You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We made CSV data sources appendable to have a quick proof of concept, used with a single-process server to prototype flyscanning. But appending file-based storage is not safe/robust in a horizontally scaled deployment. #779 provides a better way. It includes support for single-process (DuckDB) or scaled (PostgreSQL) storage. I think therefore we should remove support for appendable CSVs.
This is blocked by releasing #779 and then releasing a version of TiledWriter in bluesky that uses SQL instead of CSV.
The text was updated successfully, but these errors were encountered:
We made CSV data sources appendable to have a quick proof of concept, used with a single-process server to prototype flyscanning. But appending file-based storage is not safe/robust in a horizontally scaled deployment. #779 provides a better way. It includes support for single-process (DuckDB) or scaled (PostgreSQL) storage. I think therefore we should remove support for appendable CSVs.
This is blocked by releasing #779 and then releasing a version of TiledWriter in bluesky that uses SQL instead of CSV.
The text was updated successfully, but these errors were encountered: