Skip to content

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

v1.10.5 - 2025-09-05

Changes

  • Added a free-text field to optionally add data description comments to each column in #564.
  • Updated the version of all Java-based services from 17 (EOL) to 21 and updated the version of all Node-based services from 20 (EOL) to 24 in #562.

v1.10.4 - 2025-08-04

Changes

  • Added the language next to descriptions and titles in #559.
  • Added the verbs ListRecords and ListSets to comply with OpenAIRE Provide and make the records findable using the OAI-PMH harvesting protocol.

v1.10.3 - 2025-07-28

Fixes

  • Fixed a wrong configuration of the Auth Service in the Helm Chart, where new pods reset the realm settings, leading to a deletion of e.g. the SSO settings in #550.

v1.10.2 - 2025-07-12

Fixes

  • Fixed a bug where saving a PID multiple times created duplicate titles, creators, descriptions, funders, etc. in #545.

v1.10.1 - 2025-07-10

Fixes

  • Fixed a bug where PID presentation in the UI did not select the newest PID.
  • Fixed a bug where a user could get a different id from the Identity Service when using a different authentication provider (i.e. SAML)

Changes

  • Changed the mechanism to obtain the subset columns to use DuckDB and now create views anymore.
  • Changed the internal user identifier from using the user id (i.e. java.util.UUID) to the username (i.e. java.lang.String) for all interfaces including the query store due to continued instability from SSO-integrations relying on the EntryUUID in the Identity Service.

v1.10.0 - 2025-07-05

Removals

  • Removed the Analyse Service in favor of DuckDB in the Data Service covering the same functionality.

Fixes

  • Fixed a bug where saving a PID multiple times broke the entity linking within Hibernate in #545.

v1.9.3 - 2025-06-06

Features

  • Added a Jupyter Starter notebook for data scientists that is pre-configured to use DBRepo in #529.

Changes

  • Gave the documentation website a major overhaul with detailed user guide, maintainer guide, animated step-by-step manuals and flowcharts for better overview in #514.

Fixes

  • Fixed a bug where identifiers could not be created because of missing ordinal_position field for creators.
  • Fixed multiple bugs where access grants were mapped wrongly.
  • Fixed a bug where subsets could not be created because the wrong data source (table) was used in the data service.
  • Fixed a bug in the UI where grants were not displayed once created/update/revoked.

v1.9.2 - 2025-06-05

Features

  • Added an endpoint to check for database grants and display them in the UI (Database > Settings) for the owner in #536.

v1.9.1 - 2025-05-31

Changes

  • Release a new patch since version 1.9.0 was accidentally released on PyPI.

v1.9.0 - 2025-05-30

Fixes

  • Fixed a bug where another PID was registered when using the DOI profile in the Metadata Service.
  • Fixed a bug where titles, descriptions, creators, etc. were not sorted to the user-specified ordering in #531.
  • Fixed a design issue where the get_identifier_data method in the Python library only fetched the first 10.000 rows and could not paginate in #527.
  • Fixed a bug where Spark did not map the column headers correct when importing a dataset in #518.

Changes

  • Improved S3-related mechanisms to de-duplicate uploaded datasets and remove them on successful import, various structured logging improvements in #528.

v1.8.2 - 2025-05-15

Fixes

  • Fixed a bug in the UI where the resource status was displayed as identifier when a draft identifier has been created.

Features

  • Added structured logging through the fluentd protocol via the lightweight fluentbit in a separate Logging Service. in #524.
  • Added a separate database for the Dashboard Service for high-availability deployment of Grafana in #526.

Changes

  • Improved internal packaging mechanism that is compatible with multiarch deployments in #523.

v1.8.1 - 2025-04-13

Changes

  • Improved default mask for PIDs in #521.
  • Added a visual filter for displaying starred/unstarred/all subsets in the UI.
  • Specified image platform as linux/amd64 in docker-compose.yaml deployment to enable host platform (e.g. ARM) to emulate it.
  • Specified resource limits in the docker-compose.yaml deployment in #517.

Fixes

  • Fixed a bug in the UI that displays the "Create View" button only when the user has at least read access.

Removals

  • Removed the stale objects scheduler from the Data Service and pushed it to next release.

v1.8.0 - 2025-04-04

Features

  • Refactored internal Java-based testing data that improves test consistency in #510.
  • Added automated dashboard generation for all public databases where each view has an overview of its data and schema in #460.

Changes

  • Removed OpenSearch security plugin from the Docker test deployment and changed to the bitnami/opensearch image of the same version in #515.

Fixes

  • Fixed a bug where validation of missing Principal object in Java services caused a 400 error instead of a 401 error in #512.

v1.7.3 - 2025-03-17

Fixes

  • Fixed a wrong configuration where assets were not considered in the Kubernetes deployment.
  • Fixed a wrong configuration in the Docker deployment where the OIDC provider did not consider other URLs than http://localhost.

v1.7.2 - 2025-03-13

Fixes

  • Fixed a wrong configuration of caffeine in the Data Service that did not find views/subsets after table creation within the cache period of 60 seconds in #506.

v1.7.1 - 2025-03-06

Features

  • Added support to download pandas DataFrame by PID in #503.
  • Added the possibility to create and fill a table from a pandas DataFrame (or optionally just create the schema) in #496.

