--config¶
Why we added this flag: idxr needs a concrete mapping between models and source files; the config flag lets you point to that JSON document each time you run the pipeline.
What it does¶
- Path to the prepare-datasets JSON config (see Config Reference).
- Required so idxr knows which source files to read and how to map columns.
- Supports relative paths; they are resolved from the current working directory.
Typical usage¶
idxr prepare_datasets \
--model "$IDXR_MODEL" \
--config workdir/configs/prepare_datasets_contracts.json \
--output-root workdir/partitions
Tips¶
- Keep configs under
workdir/prepare_datasets(or similar) so they are easy to diff in Git. - Use different configs for full exports vs. patch drops to keep manifests easy to audit.