-install-recommends chromium chromium-common chromium-driver xvfb dumb-init \ && apt-get install -y --no-install-recommends chromium chromium-common chromium-driver xvfb \
procps curl vim xauth \
# Remove temporary files and hardware decoding libraries # Remove temporary files and hardware decoding libraries
&& rm -rf /var/lib/apt/lists/* \ && rm -rf /var/lib/apt/lists/* \
&& rm -f /usr/lib/x86_64-linux-gnu/libmfxhw* \ && rm -f /usr/lib/x86_64-linux-gnu/libmfxhw* \
@ -48,31 +47,19 @@ RUN pip install -r requirements.txt \
USER flaresolverr USER flaresolverr
RUN mkdir -p "/app/.config/chromium/Crash Reports/pending"
COPY src . COPY src .
COPY package.json ../ COPY package.json ../
EXPOSE 8191 EXPOSE 8191
EXPOSE 8192
# dumb-init avoids zombie chromium processes
ENTRYPOINT ["/usr/bin/dumb-init", "--"]
CMD ["/usr/local/bin/python", "-u", "/app/flaresolverr.py"] CMD ["/usr/local/bin/python", "-u", "/app/flaresolverr.py"]
# Local build # Local build
# docker build -t ngosang/flaresolverr:3.3.24 . # docker build -t ngosang/flaresolverr:3.0.0 .
# docker run -p 8191:8191 ngosang/flaresolverr:3.3.24 # docker run -p 8191:8191 ngosang/flaresolverr:3.0.0
# Multi-arch build # Multi-arch build
# docker run --rm --privileged multiarch/qemu-user-static --reset -p yes # docker run --rm --privileged multiarch/qemu-user-static --reset -p yes
# docker buildx create --use # docker buildx create --use
# docker buildx build -t ngosang/flaresolverr:3.3.24 --platform linux/386,linux/amd64,linux/arm/v7,linux/arm64/v8 . # docker buildx build -t ngosang/flaresolverr:3.0.0 --platform linux/386,linux/amd64,linux/arm/v7,linux/arm64/v8 .
# add --push to publish in DockerHub # 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:3.3.24 --platform linux/arm/v7 --load .
# docker run -p 8191:8191 --platform linux/arm/v7 ngosang/flaresolverr:3.3.24

147
README.md
View File

@ -45,8 +45,7 @@ Supported architectures are:
| ARM32 | linux/arm/v7 | | ARM32 | linux/arm/v7 |
| ARM64 | linux/arm64 | | ARM64 | linux/arm64 |
We provide a `docker-compose.yml` configuration file. Clone this repository and execute We provide a `docker-compose.yml` configuration file. Clone this repository and execute `docker-compose up -d` to start
`docker-compose up -d` _(Compose V1)_ or `docker compose up -d` _(Compose V2)_ to start
the container. the container.
If you prefer the `docker cli` execute the following command. If you prefer the `docker cli` execute the following command.
@ -65,76 +64,38 @@ Remember to restart the Docker daemon and the container after the update.
### Precompiled binaries ### 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. This is the recommended way for Windows users.
* Download the [FlareSolverr executable](https://github.com/FlareSolverr/FlareSolverr/releases) from the release's page. It is available for Windows x64 and Linux x64. * Download the [FlareSolverr zip](https://github.com/FlareSolverr/FlareSolverr/releases) from the release's assets. It is available for Windows and Linux.
* 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. * Execute FlareSolverr binary. In the environment variables section you can find how to change the configuration.
### From source code ### From source code
> **Warning** This is the recommended way for macOS users and for developers.
> Installing from source code only works for x64 architecture. For other architectures see Docker images. * Install [Python 3.10](https://www.python.org/downloads/).
* 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/). * (Only in Linux / macOS) Install [Xvfb](https://en.wikipedia.org/wiki/Xvfb) package.
* 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) Install [Xvfb](https://en.wikipedia.org/wiki/Xvfb) package.
* (Only in macOS) Install [XQuartz](https://www.xquartz.org/) package.
* Clone this repository and open a shell in that path. * Clone this repository and open a shell in that path.
* Run `pip install -r requirements.txt` command to install FlareSolverr dependencies. * Run `pip install -r requirements.txt` command to install FlareSolverr dependencies.
* Run `python src/flaresolverr.py` command to start FlareSolverr. * Run `python src/flaresolverr.py` command to start FlareSolverr.
### From source code (FreeBSD/TrueNAS CORE)
* Run `pkg install chromium python311 py311-pip xorg-vfbserver` command to install the required dependencies.
* Clone this repository and open a shell in that path.
* Run `python3.11 -m pip install -r requirements.txt` command to install FlareSolverr dependencies.
* Run `python3.11 src/flaresolverr.py` command to start FlareSolverr.
### Systemd service ### Systemd service
We provide an example Systemd unit file `flaresolverr.service` as reference. You have to modify the file to suit your needs: paths, user and environment variables. We provide an example Systemd unit file `flaresolverr.service` as reference. You have to modify the file to suit your needs: paths, user and environment variables.
## Usage ## Usage
Example Bash request: Example request:
```bash ```bash
curl -L -X POST 'http://localhost:8191/v1' \ curl -L -X POST 'http://localhost:8191/v1' \
-H 'Content-Type: application/json' \ -H 'Content-Type: application/json' \
--data-raw '{ --data-raw '{
"cmd": "request.get", "cmd": "request.get",
"url": "http://www.google.com/", "url":"http://www.google.com/",
"maxTimeout": 60000 "maxTimeout": 60000
}' }'
``` ```
Example Python request:
```py
import requests
url = "http://localhost:8191/v1"
headers = {"Content-Type": "application/json"}
data = {
"cmd": "request.get",
"url": "http://www.google.com/",
"maxTimeout": 60000
}
response = requests.post(url, headers=headers, json=data)
print(response.text)
```
Example PowerShell request:
```ps1
$body = @{
cmd = "request.get"
url = "http://www.google.com/"
maxTimeout = 60000
} | ConvertTo-Json
irm -UseBasicParsing 'http://localhost:8191/v1' -Headers @{"Content-Type"="application/json"} -Method Post -Body $body
```
### Commands ### Commands
#### + `sessions.create` #### + `sessions.create`
@ -145,10 +106,10 @@ cookies for the browser to use.
This also speeds up the requests since it won't have to launch a new browser instance for every request. This also speeds up the requests since it won't have to launch a new browser instance for every request.
| Parameter | Notes | | Parameter | Notes |
|-----------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| |-----------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| session | Optional. The session ID that you want to be assigned to the instance. If isn't set a random UUID will be assigned. | | session | Optional. The session ID that you want to be assigned to the instance. If isn't set a random UUID will be assigned. |
| 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 supported. Eg: `"proxy": {"url": "http://127.0.0.1:8888", "username": "testuser", "password": "testpass"}` | | 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. |
#### + `sessions.list` #### + `sessions.list`
@ -179,18 +140,16 @@ session. When you no longer need to use a session you should make sure to close
#### + `request.get` #### + `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. |
| session_ttl_minutes | Optional. FlareSolverr will automatically rotate expired sessions based on the TTL provided in minutes. | | maxTimeout | Optional, default value 60000. Max timeout to solve the challenge in milliseconds. |
| maxTimeout | Optional, default value 60000. Max timeout to solve the challenge in milliseconds. | | 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. |
| cookies | Optional. Will be used by the headless browser. Eg: `"cookies": [{"name": "cookie1", "value": "value1"}, {"name": "cookie2", "value": "value2"}]`. | | returnOnlyCookies | Optional, default false. Only returns the cookies. Response data, headers and other parts of the response are removed. |
| 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`.) |
| 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** :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.
> 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: Example response from running the `curl` above:
@ -261,63 +220,33 @@ This is the same as `request.get` but it takes one more param:
## Environment variables ## Environment variables
| Name | Default | Notes | | Name | Default | Notes |
|--------------------|------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------| |-----------------|------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------|
| LOG_LEVEL | info | Verbosity of the logging. Use `LOG_LEVEL=debug` for more information. | | LOG_LEVEL | info | Verbosity of the logging. Use `LOG_LEVEL=debug` for more information. |
| LOG_HTML | false | Only for debugging. If `true` all HTML that passes through the proxy will be logged to the console in `debug` level. | | LOG_HTML | false | Only for debugging. If `true` all HTML that passes through the proxy will be logged to the console in `debug` level. |
| CAPTCHA_SOLVER | none | Captcha solving method. It is used when a captcha is encountered. See the Captcha Solvers section. | | CAPTCHA_SOLVER | none | Captcha solving method. It is used when a captcha is encountered. See the Captcha Solvers section. |
| TZ | UTC | Timezone used in the logs and the web browser. Example: `TZ=Europe/London`. | | TZ | UTC | Timezone used in the logs and the web browser. Example: `TZ=Europe/London`. |
| LANG | none | Language used in the web browser. Example: `LANG=en_GB`. | | HEADLESS | true | Only for debugging. To run the web browser in headless mode or visible. |
| HEADLESS | true | Only for debugging. To run the web browser in headless mode or visible. | | BROWSER_TIMEOUT | 40000 | If you are experiencing errors/timeouts because your system is slow, you can try to increase this value. Remember to increase the `maxTimeout` parameter too. |
| BROWSER_TIMEOUT | 40000 | If you are experiencing errors/timeouts because your system is slow, you can try to increase this value. Remember to increase the `maxTimeout` parameter too. | | TEST_URL | https://www.google.com | FlareSolverr makes a request on start to make sure the web browser is working. You can change that URL if it is blocked in your country. |
| TEST_URL | https://www.google.com | FlareSolverr makes a request on start to make sure the web browser is working. You can change that URL if it is blocked in your country. | | PORT | 8191 | Listening port. You don't need to change this if you are running on Docker. |
| PORT | 8191 | Listening port. You don't need to change this if you are running on Docker. | | HOST | 0.0.0.0 | Listening interface. You don't need to change this if you are running on Docker. |
| HOST | 0.0.0.0 | Listening interface. You don't need to change this if you are running on Docker. |
| PROMETHEUS_ENABLED | false | Enable Prometheus exporter. See the Prometheus section below. |
| PROMETHEUS_PORT | 8192 | Listening port for Prometheus exporter. See the Prometheus section below. |
Environment variables are set differently depending on the operating system. Some examples: Environment variables are set differently depending on the operating system. Some examples:
* Docker: Take a look at the Docker section in this document. Environment variables can be set in the `docker-compose.yml` file or in the Docker CLI command. * Docker: Take a look at the Docker section in this document. Environment variables can be set in the `docker-compose.yml` file or in the Docker CLI command.
* Linux: Run `export LOG_LEVEL=debug` and then run `flaresolverr` in the same shell. * Linux: Run `export LOG_LEVEL=debug` and then start FlareSolverr in the same shell.
* Windows: Open `cmd.exe`, run `set LOG_LEVEL=debug` and then run `flaresolverr.exe` in the same shell. * Windows: Open `cmd.exe`, run `set LOG_LEVEL=debug` and then start FlareSolverr in the same shell.
## Prometheus exporter
The Prometheus exporter for FlareSolverr is disabled by default. It can be enabled with the environment variable `PROMETHEUS_ENABLED`. If you are using Docker make sure you expose the `PROMETHEUS_PORT`.
Example metrics:
```shell
# HELP flaresolverr_request_total Total requests with result
# TYPE flaresolverr_request_total counter
flaresolverr_request_total{domain="nowsecure.nl",result="solved"} 1.0
# HELP flaresolverr_request_created Total requests with result
# TYPE flaresolverr_request_created gauge
flaresolverr_request_created{domain="nowsecure.nl",result="solved"} 1.690141657157109e+09
# HELP flaresolverr_request_duration Request duration in seconds
# TYPE flaresolverr_request_duration histogram
flaresolverr_request_duration_bucket{domain="nowsecure.nl",le="0.0"} 0.0
flaresolverr_request_duration_bucket{domain="nowsecure.nl",le="10.0"} 1.0
flaresolverr_request_duration_bucket{domain="nowsecure.nl",le="25.0"} 1.0
flaresolverr_request_duration_bucket{domain="nowsecure.nl",le="50.0"} 1.0
flaresolverr_request_duration_bucket{domain="nowsecure.nl",le="+Inf"} 1.0
flaresolverr_request_duration_count{domain="nowsecure.nl"} 1.0
flaresolverr_request_duration_sum{domain="nowsecure.nl"} 5.858
# HELP flaresolverr_request_duration_created Request duration in seconds
# TYPE flaresolverr_request_duration_created gauge
flaresolverr_request_duration_created{domain="nowsecure.nl"} 1.6901416571570296e+09
```
## Captcha Solvers ## Captcha Solvers
> **Warning** :warning: At this time none of the captcha solvers work. You can check the status in the open issues. Any help is welcome.
> 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 Sometimes CloudFlare not only gives mathematical computations and browser tests, sometimes they also require the user to
solve a captcha. solve a captcha.
If this is the case, FlareSolverr will return the error `Captcha detected but no automatic solver is configured.` 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 CAPTCHA automatically by setting the environment variable `CAPTCHA_SOLVER` FlareSolverr can be customized to solve the captchas automatically by setting the environment variable `CAPTCHA_SOLVER`
to the file name of one of the adapters inside the `/captcha` directory. to the file name of one of the adapters inside the [/captcha](src/captcha) directory.
## Related projects ## Related projects

