5684e896af
Fixes #2259 This PR brings backend and frontend support for the new autoclick behavior in Browsertrix, introduces in Browsertrix 1.5.0+ On the backend, we introduce `min_autoclick_crawler_image` to `values.yaml`, with a default value of `"docker.io/webrecorder/browsertrix-crawler:1.5.0"`. If this is set and the crawler version for a new crawl is less than this value, the autoclick behavior is removed from the behaviors list in the configmap created for the crawl. The one caveat for this is that a crawler image tag like "latest" will always be parsed as greater than `min_autoclick_crawler_image`, so there is the potential for the crawler to run into issues if using a non-numeric image tag with an older version of the crawler. For production we use hardcoded specific versions of the crawler except for the dev channel, which from here on out will including autoclick support, so I think this should be okay (and is also true of the existing implementation for checking `min_qa_crawler_image`). On the frontend, I've added a checkbox (unchecked by default) in the "Limits" section just below the current checkbox for autoscroll. We might want to move these to a different section eventually - I'm not sure Limits is the right place for them - but I wanted to be consistent with things as they are. --------- Co-authored-by: Ilya Kreymer <ikreymer@users.noreply.github.com> |
||
|---|---|---|
| .github | ||
| .vscode | ||
| ansible | ||
| assets | ||
| backend | ||
| chart | ||
| configs | ||
| frontend | ||
| scripts | ||
| test | ||
| .gitattributes | ||
| .gitignore | ||
| .pre-commit-config.yaml | ||
| btrix | ||
| CHANGES.md | ||
| LICENSE | ||
| NOTICE | ||
| pylintrc | ||
| README.md | ||
| update-version.sh | ||
| version.txt | ||
Browsertrix is a cloud-native, high-fidelity, browser-based crawling service designed to make web archiving easier and more accessible for everyone.
The service provides an API and UI for scheduling crawls and viewing results, and managing all aspects of crawling process. This system provides the orchestration and management around crawling, while the actual crawling is performed using Browsertrix Crawler containers, which are launched for each crawl.
See webrecorder.net/browsertrix for a feature overview and information about how to sign up for Webrecorder's hosted Browsertrix service.
Documentation
The full docs for using, deploying, and developing Browsertrix are available at docs.browsertrix.com.
Our docs are created with Material for MKDocs.
Deployment
The latest deployment documentation is available at docs.browsertrix.com/deploy.
The docs cover deploying Browsertrix in different environments using Kubernetes, from a single-node setup to scalable clusters in the cloud.
Early on, Browsertrix also supported Docker Compose and podman-based deployment. This was deprecated due to the complexity of maintaining feature parity across different setups, and with various Kubernetes deployment options being available and easy to deploy, even on a single machine.
Making deployment of Browsertrix as easy as possible remains a key goal, and we welcome suggestions for how we can further improve our Kubernetes deployment options.
If you are looking to just try running a single crawl, you may want to try Browsertrix Crawler first to test out the crawling capabilities.
Contributing
Though the system and backend API is fairly stable, we are working on many additional features. Please see the GitHub issues and this GitHub Project for our current project plan and tasks.
Guides for getting started with local development are available at docs.browsertrix.com/develop.
Translation
We use Weblate to manage translation contributions.
License
Browsertrix is made available under the AGPLv3 License.
Documentation is made available under the Creative Commons Attribution 4.0 International License