What’s new
Version 2.1
Version 2.1 has two major updates:
It was migrated to the new CDS toolbox and the CDS API.
The local setup now prefers to use DuckDB instead of SQLite.
DuckDB integrates very well with pandas dataframes which allows to dump dataframes directly towards DuckDB tables. This avoids the overhead of going through SQL INSERT() statements or through intermediate CSV files for database loading. In fact, this works so well that duckdb is now the default setup. If there is no specific need for a relational client/server database (like MySQL, PostgreSQL), I suggest to stay with DuckDB as it performs extremely well and simplifies things a lot.
Version 2.0
Version 2.0 of AgERA5tools is completely restructured and can be used to set up a local mirror of AgERA5. The software will automatically download the requested AgERA5 data from the Copernicus Climate Data Store, store it in a relational database and make it available through a http API. A call to this API can be used to retrieve a time-series of AgERA5 data as JSON. This will allow many applications that require time-series data to efficiently query agERA5 data.