README.md

well-known-crawler

Crawl well-known Resources introduced by The Privacy Sandbox:

• Related Website Sets

  • /.well-known/related-website-set.json
  • HTTPS only
  • on ETLD+1 only, where PSL is authoritative source for ETLD
  • different JSON format depending if primary or other
  • There is a canonical public list of sets, but some (like google.com/youtube.com) are missing from it for instance
  • Generator

• Attestation File

  • /.well-known/privacy-sandbox-attestations.json
  • Submit a form, JSON file sent by Google
  • No public list of who participates

Dependencies

A Dockerfile is provided under .devcontainer/; for direct integration with VS Code or to manually build the image and deploy the Docker container, follow the instructions in this guide.

Environment Variables

Required:

• CRUX_URL: The URL to the cached version of CrUX to use (https://github.com/zakird/crux-top-lists/raw/main/data/global/current.csv.gz)
• CRUX_TOP: Specify how many first top origins to crawl (1000000)
• RWS_URL: The URL to the RWS canonical set (https://raw.githubusercontent.com/GoogleChrome/related-website-sets/main/related_website_sets.JSON)

Optional:

• S3_DATA_BUCKET: The s3 bucket where the crawl raw results are saved, if undefined, we are assuming local run.

Usage

./crawl_crux.sh

Gitlab CI/CD Variables

Define the following CI variables to have Gitlab CI building and pushing the Docker image automatically so that ECS task is up to date:

• AWS_ACCOUNT_ID: the AWS account ID
• AWS_REGION: the AWS region to use
• AWS_ACCESS_KEY_ID: of an IAM user with the AmazonEC2ContainerRegistryPowerUser policy
• AWS_SECRET_ACCESS_KEY: of an IAM user with the AmazonEC2ContainerRegistryPowerUser policy