View File

@ -1,19 +0,0 @@
[Unit]
Description=FlareSolverr
After=network.target
[Service]
SyslogIdentifier=flaresolverr
Restart=always
RestartSec=5
Type=simple
User=flaresolverr
Group=flaresolverr
Environment="LOG_LEVEL=info"
Environment="CAPTCHA_SOLVER=none"
WorkingDirectory=/opt/flaresolverr
ExecStart=/opt/flaresolverr/flaresolverr
TimeoutStopSec=30
[Install]
WantedBy=multi-user.target

View File

@ -1,7 +1,7 @@
{ {
"name": "flaresolverr", "name": "flaresolverr",
"version": "3.3.24", "version": "3.0.0",
"description": "Proxy server to bypass Cloudflare protection", "description": "Proxy server to bypass Cloudflare protection",
"author": "Diego Heras (ngosang / ngosang@hotmail.es)", "author": "Diego Heras (ngosang / ngosang@hotmail.es)",
"license": "MIT" "license": "MIT"
} }

View File

@ -1,13 +1,9 @@
bottle==0.12.25 bottle==0.12.23
waitress==3.0.1 waitress==2.1.2
selenium==4.15.2 selenium==4.4.3
func-timeout==4.3.5 func-timeout==4.3.5
prometheus-client==0.17.1
# required by undetected_chromedriver # required by undetected_chromedriver
requests==2.32.3 requests==2.28.1
certifi==2024.07.04 websockets==10.3
websockets==11.0.3 # only required for linux
# only required for linux and macos xvfbwrapper==0.2.9
xvfbwrapper==0.2.9; platform_system != "Windows"
# only required for windows
pefile==2023.2.7; platform_system == "Windows"

Binary file not shown.

Before

Width:  |  Height:  |  Size: 8.8 KiB

View File

@ -1,66 +0,0 @@
import logging
import os
import urllib.parse
from bottle import request
from dtos import V1RequestBase, V1ResponseBase
from metrics import start_metrics_http_server, REQUEST_COUNTER, REQUEST_DURATION
PROMETHEUS_ENABLED = os.environ.get('PROMETHEUS_ENABLED', 'false').lower() == 'true'
PROMETHEUS_PORT = int(os.environ.get('PROMETHEUS_PORT', 8192))
def setup():
if PROMETHEUS_ENABLED:
start_metrics_http_server(PROMETHEUS_PORT)
def prometheus_plugin(callback):
"""
Bottle plugin to expose Prometheus metrics
http://bottlepy.org/docs/dev/plugindev.html
"""
def wrapper(*args, **kwargs):
actual_response = callback(*args, **kwargs)
if PROMETHEUS_ENABLED:
try:
export_metrics(actual_response)
except Exception as e:
logging.warning("Error exporting metrics: " + str(e))
return actual_response
def export_metrics(actual_response):
res = V1ResponseBase(actual_response)
if res.startTimestamp is None or res.endTimestamp is None:
# skip management and healthcheck endpoints
return
domain = "unknown"
if res.solution and res.solution.url:
domain = parse_domain_url(res.solution.url)
else:
# timeout error
req = V1RequestBase(request.json)
if req.url:
domain = parse_domain_url(req.url)
run_time = (res.endTimestamp - res.startTimestamp) / 1000
REQUEST_DURATION.labels(domain=domain).observe(run_time)
result = "unknown"
if res.message == "Challenge solved!":
result = "solved"
elif res.message == "Challenge not detected!":
result = "not_detected"
elif res.message.startswith("Error"):
result = "error"
REQUEST_COUNTER.labels(domain=domain, result=result).inc()
def parse_domain_url(url):
parsed_url = urllib.parse.urlparse(url)
return parsed_url.hostname
return wrapper

