Announcing pandaSDMX 1.10.0 on PyPI and conda-forge

[Leo]

Hi,

v1.10.0 is out. It supports the new API of Eurostat as well as three new data sources provided by the European Commission, namely

- EC_COMP: DG Competition

- EC_EMPL: DG Employmentt, and

- EC_GROW: DG Growth..

Further, v1.10.0 allows you to have the pandas dtype be selected by casting the data types ("facet value types") specified in the datastructure definition to corresponding pandas dtypes. `float64` remains the default as in all previous versions. And the option to use the pandas dtype "object" as a fallback is still available. The usefulness of the new feature depends, of course, on the ambition of the data providers to specify the most appropriate data type in their DSD's. I saw some DSD's specifying just string as data types whereas some decimal type would have appeared as the better choice.

Moreover, coded dimensions can be represented as dtype categorical in pandas DataFrames/Series rather than object/str.

v1.10.0 requires Python 1.9.6+ including support for Python 3.11.x. I hope this won't bother too many.

Feedback and code contributions are highly appreciated as always.

Have fun!

Leo