# `data`
## Scientific scope
Large modern astronomy datasets such as JPL ephemerides and SPICE kernels are
too large and too frequently revised to be treated like small compile-time
constants. This module exists to manage those external reference products in a
reproducible way: a caller can identify a dataset, fetch it into a local cache,
verify its integrity, and then hand the resulting bytes to the scientific code
that understands the format.
The scientific content of the data is not defined here; this module is the
logistics layer that keeps access to authoritative upstream products explicit,
auditable, and byte-stable.
## Technical scope
What you will find here:
- `registry.rs`: the pinned list of supported dataset identifiers and metadata.
- `manager.rs`, `download.rs`, `cache.rs`: dataset acquisition and cache
management when the `runtime-data` feature is enabled.
- `daf.rs`: Double Precision Array File parsing helpers.
- `spk.rs`: SPK-specific parsing primitives.
- `DataError`: the common error type for IO, integrity, and parsing failures.
Without `runtime-data`, the registry and parser-facing types remain available,
so dependent code can reason about datasets without pulling in network or cache
machinery.
## References
- NASA NAIF. SPK Required Reading.
<https://naif.jpl.nasa.gov/pub/naif/toolkit_docs/C/req/spk.html>
- Acton, C. H., Bachman, N., Semenov, B., and Wright, E. (2018). "A look
towards the future in the handling of space science mission geometry."
Planetary and Space Science 150, 9-12.
- Astropy Project documentation. Downloadable data management.
<https://docs.astropy.org/en/latest/utils/data.html>