Compare commits

..

53 Commits

Author SHA1 Message Date
ilike2burnthing
b903a5dd84 Bump version 3.0.4 2023-03-07 03:54:10 +00:00
ilike2burnthing
7e9d5f424f Update changelog 2023-03-07 03:54:05 +00:00
ilike2burnthing
fc6d2d9095 Click on the Cloudflare's 'Verify you are human' button if necessary
Co-authored-by: furdarius <furdarius@users.noreply.github.com>

69e023b946 minus screenshot function
2023-03-07 03:51:17 +00:00
ilike2burnthing
aef9b2d4d6 Update changelog 2023-03-06 14:02:56 +00:00
ilike2burnthing
6dc279a9d3 Bump version 3.0.3 2023-03-06 13:59:20 +00:00
Artemiy Ryabinkov
96fcd21174 Update undetected_chromedriver version to 3.4.6 (#715)
Co-authored-by: ilike2burnthing <59480337+ilike2burnthing@users.noreply.github.com>
2023-03-06 13:57:38 +00:00
ngosang
3a6e8e0f92 Update GitHub bug report template 2023-01-28 18:00:57 +01:00
ilike2burnthing
2d97f88276 Update README.md 2023-01-09 21:51:49 +00:00
ngosang
ac5c64319e Bump version 3.0.2 2023-01-08 20:48:20 +01:00
ngosang
c93834e2f0 Check Chrome / Chromium web browser is installed correctly 2023-01-08 20:46:11 +01:00
ngosang
e3b4200d94 Detect Cloudflare blocked access 2023-01-08 20:40:10 +01:00
ngosang
0941861f80 Update changelog 2023-01-06 18:49:05 +01:00
ngosang
8a10eb27a6 Bump version 3.0.1 2023-01-06 18:33:02 +01:00
ngosang
e9c08c84ef Update GitHub actions 2023-01-06 18:32:34 +01:00
ngosang
2aa1744476 Add more selectors to detect blocked access 2023-01-06 18:12:26 +01:00
ngosang
a89679a52d Disable Zygote sandbox in Chromium browser 2023-01-06 18:05:23 +01:00
ngosang
410ee7981f Apply undetected-chromedriver patches
* Hide Chrome window in Windows/NT
* Not use subprocess by default (independent process)
* Kill Chromium processes properly to avoid defunct/zombie processes
2023-01-06 17:50:52 +01:00
ngosang
e163019f28 Update undetected-chromedriver 2023-01-06 17:19:11 +01:00
ngosang
7d84f1b663 Kill Chromium processes properly to avoid defunct/zombie processes 2023-01-06 13:58:24 +01:00
ngosang
4807e9dbe2 Include procps (ps), curl and vim packages in the Docker image 2023-01-05 13:25:45 +01:00
ngosang
85360df336 Bump version 3.0.0 2023-01-05 01:45:54 +01:00
ngosang
fd42fcee66 Detect challenges by title 2023-01-05 01:45:43 +01:00
ngosang
90a831a13b Update changelog 2023-01-04 22:38:46 +01:00
ngosang
904072267a Update GitHub actions to build linux/386 image 2023-01-04 22:38:38 +01:00
ngosang
f63816905b Update Dockerfile with latest Python and Chrome 2023-01-04 22:37:38 +01:00
ngosang
258225a091 Detect Cloudflare access denied 2023-01-04 22:15:19 +01:00
ngosang
da2263c85b Clean up Dockerfile 2023-01-04 21:59:19 +01:00
ngosang
70a423805d Fix selector for new challenge 2023-01-04 21:28:04 +01:00
ngosang
99d074084a Fix Dockerfile for linux/386 architecture 2023-01-04 21:05:49 +01:00
ngosang
c7251da54f Install undetected_chromedriver dependencies 2023-01-04 21:05:42 +01:00
ngosang
676a930f02 Reuse patched chromedriver 2023-01-04 21:05:32 +01:00
ngosang
ecaac2e1d9 Fix Chrome version detection on Windows 2023-01-04 21:05:27 +01:00
ngosang
60a22625be Add browser headless mode for Windows 2023-01-04 21:05:19 +01:00
ngosang
b32d7b70be Fork undetected-chromedriver 3.1.5.post4 2023-01-04 21:05:12 +01:00
ngosang
f10f7269ca Reduce Docker image size 2023-01-04 21:05:07 +01:00
ngosang
33bed9428e Update readme 2023-01-04 21:05:01 +01:00
ngosang
5092b1eef9 Add browser headless mode for Linux 2023-01-04 21:04:38 +01:00
ngosang
0cfcb57e3a Add tests for several known sites 2023-01-04 21:04:31 +01:00
ngosang
59bd7c8bdd Show ReqId only in Debug traces 2023-01-04 21:04:26 +01:00
ngosang
b90d263159 Detect Cloudflare Access Denied 2023-01-04 21:04:18 +01:00
ngosang
5257f47d34 Update readme 2023-01-04 21:04:08 +01:00
ngosang
613e49e6cb Add Docker image and Docker compose 2023-01-04 21:03:54 +01:00
ngosang
1505595591 Rewrite FlareSolverr from scratch in Python + Selenium 2023-01-04 21:03:40 +01:00
ngosang
8d1ac09bf2 Update package.json 2023-01-04 21:03:31 +01:00
ngosang
31265a510d Update license, remove FlareSolverr v1 / v2 authors 2023-01-04 21:02:14 +01:00
ngosang
e967e135a3 Prepare .gitignore for Python project 2023-01-04 21:01:37 +01:00
ngosang
383025032b Prepare for version 3.0, remove JS code 2023-01-04 20:57:23 +01:00
ilike2burnthing
345628e3e4 issue template: switch to Jackett-style required form (#631) 2022-12-21 01:59:51 +00:00
ilike2burnthing
35c2f09202 Bump version 2.2.10 (#560) 2022-10-22 14:39:18 +01:00
Athorcis
197258e921 detect DDos-Guard through title content. resolves #546 (#559) 2022-10-22 14:34:29 +01:00
ngosang
c99101f74b Bump version 2.2.9 2022-09-25 00:58:55 +02:00
ngosang
5703caa9d3 Commit the complete changelog 2022-09-25 00:57:31 +02:00
ngosang
aa254eb830 Detect Cloudflare Access Denied 2022-09-25 00:35:39 +02:00
25 changed files with 2276 additions and 1948 deletions

