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.
- Configuration parameter to disable elastic indexing
- Mimimal configuration file for an RE only indexer
- Configuration parameter for max objects in a workspace to skip narrative reindex
- Configuration option to skip narrative reindex for certain workspaces
- Configuration option for poll timeout
- Configuration option to skip narrative reindexing
- Fixing Samples Releng indexer to work with RE api time-travel.
- Added prefix to default_search alias
- Removed reads_1 from reads alias
- Removed pangenome_orthologfamily from default_search
- adding narrative_config.yaml spec file for separating narrative indexing.
- Configuration update for sample releng indexer.
- Handle SET_PERMISSION events and update the "shared_users" field
- Don't throw a RuntimeError on an unhandled event; only log a warning
- AMA index runner configuration bug fix
- bug fixes for sampleset relation engine indexer
- Bugfixes related to setting default field values for new ES docs.
- Sample relation engine indexer. indexes based on SampleSet.
- Centralize the configuration for Elasticsearch indexer modules into
spec/elasticsearch_modules.yaml
- Fix the permissions errors associated with copying a SampleSet object in a narrative. Updates how default fields are merged on index
- Save the index runner app version in every Elasticsearch document we save
- Fix the signal handlers in the main index runner process
- Reindex the narrative (if present) on any indexing of an object
- Testing now uses CI as an effort to migrate away from server response mocks
- Removing AMA information from config and adding ama_config file.
- AMA features are now only indexed as versioned objects.
- Added
sample_set
,sample_set_version
, and sample indices to config.yaml - Updating the sample indexer to include support for multiple source WS objects
- Fixed a bug in a bulk update Elasticsearch request function
- Moved the workspace type blacklist into spec/config.yaml
- RE importer now checks against the type blacklist as well
- Default indexes will automatically create an alias to
default_search
- Skip indexing of temporary narratives
- Updated configuration aliases to include most recent indexes under
default_search
- No longer copying publication title/author to agg_fields for genome_2
- Add more thorough spec validation and testing
- Fix some latest version alias names in the spec
- Fix a typo in the spec
- Using Github Actions for CI instead of Travis
- Added docker build and deployment to the github action
- Fetch the object type from the workspace when it is not provided by the kafka message
- Clean up logger and WorkspaceClient imports and initialization
- Index static narrative data from the workspace info
- Add workspace type whitelist/blacklist options (
ALLOW_TYPES
andSKIP_TYPES
) - Add unit testing setup with pytest
- Add a failure count limit in the consumer
- Move the
config.yaml
fromindex_runner_spec
tospec/config.yaml
- Centralize app version under
VERSION
- Increase batch write size and make configurable
- Consumer commits using the current message offset/partition
- Docs updated
- Fixed permissions errors for sample sets
- Sample and SampleSet indexers
- Prevent crash on workspace errors in the admin CLI