View File

@ -1,110 +0,0 @@
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 = "1453032" if os.name == 'nt' else '1453031'
arch = 'Win_x64' 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', '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 ':'
result = subprocess.run([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, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
if result.returncode != 0:
print(result.stderr.decode('utf-8'))
raise Exception("Error running pyInstaller")
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

View File

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

View File

@ -3,13 +3,11 @@ import logging
import os import os
import sys import sys
import certifi from bottle import run, response, Bottle, request
from bottle import run, response, Bottle, request, ServerAdapter
from bottle_plugins.error_plugin import error_plugin from bottle_plugins.error_plugin import error_plugin
from bottle_plugins.logger_plugin import logger_plugin from bottle_plugins.logger_plugin import logger_plugin
from bottle_plugins import prometheus_plugin from dtos import IndexResponse, V1RequestBase
from dtos import V1RequestBase
import flaresolverr_service import flaresolverr_service
import utils import utils
@ -25,6 +23,10 @@ class JSONErrorBottle(Bottle):
app = JSONErrorBottle() app = JSONErrorBottle()
# plugin order is important
app.install(logger_plugin)
app.install(error_plugin)
@app.route('/') @app.route('/')
def index(): def index():
@ -58,22 +60,6 @@ def controller_v1():
if __name__ == "__main__": 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 for HEADLESS=false in Windows binary
# https://stackoverflow.com/a/27694505
if os.name == 'nt':
import multiprocessing
multiprocessing.freeze_support()
# 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 # validate configuration
log_level = os.environ.get('LOG_LEVEL', 'info').upper() log_level = os.environ.get('LOG_LEVEL', 'info').upper()
log_html = utils.get_config_log_html() log_html = utils.get_config_log_html()
@ -101,25 +87,9 @@ if __name__ == "__main__":
logging.info(f'FlareSolverr {utils.get_flaresolverr_version()}') logging.info(f'FlareSolverr {utils.get_flaresolverr_version()}')
logging.debug('Debug log enabled') logging.debug('Debug log enabled')
# Get current OS for global variable
utils.get_current_platform()
# test browser installation # test browser installation
flaresolverr_service.test_browser_installation() flaresolverr_service.test_browser_installation()
# start bootle plugins
# plugin order is important
app.install(logger_plugin)
app.install(error_plugin)
prometheus_plugin.setup()
app.install(prometheus_plugin.prometheus_plugin)
# start webserver # start webserver
# default server 'wsgiref' does not support concurrent requests # default server 'wsgiref' does not support concurrent requests
# https://github.com/FlareSolverr/FlareSolverr/issues/680 run(app, host=server_host, port=server_port, quiet=True, server='waitress')
# 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)

View File

@ -1,79 +1,42 @@
import logging import logging
import platform
import sys
import time import time
from datetime import timedelta from urllib.parse import unquote
from html import escape
from urllib.parse import unquote, quote
from func_timeout import FunctionTimedOut, func_timeout from func_timeout import func_timeout, FunctionTimedOut
from selenium.common import TimeoutException from selenium.common import TimeoutException
from selenium.webdriver.chrome.webdriver import WebDriver from selenium.webdriver.chrome.webdriver import WebDriver
from selenium.webdriver.common.by import By from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
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.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 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 = [ ACCESS_DENIED_SELECTORS = [
# Cloudflare # 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_TITLES = [ CHALLENGE_TITLE = [
# Cloudflare # Cloudflare
'Just a moment...', 'Just a moment...',
# DDoS-GUARD # DDoS-GUARD
'DDoS-Guard' 'DDOS-GUARD',
] ]
CHALLENGE_SELECTORS = [ CHALLENGE_SELECTORS = [
# Cloudflare # Cloudflare
'#cf-challenge-running', '.ray_id', '.attack-box', '#cf-please-wait', '#challenge-spinner', '#trk_jschal_js', '#turnstile-wrapper', '.lds-ring', '#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 # 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 = 1 SHORT_TIMEOUT = 10
SESSIONS_STORAGE = SessionsStorage()
def test_browser_installation(): def test_browser_installation():
logging.info("Testing web 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() user_agent = utils.get_user_agent()
logging.info("FlareSolverr User-Agent: " + user_agent) logging.info("FlareSolverr User-Agent: " + user_agent)
logging.info("Test successful!") logging.info("Test successful")
def index_endpoint() -> IndexResponse: def index_endpoint() -> IndexResponse:
@ -121,17 +84,17 @@ def _controller_v1_handler(req: V1RequestBase) -> V1ResponseBase:
logging.warning("Request parameter 'userAgent' was removed in FlareSolverr v2.") logging.warning("Request parameter 'userAgent' was removed in FlareSolverr v2.")
# set default values # set default values
if req.maxTimeout is None or int(req.maxTimeout) < 1: if req.maxTimeout is None or req.maxTimeout < 1:
req.maxTimeout = 60000 req.maxTimeout = 60000
# execute the command # execute the command
res: V1ResponseBase res: V1ResponseBase
if req.cmd == 'sessions.create': if req.cmd == 'sessions.create':
res = _cmd_sessions_create(req) raise Exception("Not implemented yet.")
elif req.cmd == 'sessions.list': elif req.cmd == 'sessions.list':
res = _cmd_sessions_list(req) raise Exception("Not implemented yet.")
elif req.cmd == 'sessions.destroy': elif req.cmd == 'sessions.destroy':
res = _cmd_sessions_destroy(req) raise Exception("Not implemented yet.")
elif req.cmd == 'request.get': elif req.cmd == 'request.get':
res = _cmd_request_get(req) res = _cmd_request_get(req)
elif req.cmd == 'request.post': elif req.cmd == 'request.post':
@ -178,108 +141,19 @@ def _cmd_request_post(req: V1RequestBase) -> V1ResponseBase:
return res 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: def _resolve_challenge(req: V1RequestBase, method: str) -> ChallengeResolutionT:
timeout = int(req.maxTimeout) / 1000 timeout = req.maxTimeout / 1000
driver = None driver = None
try: try:
if req.session: driver = utils.get_webdriver()
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)) return func_timeout(timeout, _evil_logic, (req, driver, method))
except FunctionTimedOut: except FunctionTimedOut:
raise Exception(f'Error solving the challenge. Timeout after {timeout} seconds.') raise Exception(f'Error solving the challenge. Timeout after {timeout} seconds.')
except Exception as e: except Exception as e:
raise Exception('Error solving the challenge. ' + str(e).replace('\n', '\\n')) raise Exception('Error solving the challenge. ' + str(e))
finally: finally:
if not req.session and driver is not None: if driver is not None:
if utils.PLATFORM_VERSION == "nt":
driver.close()
driver.quit() 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...")
actions = ActionChains(driver)
actions.pause(5).send_keys(Keys.TAB).pause(1).send_keys(Keys.SPACE).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:
logging.debug("The Cloudflare 'Verify you are human' button not found on the page.")
time.sleep(2)
def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> ChallengeResolutionT: def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> ChallengeResolutionT:
@ -287,37 +161,18 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
res.status = STATUS_OK res.status = STATUS_OK
res.message = "" res.message = ""
# navigate to the page # navigate to the page
logging.debug(f'Navigating to... {req.url}') logging.debug(f'Navigating to... {req.url}')
if method == 'POST': if method == 'POST':
_post_request(req, driver) _post_request(req, driver)
else: else:
driver.get(req.url) driver.get(req.url)
# 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(): if utils.get_config_log_html():
logging.debug(f"Response HTML:\n{driver.page_source}") 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 # wait for the page
for title in ACCESS_DENIED_TITLES: html_element = driver.find_element(By.TAG_NAME, "html")
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 # find access denied selectors
for selector in ACCESS_DENIED_SELECTORS: for selector in ACCESS_DENIED_SELECTORS:
found_elements = driver.find_elements(By.CSS_SELECTOR, selector) found_elements = driver.find_elements(By.CSS_SELECTOR, selector)
@ -327,10 +182,11 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
# find challenge by title # find challenge by title
challenge_found = False challenge_found = False
for title in CHALLENGE_TITLES: page_title = driver.title
if title.lower() == page_title.lower(): for title in CHALLENGE_TITLE:
if title == page_title:
challenge_found = True challenge_found = True
logging.info("Challenge detected. Title found: " + page_title) logging.info("Challenge detected. Title found: " + title)
break break
if not challenge_found: if not challenge_found:
# find challenge by selectors # find challenge by selectors
@ -341,19 +197,17 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
logging.info("Challenge detected. Selector found: " + selector) logging.info("Challenge detected. Selector found: " + selector)
break break
attempt = 0
if challenge_found: if challenge_found:
while True: while True:
try: try:
attempt = attempt + 1 # wait until the title change
# wait until the title changes for title in CHALLENGE_TITLE:
for title in CHALLENGE_TITLES: logging.debug("Waiting for title: " + title)
logging.debug("Waiting for title (attempt " + str(attempt) + "): " + title)
WebDriverWait(driver, SHORT_TIMEOUT).until_not(title_is(title)) WebDriverWait(driver, SHORT_TIMEOUT).until_not(title_is(title))
# then wait until all the selectors disappear # then wait until all the selectors disappear
for selector in CHALLENGE_SELECTORS: for selector in CHALLENGE_SELECTORS:
logging.debug("Waiting for selector (attempt " + str(attempt) + "): " + selector) logging.debug("Waiting for selector: " + selector)
WebDriverWait(driver, SHORT_TIMEOUT).until_not( WebDriverWait(driver, SHORT_TIMEOUT).until_not(
presence_of_element_located((By.CSS_SELECTOR, selector))) presence_of_element_located((By.CSS_SELECTOR, selector)))
@ -362,9 +216,6 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
except TimeoutException: except TimeoutException:
logging.debug("Timeout waiting for selector") logging.debug("Timeout waiting for selector")
click_verify(driver)
# update the html (cloudflare reloads the page every 5 s) # update the html (cloudflare reloads the page every 5 s)
html_element = driver.find_element(By.TAG_NAME, "html") html_element = driver.find_element(By.TAG_NAME, "html")
@ -414,7 +265,7 @@ def _post_request(req: V1RequestBase, driver: WebDriver):
value = unquote(parts[1]) value = unquote(parts[1])
except Exception: except Exception:
value = parts[1] value = parts[1]
post_form += f'<input type="text" name="{escape(quote(name))}" value="{escape(quote(value))}"><br>' post_form += f'<input type="text" name="{name}" value="{value}"><br>'
post_form += '</form>' post_form += '</form>'
html_content = f""" html_content = f"""
<!DOCTYPE html> <!DOCTYPE html>
@ -424,4 +275,4 @@ def _post_request(req: V1RequestBase, driver: WebDriver):
<script>document.getElementById('hackForm').submit();</script> <script>document.getElementById('hackForm').submit();</script>
</body> </body>
</html>""" </html>"""
driver.get("data:text/html;charset=utf-8,{html_content}".format(html_content=html_content)) driver.get("data:text/html;charset=utf-8," + html_content)