View File

@@ -1,32 +0,0 @@
**Please use the search bar** at the top of the page and make sure you are not creating an already submitted issue.
Check closed issues as well, because your issue may have already been fixed.
### How to enable debug and html traces
[Follow the instructions from this wiki page](https://github.com/FlareSolverr/FlareSolverr/wiki/How-to-enable-debug-and-html-trace)
### Environment
* **FlareSolverr version**:
* **Last working FlareSolverr version**:
* **Operating system**:
* **Are you using Docker**: [yes/no]
* **FlareSolverr User-Agent (see log traces or / endpoint)**:
* **Are you using a proxy or VPN?** [yes/no]
* **Are you using Captcha Solver:** [yes/no]
* **If using captcha solver, which one:**
* **URL to test this issue:**
### Description
[List steps to reproduce the error and details on what happens and what you expected to happen]
### Logged Error Messages
[Place any relevant error messages you noticed from the logs here.]
[Make sure you attach the full logs with your personal information removed in case we need more information]
### Screenshots
[Place any screenshots of the issue here if needed]

64
.github/ISSUE_TEMPLATE/bug_report.yml vendored Normal file
View File

@@ -0,0 +1,64 @@
name: Bug report
description: Create a report of your issue
body:
- type: checkboxes
attributes:
label: Have you checked our README?
description: Please check the <a href="https://github.com/FlareSolverr/FlareSolverr/blob/master/README.md">README</a>.
options:
- label: I have checked the README
required: true
- type: checkboxes
attributes:
label: Is there already an issue for your problem?
description: Please make sure you are not creating an already submitted <a href="https://github.com/FlareSolverr/FlareSolverr/issues">Issue</a>. Check closed issues as well, because your issue may have already been fixed.
options:
- label: I have checked older issues, open and closed
required: true
- type: checkboxes
attributes:
label: Have you checked the discussions?
description: Please read our <a href="https://github.com/FlareSolverr/FlareSolverr/discussions">Discussions</a> before submitting your issue, some wider problems may be dealt with there.
options:
- label: I have read the Discussions
required: true
- type: textarea
attributes:
label: Environment
description: Please provide the details of the system FlareSolverr is running on.
value: |
- FlareSolverr version:
- Last working FlareSolverr version:
- Operating system:
- Are you using Docker: [yes/no]
- FlareSolverr User-Agent (see log traces or / endpoint):
- Are you using a VPN: [yes/no]
- Are you using a Proxy: [yes/no]
- Are you using Captcha Solver: [yes/no]
- If using captcha solver, which one:
- URL to test this issue:
render: markdown
validations:
required: true
- type: textarea
attributes:
label: Description
description: List steps to reproduce the error and details on what happens and what you expected to happen.
validations:
required: true
- type: textarea
attributes:
label: Logged Error Messages
description: |
Place any relevant error messages you noticed from the logs here.
Make sure you attach the full logs with your personal information removed in case we need more information.
If you wish to provide debug logs, follow the instructions from this <a href="https://github.com/FlareSolverr/FlareSolverr/wiki/How-to-enable-debug-and-html-trace">wiki page</a>.
render: text
validations:
required: true
- type: textarea
attributes:
label: Screenshots
description: Place any screenshots of the issue here if needed
validations:
required: false

