Skip to main content

v0.4.0

· 2 min read

Hello,

We are pleased to announce the release of version 0.4.0, which introduces contaminated data cleanup, user aliasing, and a simplified Docker deployment model.

New Features​

  • Contaminated data cleanup: Hyppot now detects when a single person was exposed to more than one variant of the same experiment — either under the same user ID or across linked identities — marks those impressions as contaminated, and excludes them from conversion and metric statistics. Detection runs at alias time, via a background job (every 10 minutes), and at statistics time. The experiment Statistics tab now reports Contaminated users excluded: N. See Contaminated data cleanup for details.

  • User aliasing: You can now link multiple user IDs as belonging to the same person (for example, anonymous and logged-in IDs at login). New REST API endpoints: POST {pathPrefix}/userAlias and GET {pathPrefix}/userAlias/{userId}. Available in-process via IUserAliasService and through the Hyppot.Sdk package. Aliases are retained for a configurable period (UserAliasRetentionPeriod, default 30 days) and cleaned up automatically. User aliasing must be performed from your backend — the browser SDK does not expose it.

Breaking Changes​

  • Docker image is now PostgreSQL-only: The Docker version no longer supports selecting different database providers. The Docker image now works exclusively with PostgreSQL using the DB_CONNECTION_STRING environment variable. If you were previously using SQLite or SQL Server with the Docker image, you must migrate to PostgreSQL. Contact us for assistance to ensure a seamless migration. SQLite and SQL Server remain supported via NuGet integration.