View File

@ -1,32 +0,0 @@
import logging
from prometheus_client import Counter, Histogram, start_http_server
import time
REQUEST_COUNTER = Counter(
name='flaresolverr_request',
documentation='Total requests with result',
labelnames=['domain', 'result']
)
REQUEST_DURATION = Histogram(
name='flaresolverr_request_duration',
documentation='Request duration in seconds',
labelnames=['domain'],
buckets=[0, 10, 25, 50]
)
def serve(port):
start_http_server(port=port)
while True:
time.sleep(600)
def start_metrics_http_server(prometheus_port: int):
logging.info(f"Serving Prometheus exporter on http://0.0.0.0:{prometheus_port}/metrics")
from threading import Thread
Thread(
target=serve,
kwargs=dict(port=prometheus_port),
daemon=True,
).start()

View File

@ -1,84 +0,0 @@
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)
if utils.PLATFORM_VERSION == "nt":
session.driver.close()
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())

View File

@ -1,5 +1,4 @@
import unittest import unittest
from typing import Optional
from webtest import TestApp from webtest import TestApp
@ -8,7 +7,7 @@ import flaresolverr
import utils import utils
def _find_obj_by_key(key: str, value: str, _list: list) -> Optional[dict]: def _find_obj_by_key(key: str, value: str, _list: list) -> dict | None:
for obj in _list: for obj in _list:
if obj[key] == value: if obj[key] == value:
return obj return obj
@ -24,13 +23,10 @@ class TestFlareSolverr(unittest.TestCase):
cloudflare_url = "https://nowsecure.nl" cloudflare_url = "https://nowsecure.nl"
cloudflare_url_2 = "https://idope.se/torrent-list/harry/" cloudflare_url_2 = "https://idope.se/torrent-list/harry/"
ddos_guard_url = "https://anidex.info/" ddos_guard_url = "https://anidex.info/"
fairlane_url = "https://www.pararius.com/apartments/amsterdam"
custom_cloudflare_url = "https://www.muziekfabriek.org" custom_cloudflare_url = "https://www.muziekfabriek.org"
cloudflare_blocked_url = "https://cpasbiens3.fr/index.php?do=search&subaction=search" cloudflare_blocked_url = "https://cpasbiens3.fr/index.php?do=search&subaction=search"
app = TestApp(flaresolverr.app) app = TestApp(flaresolverr.app)
# wait until the server is ready
app.get('/')
def test_wrong_endpoint(self): def test_wrong_endpoint(self):
res = self.app.get('/wrong', status=404) res = self.app.get('/wrong', status=404)
@ -68,7 +64,7 @@ class TestFlareSolverr(unittest.TestCase):
self.assertEqual("Error: Request parameter 'cmd' = 'request.bad' is invalid.", body.message) self.assertEqual("Error: Request parameter 'cmd' = 'request.bad' is invalid.", body.message)
self.assertGreater(body.startTimestamp, 10000) self.assertGreater(body.startTimestamp, 10000)
self.assertGreaterEqual(body.endTimestamp, body.startTimestamp) 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): def test_v1_endpoint_request_get_no_cloudflare(self):
res = self.app.post_json('/v1', { res = self.app.post_json('/v1', {
@ -82,7 +78,7 @@ class TestFlareSolverr(unittest.TestCase):
self.assertEqual("Challenge not detected!", body.message) self.assertEqual("Challenge not detected!", body.message)
self.assertGreater(body.startTimestamp, 10000) self.assertGreater(body.startTimestamp, 10000)
self.assertGreaterEqual(body.endTimestamp, body.startTimestamp) 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 solution = body.solution
self.assertIn(self.google_url, solution.url) self.assertIn(self.google_url, solution.url)
@ -104,7 +100,7 @@ class TestFlareSolverr(unittest.TestCase):
self.assertEqual("Challenge solved!", body.message) self.assertEqual("Challenge solved!", body.message)
self.assertGreater(body.startTimestamp, 10000) self.assertGreater(body.startTimestamp, 10000)
self.assertGreaterEqual(body.endTimestamp, body.startTimestamp) 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 solution = body.solution
self.assertIn(self.cloudflare_url, solution.url) self.assertIn(self.cloudflare_url, solution.url)
@ -130,7 +126,7 @@ class TestFlareSolverr(unittest.TestCase):
self.assertEqual("Challenge solved!", body.message) self.assertEqual("Challenge solved!", body.message)
self.assertGreater(body.startTimestamp, 10000) self.assertGreater(body.startTimestamp, 10000)
self.assertGreaterEqual(body.endTimestamp, body.startTimestamp) 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 solution = body.solution
self.assertIn(self.cloudflare_url_2, solution.url) self.assertIn(self.cloudflare_url_2, solution.url)
@ -156,7 +152,7 @@ class TestFlareSolverr(unittest.TestCase):
self.assertEqual("Challenge solved!", body.message) self.assertEqual("Challenge solved!", body.message)
self.assertGreater(body.startTimestamp, 10000) self.assertGreater(body.startTimestamp, 10000)
self.assertGreaterEqual(body.endTimestamp, body.startTimestamp) 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 solution = body.solution
self.assertIn(self.ddos_guard_url, solution.url) self.assertIn(self.ddos_guard_url, solution.url)
@ -170,32 +166,6 @@ class TestFlareSolverr(unittest.TestCase):
self.assertIsNotNone(cf_cookie, "DDOS-Guard cookie not found") self.assertIsNotNone(cf_cookie, "DDOS-Guard cookie not found")
self.assertGreater(len(cf_cookie["value"]), 10) 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): def test_v1_endpoint_request_get_custom_cloudflare_js(self):
res = self.app.post_json('/v1', { res = self.app.post_json('/v1', {
"cmd": "request.get", "cmd": "request.get",
@ -208,7 +178,7 @@ class TestFlareSolverr(unittest.TestCase):
self.assertEqual("Challenge solved!", body.message) self.assertEqual("Challenge solved!", body.message)
self.assertGreater(body.startTimestamp, 10000) self.assertGreater(body.startTimestamp, 10000)
self.assertGreaterEqual(body.endTimestamp, body.startTimestamp) 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 solution = body.solution
self.assertIn(self.custom_cloudflare_url, solution.url) self.assertIn(self.custom_cloudflare_url, solution.url)
@ -239,45 +209,7 @@ class TestFlareSolverr(unittest.TestCase):
self.assertGreaterEqual(body.endTimestamp, body.startTimestamp) 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_cookies_param(self): # todo: test Cmd 'request.get' should return OK with 'cookies' param
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): def test_v1_endpoint_request_get_returnOnlyCookies_param(self):
res = self.app.post_json('/v1', { res = self.app.post_json('/v1', {
@ -302,124 +234,10 @@ class TestFlareSolverr(unittest.TestCase):
self.assertGreater(len(solution.cookies), 0) self.assertGreater(len(solution.cookies), 0)
self.assertIn("Chrome/", solution.userAgent) self.assertIn("Chrome/", solution.userAgent)
def test_v1_endpoint_request_get_proxy_http_param(self): # todo: test Cmd 'request.get' should return OK with HTTP 'proxy' param
""" # todo: test Cmd 'request.get' should return OK with HTTP 'proxy' param with credentials
To configure TinyProxy in local: # todo: test Cmd 'request.get' should return OK with SOCKSv5 'proxy' param
* sudo vim /etc/tinyproxy/tinyproxy.conf # todo: test Cmd 'request.get' should fail with wrong 'proxy' param
* 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_http_param_with_credentials(self):
"""
To configure TinyProxy in local:
* sudo vim /etc/tinyproxy/tinyproxy.conf
* edit => LogFile "/tmp/tinyproxy.log"
* edit => Syslog Off
* add => BasicAuth testuser testpass
* 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,
"username": "testuser",
"password": "testpass"
}
})
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): def test_v1_endpoint_request_get_fail_timeout(self):
res = self.app.post_json('/v1', { res = self.app.post_json('/v1', {
@ -533,99 +351,12 @@ class TestFlareSolverr(unittest.TestCase):
self.assertEqual(STATUS_OK, body.status) self.assertEqual(STATUS_OK, body.status)
self.assertEqual("Challenge not detected!", body.message) self.assertEqual("Challenge not detected!", body.message)
def test_v1_endpoint_sessions_create_without_session(self): # todo: test Cmd 'sessions.create' should return OK
res = self.app.post_json('/v1', { # todo: test Cmd 'sessions.create' should return OK with session
"cmd": "sessions.create" # todo: test Cmd 'sessions.list' should return OK
}) # todo: test Cmd 'sessions.destroy' should return OK
self.assertEqual(res.status_code, 200) # todo: test Cmd 'sessions.destroy' should fail
# 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__': if __name__ == '__main__':