8
.github/ISSUE_TEMPLATE/config.yml vendored Normal file
View File

@@ -0,0 +1,8 @@
blank_issues_enabled: false
contact_links:
- name: Requesting new features or changes
url: https://github.com/FlareSolverr/FlareSolverr/discussions
about: Please create a new discussion topic, grouped under "Ideas".
- name: Asking questions
url: https://github.com/FlareSolverr/FlareSolverr/discussions
about: Please create a new discussion topic, grouped under "Q&A".

View File

@@ -1,21 +1,20 @@
# todo: enable in the first release name: autotag
#name: autotag
# on:
#on: push:
# push: branches:
# branches: - "master"
# - "master"
# jobs:
#jobs: build:
# 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 with:
# with: GITHUB_TOKEN: "${{ secrets.GH_PAT }}"
# GITHUB_TOKEN: "${{ secrets.GH_PAT }}" tag_prefix: "v"
# tag_prefix: "v"

View File

@@ -11,43 +11,43 @@ jobs:
steps: steps:
- -
name: Checkout name: Checkout
uses: actions/checkout@v2 uses: actions/checkout@v3
- -
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: crazy-max/ghaction-docker-meta@v1 uses: crazy-max/ghaction-docker-meta@v3
with: with:
images: ${{ env.REPOSITORY }},ghcr.io/${{ env.REPOSITORY }} images: ${{ env.REPOSITORY }},ghcr.io/${{ env.REPOSITORY }}
tag-sha: false tag-sha: false
- -
name: Set up QEMU name: Set up QEMU
uses: docker/setup-qemu-action@v1.0.1 uses: docker/setup-qemu-action@v2
- -
name: Set up Docker Buildx name: Set up Docker Buildx
uses: docker/setup-buildx-action@v1 uses: docker/setup-buildx-action@v2
- -
name: Login to DockerHub name: Login to DockerHub
uses: docker/login-action@v1 uses: docker/login-action@v2
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@v1 uses: docker/login-action@v2
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@v2 uses: docker/build-push-action@v3
with: with:
context: . context: .
file: ./Dockerfile file: ./Dockerfile
platforms: linux/amd64,linux/arm/v7,linux/arm64 platforms: linux/386,linux/amd64,linux/arm/v7,linux/arm64/v8
push: ${{ github.event_name != 'pull_request' }} 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

@@ -11,12 +11,12 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v2 uses: actions/checkout@v3
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 - name: Setup Node
uses: actions/setup-node@v2 uses: actions/setup-node@v3
with: with:
node-version: '16' node-version: '16'

