François Massot
François Massot
**Is your feature request related to a problem? Please describe.** When using (badly) the scorer API and especially the `seek` method, I ended up in some infinite loop. It comes...
# Make semi structured-data a first-class citizen In many use cases, users/engineers don't know the data schema in advance but want to index it and query it fast. This is...
### Description Refactor a bit how we build cluster members and how we use it. One benefit is that you can define the search client pool without depending on `Cluster`...
### Description Add a metastore service with 2 implementations: - One backed by a `Metastore` instance. - One backed by a gRPC client. ### Important note The PR still needs...
In the search pool client, we update the pool with new cluster clients in a tokio task. An error can happen in the task, and the update will stop. We...
Viewed in the cluster implem: - ~~https://github.com/quickwit-oss/quickwit/blob/main/quickwit-cluster/src/cluster.rs#L257-L258~~ (false alert thanks @guilload) - https://github.com/quickwit-oss/quickwit/blob/main/quickwit-cluster/src/cluster.rs#L243-L245 I would remove them, or if there are ok (but I'm not convinced at all) this would...
## Current situation Garbage collection is runned on a indexer for each pipeline. The garbage collection of splits is intertwined with the `IndexingSplitStore`. This store is keeping splits locally for...
## Notes - Add the `IndexManagementService` in `quickwit-index-management` new project. - Implement the gRPC client so that other nodes can make gRPC calls for CRUD index operations. - Move the...
On main branch we have these routes for search and index management: - `/api/v1/{index-id}/search` - `/api/v1/{index-id}/search/stream` - `/api/v1/indexes` - `/api/v1/indexes/{index-id}` - `/api/v1/indexes/{index-id}/splits` I would prefer to have always the prefix...