View File

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

File diff suppressed because it is too large Load Diff

View File

@ -0,0 +1,259 @@
#!/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()

View File

@ -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 websockets import requests
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

View File

@ -1,193 +1,191 @@
import asyncio import asyncio
from collections.abc import Mapping import logging
from collections.abc import Sequence import time
from functools import wraps import traceback
import os 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)
if os.name == "nt": driver.quit()
driver.close()
driver.quit()

View File

@ -1,77 +1,75 @@
import atexit import multiprocessing
import logging import os
import multiprocessing import platform
import os import sys
import platform from subprocess import PIPE
import signal from subprocess import Popen
from subprocess import PIPE import atexit
from subprocess import Popen import traceback
import sys import logging
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
process = multiprocessing.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()
process.start() # receive pid from pipe
process.join() pid = reader.recv()
# receive pid from pipe REGISTERED.append(pid)
pid = reader.recv() # close pipes
REGISTERED.append(pid) writer.close()
# close pipes reader.close()
writer.close()
reader.close() return pid
process.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)

View File

@ -1,85 +1,70 @@
#!/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}
@staticmethod def handle_prefs(self, user_data_dir):
def _merge_nested(a, b): prefs = self.experimental_options.get("prefs")
""" if prefs:
merges b into a
leaf values in a are overwritten with values from b user_data_dir = user_data_dir or self._user_data_dir
""" default_path = os.path.join(user_data_dir, "Default")
for key in b: os.makedirs(default_path, exist_ok=True)
if key in a:
if isinstance(a[key], dict) and isinstance(b[key], dict): # undot prefs dict keys
ChromeOptions._merge_nested(a[key], b[key]) undot_prefs = {}
continue for key, value in prefs.items():
a[key] = b[key] undot_prefs.update(self._undot_key(key, value))
return a
prefs_file = os.path.join(default_path, "Preferences")
def handle_prefs(self, user_data_dir): if os.path.exists(prefs_file):
prefs = self.experimental_options.get("prefs") with open(prefs_file, encoding="latin1", mode="r") as f:
if prefs: undot_prefs.update(json.load(f))
user_data_dir = user_data_dir or self._user_data_dir
default_path = os.path.join(user_data_dir, "Default") with open(prefs_file, encoding="latin1", mode="w") as f:
os.makedirs(default_path, exist_ok=True) json.dump(undot_prefs, f)
# undot prefs dict keys # remove the experimental_options to avoid an error
undot_prefs = {} del self._experimental_options["prefs"]
for key, value in prefs.items():
undot_prefs = self._merge_nested( @classmethod
undot_prefs, self._undot_key(key, value) def from_options(cls, options):
) o = cls()
o.__dict__.update(options.__dict__)
prefs_file = os.path.join(default_path, "Preferences") return o
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