266
CHANGELOG.md Normal file
View File

@@ -0,0 +1,266 @@
# Changelog
## v3.0.4 (2023/03/07
* Click on the Cloudflare's 'Verify you are human' button if necessary
## v3.0.3 (2023/03/06)
* Update undetected_chromedriver version to 3.4.6
## v3.0.2 (2023/01/08)
* Detect Cloudflare blocked access
* Check Chrome / Chromium web browser is installed correctly
## v3.0.1 (2023/01/06)
* Kill Chromium processes properly to avoid defunct/zombie processes
* Update undetected-chromedriver
* Disable Zygote sandbox in Chromium browser
* Add more selectors to detect blocked access
* Include procps (ps), curl and vim packages in the Docker image
## v3.0.0 (2023/01/04)
* This is the first release of FlareSolverr v3. There are some breaking changes
* Docker images for linux/386, linux/amd64, linux/arm/v7 and linux/arm64/v8
* Replaced Firefox with Chrome
* Replaced NodeJS / Typescript with Python
* Replaced Puppeter with Selenium
* No binaries for Linux / Windows. You have to use the Docker image or install from Source code
* No proxy support
* No session support
## v2.2.10 (2022/10/22)
* Detect DDoS-Guard through title content
## v2.2.9 (2022/09/25)
* Detect Cloudflare Access Denied
* Commit the complete changelog
## v2.2.8 (2022/09/17)
* Remove 30 s delay and clean legacy code
## v2.2.7 (2022/09/12)
* Temporary fix: add 30s delay
* Update README.md
## v2.2.6 (2022/07/31)
* Fix Cloudflare detection in POST requests
## v2.2.5 (2022/07/30)
* Update GitHub actions to build executables with NodeJs 16
* Update Cloudflare selectors and add HTML samples
* Install Firefox 94 instead of the latest Nightly
* Update dependencies
* Upgrade Puppeteer (#396)
## v2.2.4 (2022/04/17)
* Detect DDoS-Guard challenge
## v2.2.3 (2022/04/16)
* Fix 2000 ms navigation timeout
* Update README.md (libseccomp2 package in Debian)
* Update README.md (clarify proxy parameter) (#307)
* Update NPM dependencies
* Disable Cloudflare ban detection
## v2.2.2 (2022/03/19)
* Fix ban detection. Resolves #330 (#336)
## v2.2.1 (2022/02/06)
* Fix max timeout error in some pages
* Avoid crashing in NodeJS 17 due to Unhandled promise rejection
* Improve proxy validation and debug traces
* Remove @types/puppeteer dependency
## v2.2.0 (2022/01/31)
* Increase default BROWSER_TIMEOUT=40000 (40 seconds)
* Fix Puppeter deprecation warnings
* Update base Docker image Alpine 3.15 / NodeJS 16
* Build precompiled binaries with NodeJS 16
* Update Puppeter and other dependencies
* Add support for Custom CloudFlare challenge
* Add support for DDoS-GUARD challenge
## v2.1.0 (2021/12/12)
* Add aarch64 to user agents to be replaced (#248)
* Fix SOCKSv4 and SOCKSv5 proxy. resolves #214 #220
* Remove redundant JSON key (postData) (#242)
* Make test URL configurable with TEST_URL env var. resolves #240
* Bypass new Cloudflare protection
* Update donation links
## v2.0.2 (2021/10/31)
* Fix SOCKS5 proxy. Resolves #214
* Replace Firefox ERS with a newer version
* Catch startup exceptions and give some advices
* Add env var BROWSER_TIMEOUT for slow systems
* Fix NPM warning in Docker images
## v2.0.1 (2021/10/24)
* Check user home dir before testing web browser installation
## v2.0.0 (2021/10/20)
FlareSolverr 2.0.0 is out with some important changes:
* It is capable of solving the automatic challenges of Cloudflare. CAPTCHAs (hCaptcha) cannot be resolved and the old solvers have been removed.
* The Chrome browser has been replaced by Firefox. This has caused some functionality to be removed. Parameters: `userAgent`, `headers`, `rawHtml` and `downloadare` no longer available.
* Included `proxy` support without user/password credentials. If you are writing your own integration with FlareSolverr, make sure your client uses the same User-Agent header and Proxy that FlareSolverr uses. Those values together with the Cookie are checked and detected by Cloudflare.
* FlareSolverr has been rewritten from scratch. From now on it should be easier to maintain and test.
* If you are using Jackett make sure you have version v0.18.1041 or higher. FlareSolverSharp v2.0.0 is out too.
Complete changelog:
* Bump version 2.0.0
* Set puppeteer timeout half of maxTimeout param. Resolves #180
* Add test for blocked IP
* Avoid reloading the page in case of error
* Improve Cloudflare detection
* Fix version
* Fix browser preferences and proxy
* Fix request.post method and clean error traces
* Use Firefox ESR for Docker images
* Improve Firefox start time and code clean up
* Improve bad request management and tests
* Build native packages with Firefox
* Update readme
* Improve Docker image and clean TODOs
* Add proxy support
* Implement request.post method for Firefox
* Code clean up, remove returnRawHtml, download, headers params
* Remove outdated chaptcha solvers
* Refactor the app to use Express server and Jest for tests
* Fix Cloudflare resolver for Linux ARM builds
* Fix Cloudflare resolver
* Replace Chrome web browser with Firefox
* Remove userAgent parameter since any modification is detected by CF
* Update dependencies
* Remove Puppeter steath plugin
## v1.2.9 (2021/08/01)
* Improve "Execution context was destroyed" error handling
* Implement returnRawHtml parameter. resolves #172 resolves #165
* Capture Docker stop signal. resolves #158
* Reduce Docker image size 20 MB
* Fix page reload after challenge is solved. resolves #162 resolves #143
* Avoid loading images/css/fonts to speed up page load
* Improve Cloudflare IP ban detection
* Fix vulnerabilities
## v1.2.8 (2021/06/01)
* Improve old JS challenge waiting. Resolves #129
## v1.2.7 (2021/06/01)
* Improvements in Cloudflare redirect detection. Resolves #140
* Fix installation instructions
## v1.2.6 (2021/05/30)
* Handle new Cloudflare challenge. Resolves #135 Resolves #134
* Provide reference Systemd unit file. Resolves #72
* Fix EACCES: permission denied, open '/tmp/flaresolverr.txt'. Resolves #120
* Configure timezone with TZ env var. Resolves #109
* Return the redirected URL in the response (#126)
* Show an error in hcaptcha-solver. Resolves #132
* Regenerate package-lock.json lockfileVersion 2
* Update issue template. Resolves #130
* Bump ws from 7.4.1 to 7.4.6 (#137)
* Bump hosted-git-info from 2.8.8 to 2.8.9 (#124)
* Bump lodash from 4.17.20 to 4.17.21 (#125)
## v1.2.5 (2021/04/05)
* Fix memory regression, close test browser
* Fix release-docker GitHub action
## v1.2.4 (2021/04/04)
* Include license in release zips. resolves #75
* Validate Chrome is working at startup
* Speedup Docker image build
* Add health check endpoint
* Update issue template
* Minor improvements in debug traces
* Validate environment variables at startup. resolves #101
* Add FlareSolverr logo. resolves #23
## v1.2.3 (2021/01/10)
* CI/CD: Generate release changelog from commits. resolves #34
* Update README.md
* Add donation links
* Simplify docker-compose.yml
* Allow to configure "none" captcha resolver
* Override docker-compose.yml variables via .env resolves #64 (#66)
## v1.2.2 (2021/01/09)
* Add documentation for precompiled binaries installation
* Add instructions to set environment variables in Windows
* Build Windows and Linux binaries. resolves #18
* Add release badge in the readme
* CI/CD: Generate release changelog from commits. resolves #34
* Add a notice about captcha solvers
* Add Chrome flag --disable-dev-shm-usage to fix crashes. resolves #45
* Fix Docker CLI documentation
* Add traces with captcha solver service. resolves #39
* Improve logic to detect Cloudflare captcha. resolves #48
* Move Cloudflare provider logic to his own class
* Simplify and document the "return only cookies" parameter
* Show message when debug log is enabled
* Update readme to add more clarifications. resolves #53 (#60)
* issue_template: typo fix (#52)
## v1.2.1 (2020/12/20)
* Change version to match release tag / 1.2.0 => v1.2.0
* CI/CD Publish release in GitHub repository. resolves #34
* Add welcome message in / endpoint
* Rewrite request timeout handling (maxTimeout) resolves #42
* Add http status for better logging
* Return an error when no selectors are found, #25
* Add issue template, fix #32
* Moving log.html right after loading the page and add one on reload, fix #30
* Update User-Agent to match chromium version, ref: #15 (#28)
* Update install from source code documentation
* Update readme to add Docker instructions (#20)
* Clean up readme (#19)
* Add docker-compose
* Change default log level to info
## v1.2.0 (2020/12/20)
* Fix User-Agent detected by CouldFlare (Docker ARM) resolves #15
* Include exception message in error response
* CI/CD: Rename GitHub Action build => publish
* Bump version
* Fix TypeScript compilation and bump minor version
* CI/CD: Bump minor version
* CI/CD: Configure GitHub Actions
* CI/CD: Configure GitHub Actions
* CI/CD: Bump minor version
* CI/CD: Configure Build GitHub Action
* CI/CD: Configure AutoTag GitHub Action (#14)
* CI/CD: Build the Docker images with GitHub Actions (#13)
* Update dependencies
* Backport changes from Cloudproxy (#11)

