Usage¶

How to use the package. Design is what it is and why; API Reference is the generated signatures.

Installation¶

pip install datarecord           # core
pip install datarecord[pypsa]    # with the PyPSA tool

Or with conda/mamba, from conda-forge:

conda install -c conda-forge datarecord

Connections¶

Every entry point takes a DuckDB connection. It is passed as a parameter throughout, never a module global, and is scoped to one record root — which is how the schema beside its layers is found without a separate argument.

from datarecord import connect

con = connect(":memory:", base_uri="s3://bucket/my-record")

connect opens a connection carrying the revisions table and the path macros, loading httpfs and S3 credentials only for a remote base_uri. base_uri defaults to the DATARECORD_BASE_URI environment variable.

The pages¶

page	what it covers
Reading a record	the `Record` protocol, frames, `flags`
The schema	declaring dims and attributes
Layers	revisions, branching, materialising
Editing	`WorkingRecord`: `set`, `add`, `commit`
Writing	`write_record`
Tools	consuming a record from a modelling framework