View File

@ -1,451 +1,276 @@
#!/usr/bin/env python3 #!/usr/bin/env python3
# this module is part of undetected_chromedriver # this module is part of undetected_chromedriver
from distutils.version import LooseVersion import io
import io import logging
import json import os
import logging import random
import os import re
import pathlib import string
import platform import sys
import random import time
import re import zipfile
import shutil from distutils.version import LooseVersion
import string from urllib.request import urlopen, urlretrieve
import sys import secrets
import time
from urllib.request import urlopen
from urllib.request import urlretrieve logger = logging.getLogger(__name__)
import zipfile
from multiprocessing import Lock IS_POSIX = sys.platform.startswith(("darwin", "cygwin", "linux"))
logger = logging.getLogger(__name__)
class Patcher(object):
IS_POSIX = sys.platform.startswith(("darwin", "cygwin", "linux", "linux2", "freebsd")) url_repo = "https://chromedriver.storage.googleapis.com"
zip_name = "chromedriver_%s.zip"
exe_name = "chromedriver%s"
class Patcher(object):
lock = Lock() platform = sys.platform
exe_name = "chromedriver%s" if platform.endswith("win32"):
zip_name %= "win32"
platform = sys.platform exe_name %= ".exe"
if platform.endswith("win32"): if platform.endswith("linux"):
d = "~/appdata/roaming/undetected_chromedriver" zip_name %= "linux64"
elif "LAMBDA_TASK_ROOT" in os.environ: exe_name %= ""
d = "/tmp/undetected_chromedriver" if platform.endswith("darwin"):
elif platform.startswith(("linux", "linux2")): zip_name %= "mac64"
d = "~/.local/share/undetected_chromedriver" exe_name %= ""
elif platform.endswith("darwin"):
d = "~/Library/Application Support/undetected_chromedriver" if platform.endswith("win32"):
else: d = "~/appdata/roaming/undetected_chromedriver"
d = "~/.undetected_chromedriver" elif platform.startswith("linux"):
data_path = os.path.abspath(os.path.expanduser(d)) d = "~/.local/share/undetected_chromedriver"
elif platform.endswith("darwin"):
def __init__( d = "~/Library/Application Support/undetected_chromedriver"
self, else:
executable_path=None, d = "~/.undetected_chromedriver"
force=False, data_path = os.path.abspath(os.path.expanduser(d))
version_main: int = 0,
user_multi_procs=False, def __init__(self, executable_path=None, force=False, version_main: int = 0):
): """
"""
Args: Args:
executable_path: None = automatic executable_path: None = automatic
a full file path to the chromedriver executable a full file path to the chromedriver executable
force: False force: False
terminate processes which are holding lock terminate processes which are holding lock
version_main: 0 = auto version_main: 0 = auto
specify main chrome version (rounded, ex: 82) specify main chrome version (rounded, ex: 82)
""" """
self.force = force
self._custom_exe_path = False self.force = force
prefix = "undetected" self.executable_path = None
self.user_multi_procs = user_multi_procs prefix = secrets.token_hex(8)
try: if not os.path.exists(self.data_path):
# Try to convert version_main into an integer os.makedirs(self.data_path, exist_ok=True)
version_main_int = int(version_main)
# check if version_main_int is less than or equal to e.g 114 if not executable_path:
self.is_old_chromedriver = version_main and version_main_int <= 114 self.executable_path = os.path.join(
except (ValueError,TypeError): self.data_path, "_".join([prefix, self.exe_name])
# Check not running inside Docker )
if not os.path.exists("/app/chromedriver"):
# If the conversion fails, log an error message if not IS_POSIX:
logging.info("version_main cannot be converted to an integer") if executable_path:
# Set self.is_old_chromedriver to False if the conversion fails if not executable_path[-4:] == ".exe":
self.is_old_chromedriver = False executable_path += ".exe"
# Needs to be called before self.exe_name is accessed self.zip_path = os.path.join(self.data_path, prefix)
self._set_platform_name()
if not executable_path:
if not os.path.exists(self.data_path): self.executable_path = os.path.abspath(
os.makedirs(self.data_path, exist_ok=True) os.path.join(".", self.executable_path)
)
if not executable_path:
if sys.platform.startswith("freebsd"): self._custom_exe_path = False
self.executable_path = os.path.join(
self.data_path, self.exe_name if executable_path:
) self._custom_exe_path = True
else: self.executable_path = executable_path
self.executable_path = os.path.join( self.version_main = version_main
self.data_path, "_".join([prefix, self.exe_name]) self.version_full = None
)
def auto(self, executable_path=None, force=False, version_main=None):
if not IS_POSIX: """"""
if executable_path: if executable_path:
if not executable_path[-4:] == ".exe": self.executable_path = executable_path
executable_path += ".exe" self._custom_exe_path = True
self.zip_path = os.path.join(self.data_path, prefix) if self._custom_exe_path:
ispatched = self.is_binary_patched(self.executable_path)
if not executable_path: if not ispatched:
if not self.user_multi_procs: return self.patch_exe()
self.executable_path = os.path.abspath( else:
os.path.join(".", self.executable_path) return
)
if version_main:
if executable_path: self.version_main = version_main
self._custom_exe_path = True if force is True:
self.executable_path = executable_path self.force = force
# Set the correct repository to download the Chromedriver from try:
if self.is_old_chromedriver: os.unlink(self.executable_path)
self.url_repo = "https://chromedriver.storage.googleapis.com" except PermissionError:
else: if self.force:
self.url_repo = "https://googlechromelabs.github.io/chrome-for-testing" self.force_kill_instances(self.executable_path)
return self.auto(force=not self.force)
self.version_main = version_main try:
self.version_full = None if self.is_binary_patched():
# assumes already running AND patched
def _set_platform_name(self): return True
""" except PermissionError:
Set the platform and exe name based on the platform undetected_chromedriver is running on pass
in order to download the correct chromedriver. # return False
""" except FileNotFoundError:
if self.platform.endswith("win32"): pass
self.platform_name = "win32"
self.exe_name %= ".exe" release = self.fetch_release_number()
if self.platform.endswith(("linux", "linux2")): self.version_main = release.version[0]
self.platform_name = "linux64" self.version_full = release
self.exe_name %= "" self.unzip_package(self.fetch_package())
if self.platform.endswith("darwin"): return self.patch()
if self.is_old_chromedriver:
self.platform_name = "mac64" def patch(self):
else: self.patch_exe()
self.platform_name = "mac-x64" return self.is_binary_patched()
self.exe_name %= ""
if self.platform.startswith("freebsd"): def fetch_release_number(self):
self.platform_name = "freebsd" """
self.exe_name %= "" Gets the latest major version available, or the latest major version of self.target_version if set explicitly.
:return: version string
def auto(self, executable_path=None, force=False, version_main=None, _=None): :rtype: LooseVersion
""" """
path = "/latest_release"
Args: if self.version_main:
executable_path: path += f"_{self.version_main}"
force: path = path.upper()
version_main: logger.debug("getting release number from %s" % path)
return LooseVersion(urlopen(self.url_repo + path).read().decode())
Returns:
def parse_exe_version(self):
""" with io.open(self.executable_path, "rb") as f:
p = pathlib.Path(self.data_path) for line in iter(lambda: f.readline(), b""):
if self.user_multi_procs: match = re.search(rb"platform_handle\x00content\x00([0-9.]*)", line)
with Lock(): if match:
files = list(p.rglob("*chromedriver*")) return LooseVersion(match[1].decode())
most_recent = max(files, key=lambda f: f.stat().st_mtime)
files.remove(most_recent) def fetch_package(self):
list(map(lambda f: f.unlink(), files)) """
if self.is_binary_patched(most_recent): Downloads ChromeDriver from source
self.executable_path = str(most_recent)
return True :return: path to downloaded file
"""
if executable_path: u = "%s/%s/%s" % (self.url_repo, self.version_full.vstring, self.zip_name)
self.executable_path = executable_path logger.debug("downloading from %s" % u)
self._custom_exe_path = True # return urlretrieve(u, filename=self.data_path)[0]
return urlretrieve(u)[0]
if self._custom_exe_path:
ispatched = self.is_binary_patched(self.executable_path) def unzip_package(self, fp):
if not ispatched: """
return self.patch_exe() Does what it says
else:
return :return: path to unpacked executable
"""
if version_main: logger.debug("unzipping %s" % fp)
self.version_main = version_main try:
if force is True: os.unlink(self.zip_path)
self.force = force except (FileNotFoundError, OSError):
pass
if self.platform_name == "freebsd": os.makedirs(self.zip_path, mode=0o755, exist_ok=True)
chromedriver_path = shutil.which("chromedriver") with zipfile.ZipFile(fp, mode="r") as zf:
zf.extract(self.exe_name, self.zip_path)
if not os.path.isfile(chromedriver_path) or not os.access(chromedriver_path, os.X_OK): os.rename(os.path.join(self.zip_path, self.exe_name), self.executable_path)
logging.error("Chromedriver not installed!") os.remove(fp)
return os.rmdir(self.zip_path)
os.chmod(self.executable_path, 0o755)