View File

@@ -1,4 +1,4 @@
FROM python:3.10-slim-bullseye 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 \
@@ -12,28 +12,25 @@ 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:3.10-slim-bullseye FROM python:3.11-slim-bullseye
# Copy dummy packages # Copy dummy packages
COPY --from=builder /*.deb / COPY --from=builder /*.deb /
# Install dependencies and create flaresolverr user # Install dependencies and create flaresolverr user
# We have to install and old version of Chromium because its not working in Raspberry Pi / ARM
# You can test Chromium running this command inside the container: # You can test Chromium running this command inside the container:
# xvfb-run -s "-screen 0 1600x1200x24" chromium --no-sandbox # xvfb-run -s "-screen 0 1600x1200x24" chromium --no-sandbox
# The error traces is like this: "*** stack smashing detected ***: terminated" # The error traces is like this: "*** stack smashing detected ***: terminated"
# To check the package versions available you can use this command: # To check the package versions available you can use this command:
# apt-cache madison chromium # apt-cache madison chromium
WORKDIR /app WORKDIR /app
RUN echo "\ndeb http://snapshot.debian.org/archive/debian/20210519T212015Z/ bullseye main" >> /etc/apt/sources.list \
&& echo 'Acquire::Check-Valid-Until "false";' | tee /etc/apt/apt.conf.d/00snapshot \
# Install dummy packages # Install dummy packages
&& dpkg -i /libgl1-mesa-dri.deb \ 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=89.0.4389.114-1 chromium-common=89.0.4389.114-1 \ && apt-get install -y --no-install-recommends chromium chromium-common chromium-driver xvfb dumb-init \
chromium-driver=89.0.4389.114-1 xvfb \ procps curl vim \
# 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* \
@@ -47,8 +44,7 @@ RUN echo "\ndeb http://snapshot.debian.org/archive/debian/20210519T212015Z/ bull
COPY requirements.txt . COPY requirements.txt .
RUN pip install -r requirements.txt \ RUN pip install -r requirements.txt \
# Remove temporary files # Remove temporary files
&& rm -rf /root/.cache \ && rm -rf /root/.cache
&& find / -name '*.pyc' -delete
USER flaresolverr USER flaresolverr
@@ -57,13 +53,17 @@ COPY package.json ../
EXPOSE 8191 EXPOSE 8191
# 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:3.0.0.beta2 . # docker build -t ngosang/flaresolverr:3.0.0 .
# docker run -p 8191:8191 ngosang/flaresolverr:3.0.0.beta2 # 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 buildx create --use # docker buildx create --use
# docker buildx build -t ngosang/flaresolverr:3.0.0.beta2 --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