Fixes

  • Fixed a bug where quick interaction with the UI caused the user to trigger the brute-force login detection in #501.

v1.7.0 - 2025-03-03

Contains Breaking Changes

This release updates the Metadata Database schema which is incompatible to v1.6.3! Follow the steps:

  1. Make a backup of the database with mariadb-dump.
  2. Apply the schema changes script: schema.sql: 
    mariadb -h 127.0.0.1 -p3306 -u root --password=<password> -D dbrepo < schema.sql
  3. Install the dependencies from the requirements.txt file or use your local environment: 
    pip install dbrepo==1.6.5rc15
  4. Run the data migration script data.py: 
    python data.py > data.sql
    It generates the SQL statements used for migrating to the new schema.
  5. Run the generated data.sql script: 
    mariadb -h 127.0.0.1 -p3306 -u root --password=<password> -D dbrepo < data.sql

Features

  • Implemented a basic brute-force security defense strategy in the Auth Service that increments the wait time on wrong logins in #494.
  • Implemented a password policy in #495.

Changes

  • Replaced sequential numerical ids with non-guessable random ids in the Metadata Database in #491.
  • Changed the interface for executing query in subsets/views in #493.

Removals

  • Removed the Upload Service in favor of an internal stable upload endpoint in the Data Service in #492.

v1.6.5 - 2025-02-18

Fixes

  • Fixed a bug where listing the views in the Python library did not work.
  • Fixed a wrong MariaDB configuration where the innodb_buffer_pool_size variable was not configured to 70% of the available memory in the Helm chart.

v1.6.4 - 2025-02-14

Fixes

  • Fixed a bug where the users were not synced with the Metadata Database in #489.

v1.6.3 - 2025-02-05

Changes

  • Refactored the UI to support OIDC and added an event listener to the Auth Service that syncs users on creation to the Metadata DB in #488.

v1.6.2 - 2025-01-24

Changes

  • Added interface tests for the Python library in Gitlab CI/CD pipeline in #486.

Fixes

  • Fixed a bug where no pagination was possible in #487.

v1.6.1 - 2025-01-21

Changes

  • Added privacy feature for hidden databases (and optionally tables, views, subsets) that hides them completely from e.g. the search in #482.

Fixes

  • Added init container that adds the admin user to the Metadata Database in #480.

v1.6.0 - 2025-01-07

Features

  • Added possibility to modify table description and privacy mode that hides metadata of databases, tables, subsets and views in #472.
  • Added Auth Service init container to sync internal system ("admin") user into Metadata Database in #470

Changes

  • Drop MariaDB Galera chart version from 14.0.12 to 13.2.7 because of stability issues with the newer versions on the OpenShift Kubernetes cluster.
  • Bumped Grafana chart version from 10.1.1 to 11.4.2.
  • Bumped NGINX chart version from 18.2.6 to 18.3.1.
  • Bumped SeaweedFS chart version from 1.0.2 to 4.2.1 (created a pull request #31192).
  • Improve RESTfulness of HTTP API to e.g. not include the database image due to size in #463.

Fixes

  • Fixed a wrong JPA entity configuration in Hibernate that in some cases causes the SERIAL increment to not trigger in MariaDB Galera. Removed the deprecated @GenericGenerator annotation on JPA entities.
  • Fixed a bug where the dataset separator was being ignored for imports in #478.

v1.5.3 - 2024-12-13

Fixes

  • Fixed a bug where subsets containing sub-queries are not able to retrieve data in #476.

v1.5.2 - 2024-12-03

Changes

  • Adapt Helm chart to support runAsNonRoot throughout and specify resource presets for the highly-constrained OpenShift Kubernetes environment in #467.
  • Require authentication for uploading files in #466.

Fixes

  • Fixed a validation problem failing to validate UUIDs in #471.
  • Fixed the dist.tar.gz file not being found in the CI/CD pipeline on release- branches in #465.

v1.5.1 - 2024-11-09

Fixes

  • Bug where the data volume could not be calculated when the data length column in the Metadata Database is null in #462.
  • Bug where the schema could not be created manually in #461.

v1.5.0 - 2024-11-06

Contains Breaking Changes

This release updates the Metadata Database schema which is incompatible to v1.4.6! Use the migration script schema_1.4.5-to-1.5.0.sql to apply the changes manually.

Features

  • Added SERIAL data type to create incrementing key in #454

Changes

  • Remove the Data Database Sidecar and replace it with Apache Spark in #458.
  • Allow anonymous users to create subsets for public databases in #449.
  • Show file upload progress in #448.
  • Change the Docker image of the Auth Service to Bitnami-maintained similar to Kubernetes deployment with accompanying Auth Database change to PostgreSQL in #455

Fixes

  • Preventing the semicolon ; to be used in UI and fixed cryptic subset execution error messages in #456.
  • Multiple UI errors in #453.
  • Fixed install script.sh in #444
  • No hardcoded data type metadata in UI but instead added it hardcoded (associated with image_id) Metadata Database.

v1.4.6 - 2024-10-11

Contains Breaking Changes

This release updates the Metadata Database schema which is incompatible to v1.4.5!

Features

Changes

  • Show the progress of dataset uploads in the UI in #448
  • Anonymous users are allowed to create (non-persistent) subsets in #449
  • Removed logic that maps True, False and null

Fixes

  • Import of datasets stabilized in the UI in #442
  • Install script in #444