version_path = os.path.join(os.path.dirname(self.executable_path), "version.txt") return self.executable_path
process = os.popen(f'"{chromedriver_path}" --version') @staticmethod
chromedriver_version = process.read().split(' ')[1].split(' ')[0] def force_kill_instances(exe_name):
process.close() """
kills running instances.
current_version = None :param: executable name to kill, may be a path as well
if os.path.isfile(version_path) or os.access(version_path, os.X_OK):
with open(version_path, 'r') as f: :return: True on success else False
current_version = f.read() """
exe_name = os.path.basename(exe_name)
if current_version != chromedriver_version: if IS_POSIX:
logging.info("Copying chromedriver executable...") r = os.system("kill -f -9 $(pidof %s)" % exe_name)
shutil.copy(chromedriver_path, self.executable_path) else:
os.chmod(self.executable_path, 0o755) r = os.system("taskkill /f /im %s" % exe_name)
return not r
with open(version_path, 'w') as f:
f.write(chromedriver_version) @staticmethod
def gen_random_cdc():
logging.info("Chromedriver executable copied!") cdc = random.choices(string.ascii_lowercase, k=26)
else: cdc[-6:-4] = map(str.upper, cdc[-6:-4])
try: cdc[2] = cdc[0]
os.unlink(self.executable_path) cdc[3] = "_"
except PermissionError: return "".join(cdc).encode()
if self.force:
self.force_kill_instances(self.executable_path) def is_binary_patched(self, executable_path=None):
return self.auto(force=not self.force) """simple check if executable is patched.
try:
if self.is_binary_patched(): :return: False if not patched, else True
# assumes already running AND patched """
return True executable_path = executable_path or self.executable_path
except PermissionError: with io.open(executable_path, "rb") as fh:
pass for line in iter(lambda: fh.readline(), b""):
# return False if b"cdc_" in line:
except FileNotFoundError: return False
pass else:
return True
release = self.fetch_release_number()
self.version_main = release.version[0] def patch_exe(self):
self.version_full = release """
self.unzip_package(self.fetch_package()) Patches the ChromeDriver binary
return self.patch() :return: False on failure, binary name on success
"""
def driver_binary_in_use(self, path: str = None) -> bool: logger.info("patching driver executable %s" % self.executable_path)
"""
naive test to check if a found chromedriver binary is linect = 0
currently in use replacement = self.gen_random_cdc()
with io.open(self.executable_path, "r+b") as fh:
Args: for line in iter(lambda: fh.readline(), b""):
path: a string or PathLike object to the binary to check. if b"cdc_" in line:
if not specified, we check use this object's executable_path fh.seek(-len(line), 1)
""" newline = re.sub(b"cdc_.{22}", replacement, line)
if not path: fh.write(newline)
path = self.executable_path linect += 1
p = pathlib.Path(path) return linect
if not p.exists(): def __repr__(self):
raise OSError("file does not exist: %s" % p) return "{0:s}({1:s})".format(
try: self.__class__.__name__,
with open(p, mode="a+b") as fs: self.executable_path,
exc = [] )
try:
def __del__(self):
fs.seek(0, 0)
except PermissionError as e: if self._custom_exe_path:
exc.append(e) # since some systems apprently allow seeking # if the driver binary is specified by user
# we conduct another test # we assume it is important enough to not delete it
try: return
fs.readline() else:
except PermissionError as e: timeout = 3 # stop trying after this many seconds
exc.append(e) t = time.monotonic()
while True:
if exc: now = time.monotonic()
if now - t > timeout:
return True # we don't want to wait until the end of time
return False logger.debug(
# ok safe to assume this is in use "could not unlink %s in time (%d seconds)"
except Exception as e: % (self.executable_path, timeout)
# logger.exception("whoops ", e) )
pass break
try:
def cleanup_unused_files(self): os.unlink(self.executable_path)
p = pathlib.Path(self.data_path) logger.debug("successfully unlinked %s" % self.executable_path)
items = list(p.glob("*undetected*")) break
for item in items: except (OSError, RuntimeError, PermissionError):
try: time.sleep(0.1)
item.unlink() continue
except: except FileNotFoundError:
pass break
def patch(self):
self.patch_exe()
return self.is_binary_patched()
def fetch_release_number(self):
"""
Gets the latest major version available, or the latest major version of self.target_version if set explicitly.
:return: version string
:rtype: LooseVersion
"""
# Endpoint for old versions of Chromedriver (114 and below)
if self.is_old_chromedriver:
path = f"/latest_release_{self.version_main}"
path = path.upper()
logger.debug("getting release number from %s" % path)
return LooseVersion(urlopen(self.url_repo + path).read().decode())
# Endpoint for new versions of Chromedriver (115+)
if not self.version_main:
# Fetch the latest version
path = "/last-known-good-versions-with-downloads.json"
logger.debug("getting release number from %s" % path)
with urlopen(self.url_repo + path) as conn:
response = conn.read().decode()
last_versions = json.loads(response)
return LooseVersion(last_versions["channels"]["Stable"]["version"])
# Fetch the latest minor version of the major version provided
path = "/latest-versions-per-milestone-with-downloads.json"
logger.debug("getting release number from %s" % path)
with urlopen(self.url_repo + path) as conn:
response = conn.read().decode()
major_versions = json.loads(response)
return LooseVersion(major_versions["milestones"][str(self.version_main)]["version"])
def parse_exe_version(self):
with io.open(self.executable_path, "rb") as f:
for line in iter(lambda: f.readline(), b""):
match = re.search(rb"platform_handle\x00content\x00([0-9.]*)", line)
if match:
return LooseVersion(match[1].decode())
def fetch_package(self):
"""
Downloads ChromeDriver from source
:return: path to downloaded file
"""
zip_name = f"chromedriver_{self.platform_name}.zip"
if self.is_old_chromedriver:
download_url = "%s/%s/%s" % (self.url_repo, self.version_full.vstring, zip_name)
else:
zip_name = zip_name.replace("_", "-", 1)
download_url = "https://storage.googleapis.com/chrome-for-testing-public/%s/%s/%s"
download_url %= (self.version_full.vstring, self.platform_name, zip_name)
logger.debug("downloading from %s" % download_url)
return urlretrieve(download_url)[0]
def unzip_package(self, fp):
"""
Does what it says
:return: path to unpacked executable
"""
exe_path = self.exe_name
if not self.is_old_chromedriver:
# The new chromedriver unzips into its own folder
zip_name = f"chromedriver-{self.platform_name}"
exe_path = os.path.join(zip_name, self.exe_name)
logger.debug("unzipping %s" % fp)
try:
os.unlink(self.zip_path)
except (FileNotFoundError, OSError):
pass
os.makedirs(self.zip_path, mode=0o755, exist_ok=True)
with zipfile.ZipFile(fp, mode="r") as zf:
zf.extractall(self.zip_path)
os.rename(os.path.join(self.zip_path, exe_path), self.executable_path)
os.remove(fp)
shutil.rmtree
os.chmod(self.executable_path, 0o755)
return self.executable_path
@staticmethod
def force_kill_instances(exe_name):
"""
kills running instances.
:param: executable name to kill, may be a path as well
:return: True on success else False
"""
exe_name = os.path.basename(exe_name)
if IS_POSIX:
r = os.system("kill -f -9 $(pidof %s)" % exe_name)
else:
r = os.system("taskkill /f /im %s" % exe_name)
return not r
@staticmethod
def gen_random_cdc():
cdc = random.choices(string.ascii_letters, k=27)
return "".join(cdc).encode()
def is_binary_patched(self, executable_path=None):
executable_path = executable_path or self.executable_path
try:
with io.open(executable_path, "rb") as fh:
return fh.read().find(b"undetected chromedriver") != -1
except FileNotFoundError:
return False
def patch_exe(self):
start = time.perf_counter()
logger.info("patching driver executable %s" % self.executable_path)
with io.open(self.executable_path, "r+b") as fh:
content = fh.read()
# match_injected_codeblock = re.search(rb"{window.*;}", content)
match_injected_codeblock = re.search(rb"\{window\.cdc.*?;\}", content)
if match_injected_codeblock:
target_bytes = match_injected_codeblock[0]
new_target_bytes = (
b'{console.log("undetected chromedriver 1337!")}'.ljust(
len(target_bytes), b" "
)
)
new_content = content.replace(target_bytes, new_target_bytes)
if new_content == content:
logger.warning(
"something went wrong patching the driver binary. could not find injection code block"
)
else:
logger.debug(
"found block:\n%s\nreplacing with:\n%s"
% (target_bytes, new_target_bytes)
)
fh.seek(0)
fh.write(new_content)
logger.debug(
"patching took us {:.2f} seconds".format(time.perf_counter() - start)
)
def __repr__(self):
return "{0:s}({1:s})".format(
self.__class__.__name__,
self.executable_path,
)
def __del__(self):
if self._custom_exe_path:
# if the driver binary is specified by user
# we assume it is important enough to not delete it
return
else:
timeout = 3 # stop trying after this many seconds
t = time.monotonic()
now = lambda: time.monotonic()
while now() - t > timeout:
# we don't want to wait until the end of time
try:
if self.user_multi_procs:
break
os.unlink(self.executable_path)
logger.debug("successfully unlinked %s" % self.executable_path)
break
except (OSError, RuntimeError, PermissionError):
time.sleep(0.01)
continue
except FileNotFoundError:
break