View File

@@ -1,6 +1,6 @@
MIT License MIT License
Copyright (c) 2022 Diego Heras (ngosang / ngosang@hotmail.es) Copyright (c) 2023 Diego Heras (ngosang / ngosang@hotmail.es)
Permission is hereby granted, free of charge, to any person obtaining a copy Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal of this software and associated documentation files (the "Software"), to deal

View File

@@ -64,14 +64,11 @@ Remember to restart the Docker daemon and the container after the update.
### Precompiled binaries ### Precompiled binaries
This is the recommended way for Windows users. Precompiled binaries are not currently available for v3. Please see https://github.com/FlareSolverr/FlareSolverr/issues/660 for updates,
* Download the [FlareSolverr zip](https://github.com/FlareSolverr/FlareSolverr/releases) from the release's assets. It is available for Windows and Linux. or below for instructions of how to build FlareSolverr from source code.
* Extract the zip file. FlareSolverr executable and firefox folder must be in the same directory.
* Execute FlareSolverr binary. In the environment variables section you can find how to change the configuration.
### From source code ### From source code
This is the recommended way for macOS users and for developers.
* Install [Python 3.10](https://www.python.org/downloads/). * 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 [Chrome](https://www.google.com/intl/en_us/chrome/) or [Chromium](https://www.chromium.org/getting-involved/download-chromium/) web browser.
* (Only in Linux / macOS) Install [Xvfb](https://en.wikipedia.org/wiki/Xvfb) package. * (Only in Linux / macOS) Install [Xvfb](https://en.wikipedia.org/wiki/Xvfb) package.

View File

@@ -1,7 +1,7 @@
{ {
"name": "flaresolverr", "name": "flaresolverr",
"version": "3.0.0.beta2", "version": "3.0.4",
"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

@@ -1,9 +1,9 @@
bottle==0.12.23 bottle==0.12.23
waitress==2.1.2 waitress==2.1.2
selenium==4.4.3 selenium==4.7.2
func-timeout==4.3.5 func-timeout==4.3.5
# required by undetected_chromedriver # required by undetected_chromedriver
requests==2.28.1 requests==2.28.1
websockets==10.3 websockets==10.4
# only required for linux # only required for linux
xvfbwrapper==0.2.9 xvfbwrapper==0.2.9

View File

@@ -1,4 +1,5 @@
import logging import logging
import sys
import time import time
from urllib.parse import unquote from urllib.parse import unquote
@@ -6,30 +7,58 @@ 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.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 from selenium.webdriver.support.expected_conditions import presence_of_element_located, staleness_of, title_is
from dtos import V1RequestBase, V1ResponseBase, ChallengeResolutionT, ChallengeResolutionResultT, IndexResponse, \ from dtos import V1RequestBase, V1ResponseBase, ChallengeResolutionT, ChallengeResolutionResultT, IndexResponse, \
HealthResponse, STATUS_OK, STATUS_ERROR HealthResponse, STATUS_OK, STATUS_ERROR
import utils import utils
ACCESS_DENIED_TITLES = [
# Cloudflare
'Access denied',
# Cloudflare http://bitturk.net/ Firefox
'Attention Required! | Cloudflare'
]
ACCESS_DENIED_SELECTORS = [ ACCESS_DENIED_SELECTORS = [
# Cloudflare # Cloudflare
'div.main-wrapper div.header.section h1 span.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 = [
# Cloudflare
'Just a moment...',
# DDoS-GUARD
'DDOS-GUARD',
] ]
CHALLENGE_SELECTORS = [ CHALLENGE_SELECTORS = [
# Cloudflare # Cloudflare
'#cf-challenge-running', '.ray_id', '.attack-box', '#cf-please-wait', '#trk_jschal_js', '#cf-challenge-running', '.ray_id', '.attack-box', '#cf-please-wait', '#challenge-spinner', '#trk_jschal_js',
# DDoS-GUARD
'#link-ddg',
# 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'
] ]
SHORT_TIMEOUT = 5 SHORT_TIMEOUT = 10
def test_browser_installation(): def test_browser_installation():
logging.info("Testing web browser installation...") logging.info("Testing web browser installation...")
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)
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")
@@ -152,6 +181,45 @@ def _resolve_challenge(req: V1RequestBase, method: str) -> ChallengeResolutionT:
driver.quit() driver.quit()
def click_verify(driver: WebDriver):
try:
logging.debug("Try to find the Cloudflare verify checkbox")
iframe = driver.find_element(By.XPATH, "//iframe[@title='Widget containing a Cloudflare security challenge']")
driver.switch_to.frame(iframe)
checkbox = driver.find_element(
by=By.XPATH,
value='//*[@id="cf-stage"]//label[@class="ctp-checkbox-label"]/input',
)
if checkbox:
actions = ActionChains(driver)
actions.move_to_element_with_offset(checkbox, 5, 7)
actions.click(checkbox)
actions.perform()
logging.debug("Cloudflare verify checkbox found and clicked")
except Exception as e:
logging.debug("Cloudflare verify checkbox not found on the page")
# print(e)
finally:
driver.switch_to.default_content()
try:
logging.debug("Try to find the Cloudflare 'Verify you are human' button")
button = driver.find_element(
by=By.XPATH,
value="//input[@type='button' and @value='Verify you are human']",
)
if button:
actions = ActionChains(driver)
actions.move_to_element_with_offset(button, 5, 7)
actions.click(button)
actions.perform()
logging.debug("The Cloudflare 'Verify you are human' button found and clicked")
except Exception as e:
logging.debug("The Cloudflare 'Verify you are human' button not found on the page")
# print(e)
time.sleep(2)
def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> ChallengeResolutionT: def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> ChallengeResolutionT:
res = ChallengeResolutionT({}) res = ChallengeResolutionT({})
res.status = STATUS_OK res.status = STATUS_OK
@@ -168,7 +236,13 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
# wait for the page # wait for the page
html_element = driver.find_element(By.TAG_NAME, "html") html_element = driver.find_element(By.TAG_NAME, "html")
page_title = driver.title
# find access denied titles
for title in ACCESS_DENIED_TITLES:
if title == page_title:
raise Exception('Cloudflare has blocked this request. '
'Probably your IP is banned for this site, check in your web browser.')
# find access denied selectors # 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)
@@ -176,21 +250,35 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
raise Exception('Cloudflare has blocked this request. ' raise Exception('Cloudflare has blocked this request. '
'Probably your IP is banned for this site, check in your web browser.') 'Probably your IP is banned for this site, check in your web browser.')
# find challenge selectors # find challenge by title
challenge_found = False challenge_found = False
for selector in CHALLENGE_SELECTORS: for title in CHALLENGE_TITLES:
found_elements = driver.find_elements(By.CSS_SELECTOR, selector) if title == page_title:
if len(found_elements) > 0:
challenge_found = True challenge_found = True
logging.info("Challenge detected. Selector found: " + selector) logging.info("Challenge detected. Title found: " + title)
break break
if not challenge_found:
# find challenge by selectors
for selector in CHALLENGE_SELECTORS:
found_elements = driver.find_elements(By.CSS_SELECTOR, selector)
if len(found_elements) > 0:
challenge_found = True
logging.info("Challenge detected. Selector found: " + selector)
break
attempt = 0
if challenge_found: if challenge_found:
while True: while True:
try: try:
attempt = attempt + 1
# wait until the title changes
for title in CHALLENGE_TITLES:
logging.debug("Waiting for title (attempt " + str(attempt) + "): " + title)
WebDriverWait(driver, SHORT_TIMEOUT).until_not(title_is(title))
# then wait until all the selectors disappear # then wait until all the selectors disappear
for selector in CHALLENGE_SELECTORS: for selector in CHALLENGE_SELECTORS:
logging.debug("Waiting for selector: " + selector) logging.debug("Waiting for selector (attempt " + str(attempt) + "): " + selector)
WebDriverWait(driver, SHORT_TIMEOUT).until_not( WebDriverWait(driver, SHORT_TIMEOUT).until_not(
presence_of_element_located((By.CSS_SELECTOR, selector))) presence_of_element_located((By.CSS_SELECTOR, selector)))
@@ -199,6 +287,9 @@ 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")
@@ -220,11 +311,11 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
challenge_res.url = driver.current_url challenge_res.url = driver.current_url
challenge_res.status = 200 # todo: fix, selenium not provides this info challenge_res.status = 200 # todo: fix, selenium not provides this info
challenge_res.cookies = driver.get_cookies() challenge_res.cookies = driver.get_cookies()
challenge_res.userAgent = utils.get_user_agent(driver)
if not req.returnOnlyCookies: if not req.returnOnlyCookies:
challenge_res.headers = {} # todo: fix, selenium not provides this info challenge_res.headers = {} # todo: fix, selenium not provides this info
challenge_res.response = driver.page_source challenge_res.response = driver.page_source
challenge_res.userAgent = utils.get_user_agent(driver)
res.result = challenge_res res.result = challenge_res
return res return res