View File

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

View File

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

View File

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

View File

@ -1,19 +1,13 @@
import json import json
import logging import logging
import os import os
import platform
import re import re
import shutil import shutil
import sys
import tempfile
import urllib.parse
from selenium.webdriver.chrome.webdriver import WebDriver from selenium.webdriver.chrome.webdriver import WebDriver
import undetected_chromedriver as uc import undetected_chromedriver as uc
FLARESOLVERR_VERSION = None FLARESOLVERR_VERSION = None
PLATFORM_VERSION = None
CHROME_EXE_PATH = None
CHROME_MAJOR_VERSION = None CHROME_MAJOR_VERSION = None
USER_AGENT = None USER_AGENT = None
XVFB_DISPLAY = None XVFB_DISPLAY = None
@ -34,139 +28,24 @@ def get_flaresolverr_version() -> str:
return FLARESOLVERR_VERSION return FLARESOLVERR_VERSION
package_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), os.pardir, 'package.json') 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: with open(package_path) as f:
FLARESOLVERR_VERSION = json.loads(f.read())['version'] FLARESOLVERR_VERSION = json.loads(f.read())['version']
return FLARESOLVERR_VERSION return FLARESOLVERR_VERSION
def get_current_platform() -> str:
global PLATFORM_VERSION
if PLATFORM_VERSION is not None:
return PLATFORM_VERSION
PLATFORM_VERSION = os.name
return PLATFORM_VERSION
def get_webdriver() -> WebDriver:
def create_proxy_extension(proxy: dict) -> str: global PATCHED_DRIVER_PATH
parsed_url = urllib.parse.urlparse(proxy['url'])
scheme = parsed_url.scheme
host = parsed_url.hostname
port = parsed_url.port
username = proxy['username']
password = proxy['password']
manifest_json = """
{
"version": "1.0.0",
"manifest_version": 2,
"name": "Chrome Proxy",
"permissions": [
"proxy",
"tabs",
"unlimitedStorage",
"storage",
"<all_urls>",
"webRequest",
"webRequestBlocking"
],
"background": {"scripts": ["background.js"]},
"minimum_chrome_version": "76.0.0"
}
"""
background_js = """
var config = {
mode: "fixed_servers",
rules: {
singleProxy: {
scheme: "%s",
host: "%s",
port: %d
},
bypassList: ["localhost"]
}
};
chrome.proxy.settings.set({value: config, scope: "regular"}, function() {});
function callbackFn(details) {
return {
authCredentials: {
username: "%s",
password: "%s"
}
};
}
chrome.webRequest.onAuthRequired.addListener(
callbackFn,
{ urls: ["<all_urls>"] },
['blocking']
);
""" % (
scheme,
host,
port,
username,
password
)
proxy_extension_dir = tempfile.mkdtemp()
with open(os.path.join(proxy_extension_dir, "manifest.json"), "w") as f:
f.write(manifest_json)
with open(os.path.join(proxy_extension_dir, "background.js"), "w") as f:
f.write(background_js)
return proxy_extension_dir
def get_webdriver(proxy: dict = None) -> WebDriver:
global PATCHED_DRIVER_PATH, USER_AGENT
logging.debug('Launching web browser...') logging.debug('Launching web browser...')
# undetected_chromedriver # undetected_chromedriver
options = uc.ChromeOptions() options = uc.ChromeOptions()
options.add_argument('--no-sandbox') options.add_argument('--no-sandbox')
options.add_argument('--window-size=1920,1080') options.add_argument('--window-size=1920,1080')
options.add_argument('--disable-search-engine-choice-screen')
# todo: this param shows a warning in chrome head-full # todo: this param shows a warning in chrome head-full
options.add_argument('--disable-setuid-sandbox') options.add_argument('--disable-setuid-sandbox')
options.add_argument('--disable-dev-shm-usage') 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
IS_ARMARCH = platform.machine().startswith(('arm', 'aarch'))
if IS_ARMARCH:
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')
language = os.environ.get('LANG', None) # note: headless mode is detected (options.headless = True)
if language is not None:
options.add_argument('--accept-lang=%s' % language)
# Fix for Chrome 117 | https://github.com/FlareSolverr/FlareSolverr/issues/910
if USER_AGENT is not None:
options.add_argument('--user-agent=%s' % USER_AGENT)
proxy_extension_dir = None
if proxy and all(key in proxy for key in ['url', 'username', 'password']):
proxy_extension_dir = create_proxy_extension(proxy)
options.add_argument("--load-extension=%s" % os.path.abspath(proxy_extension_dir))
elif 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 (headless = True)
# we launch the browser in head-full mode with the window hidden # we launch the browser in head-full mode with the window hidden
windows_headless = False windows_headless = False
if get_config_headless(): if get_config_headless():
@ -174,8 +53,6 @@ def get_webdriver(proxy: dict = None) -> WebDriver:
windows_headless = True windows_headless = True
else: else:
start_xvfb_display() start_xvfb_display()
# For normal headless mode:
# options.add_argument('--headless')
# if we are inside the Docker container, we avoid downloading the driver # if we are inside the Docker container, we avoid downloading the driver
driver_exe_path = None driver_exe_path = None
@ -188,27 +65,15 @@ def get_webdriver(proxy: dict = None) -> WebDriver:
if PATCHED_DRIVER_PATH is not None: if PATCHED_DRIVER_PATH is not None:
driver_exe_path = PATCHED_DRIVER_PATH driver_exe_path = PATCHED_DRIVER_PATH
# detect chrome path
browser_executable_path = get_chrome_exe_path()
# downloads and patches the chromedriver # downloads and patches the chromedriver
# if we don't set driver_executable_path it downloads, patches, and deletes the driver each time # if we don't set driver_executable_path it downloads, patches, and deletes the driver each time
try: 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, windows_headless=windows_headless)
driver_executable_path=driver_exe_path, version_main=version_main,
windows_headless=windows_headless, headless=get_config_headless())
except Exception as e:
logging.error("Error starting Chrome: %s" % e)
# save the patched driver to avoid re-downloads # save the patched driver to avoid re-downloads
if driver_exe_path is None: if driver_exe_path is None:
PATCHED_DRIVER_PATH = os.path.join(driver.patcher.data_path, driver.patcher.exe_name) PATCHED_DRIVER_PATH = os.path.join(driver.patcher.data_path, driver.patcher.exe_name)
if PATCHED_DRIVER_PATH != driver.patcher.executable_path: shutil.copy(driver.patcher.executable_path, PATCHED_DRIVER_PATH)
shutil.copy(driver.patcher.executable_path, PATCHED_DRIVER_PATH)
# clean up proxy extension directory
if proxy_extension_dir is not None:
shutil.rmtree(proxy_extension_dir)
# selenium vanilla # selenium vanilla
# options = webdriver.ChromeOptions() # options = webdriver.ChromeOptions()
@ -221,45 +86,23 @@ def get_webdriver(proxy: dict = None) -> WebDriver:
return driver 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: def get_chrome_major_version() -> str:
global CHROME_MAJOR_VERSION global CHROME_MAJOR_VERSION
if CHROME_MAJOR_VERSION is not None: if CHROME_MAJOR_VERSION is not None:
return CHROME_MAJOR_VERSION return CHROME_MAJOR_VERSION
if os.name == 'nt': if os.name == 'nt':
# Example: '104.0.5112.79'
try: try:
complete_version = extract_version_nt_executable(get_chrome_exe_path()) stream = os.popen(
'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: except Exception:
try: # Example: '104.0.5112.79'
complete_version = extract_version_nt_registry() complete_version = extract_version_folder()
except Exception:
# Example: '104.0.5112.79'
complete_version = extract_version_nt_folder()
else: else:
chrome_path = get_chrome_exe_path() chrome_path = uc.find_chrome_executable()
process = os.popen(f'"{chrome_path}" --version') process = os.popen(f'"{chrome_path}" --version')
# Example 1: 'Chromium 104.0.5112.79 Arch Linux\n' # Example 1: 'Chromium 104.0.5112.79 Arch Linux\n'
# Example 2: 'Google Chrome 104.0.5112.79 Arch Linux\n' # Example 2: 'Google Chrome 104.0.5112.79 Arch Linux\n'
@ -267,32 +110,24 @@ def get_chrome_major_version() -> str:
process.close() process.close()
CHROME_MAJOR_VERSION = complete_version.split('.')[0].split(' ')[-1] CHROME_MAJOR_VERSION = complete_version.split('.')[0].split(' ')[-1]
logging.info(f"Chrome major version: {CHROME_MAJOR_VERSION}")
return CHROME_MAJOR_VERSION return CHROME_MAJOR_VERSION
def extract_version_nt_executable(exe_path: str) -> str: def extract_version_registry(output) -> str:
import pefile try:
pe = pefile.PE(exe_path, fast_load=True) google_version = ''
pe.parse_data_directories( for letter in output[output.rindex('DisplayVersion REG_SZ') + 24:]:
directories=[pefile.DIRECTORY_ENTRY["IMAGE_DIRECTORY_ENTRY_RESOURCE"]] if letter != '\n':
) google_version += letter
return pe.FileInfo[0][0].StringTable[0].entries[b"FileVersion"].decode('utf-8') else:
break
return google_version.strip()
except TypeError:
return ''
def extract_version_nt_registry() -> str: def extract_version_folder() -> 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. # Check if the Chrome folder exists in the x32 or x64 Program Files folders.
for i in range(2): for i in range(2):
path = 'C:\\Program Files' + (' (x86)' if i else '') + '\\Google\\Chrome\\Application' path = 'C:\\Program Files' + (' (x86)' if i else '') + '\\Google\\Chrome\\Application'
@ -317,15 +152,11 @@ def get_user_agent(driver=None) -> str:
if driver is None: if driver is None:
driver = get_webdriver() driver = get_webdriver()
USER_AGENT = driver.execute_script("return navigator.userAgent") USER_AGENT = driver.execute_script("return navigator.userAgent")
# Fix for Chrome 117 | https://github.com/FlareSolverr/FlareSolverr/issues/910
USER_AGENT = re.sub('HEADLESS', '', USER_AGENT, flags=re.IGNORECASE)
return USER_AGENT return USER_AGENT
except Exception as e: except Exception as e:
raise Exception("Error getting browser User-Agent. " + str(e)) raise Exception("Error getting browser User-Agent. " + str(e))
finally: finally:
if driver is not None: if driver is not None:
if PLATFORM_VERSION == "nt":
driver.close()
driver.quit() driver.quit()