View File

@@ -1,5 +1,4 @@
import unittest import unittest
from datetime import datetime, timezone
from webtest import TestApp from webtest import TestApp
@@ -20,12 +19,12 @@ class TestFlareSolverr(unittest.TestCase):
proxy_url = "http://127.0.0.1:8888" proxy_url = "http://127.0.0.1:8888"
proxy_socks_url = "socks5://127.0.0.1:1080" proxy_socks_url = "socks5://127.0.0.1:1080"
google_url = "https://www.google.com" google_url = "https://www.google.com"
post_url = "https://ptsv2.com/t/qv4j3-1634496523" post_url = "https://httpbin.org/post"
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/"
custom_cloudflare_url = "https://www.muziekfabriek.org" custom_cloudflare_url = "https://www.muziekfabriek.org"
cloudflare_blocked_url = "https://avistaz.to/api/v1/jackett/torrents?in=1&type=0&search=" cloudflare_blocked_url = "https://cpasbiens3.fr/index.php?do=search&subaction=search"
app = TestApp(flaresolverr.app) app = TestApp(flaresolverr.app)
@@ -233,7 +232,7 @@ class TestFlareSolverr(unittest.TestCase):
self.assertIsNone(solution.headers) self.assertIsNone(solution.headers)
self.assertIsNone(solution.response) self.assertIsNone(solution.response)
self.assertGreater(len(solution.cookies), 0) self.assertGreater(len(solution.cookies), 0)
self.assertIsNone(solution.userAgent) self.assertIn("Chrome/", solution.userAgent)
# todo: test Cmd 'request.get' should return OK with HTTP 'proxy' param # 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 # todo: test Cmd 'request.get' should return OK with HTTP 'proxy' param with credentials
@@ -281,7 +280,7 @@ class TestFlareSolverr(unittest.TestCase):
def test_v1_endpoint_request_post_no_cloudflare(self): def test_v1_endpoint_request_post_no_cloudflare(self):
res = self.app.post_json('/v1', { res = self.app.post_json('/v1', {
"cmd": "request.post", "cmd": "request.post",
"url": self.post_url + '/post', "url": self.post_url,
"postData": "param1=value1&param2=value2" "postData": "param1=value1&param2=value2"
}) })
self.assertEqual(res.status_code, 200) self.assertEqual(res.status_code, 200)
@@ -297,22 +296,10 @@ class TestFlareSolverr(unittest.TestCase):
self.assertIn(self.post_url, solution.url) self.assertIn(self.post_url, solution.url)
self.assertEqual(solution.status, 200) self.assertEqual(solution.status, 200)
self.assertIs(len(solution.headers), 0) self.assertIs(len(solution.headers), 0)
self.assertIn("I hope you have a lovely day!", solution.response) self.assertIn('"form": {\n "param1": "value1", \n "param2": "value2"\n }', solution.response)
self.assertEqual(len(solution.cookies), 0) self.assertEqual(len(solution.cookies), 0)
self.assertIn("Chrome/", solution.userAgent) self.assertIn("Chrome/", solution.userAgent)
# check that we sent the post data
res2 = self.app.post_json('/v1', {
"cmd": "request.get",
"url": self.post_url
})
self.assertEqual(res2.status_code, 200)
body2 = V1ResponseBase(res2.json)
self.assertEqual(STATUS_OK, body2.status)
date_hour = datetime.now(timezone.utc).isoformat().split(':')[0].replace('T', ' ')
self.assertIn(date_hour, body2.solution.response)
def test_v1_endpoint_request_post_cloudflare(self): def test_v1_endpoint_request_post_cloudflare(self):
res = self.app.post_json('/v1', { res = self.app.post_json('/v1', {
"cmd": "request.post", "cmd": "request.post",

File diff suppressed because it is too large Load Diff

View File

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

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

@@ -44,6 +44,8 @@ def get_webdriver() -> WebDriver:
# 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')
# note: headless mode is detected (options.headless = True) # note: headless mode is detected (options.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
@@ -86,6 +88,10 @@ def get_webdriver() -> WebDriver:
return driver return driver
def get_chrome_exe_path() -> str:
return uc.find_chrome_executable()
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:
@@ -110,7 +116,6 @@ 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