ores.storage
Table of Contents
1. Summary
ores.storage is ORE Studio's object storage: a bucket-and-key namespace of
opaque bytes, reachable over HTTP when a caller is moving a body and over NATS
when it is sending a command. It is a composite, and the three parts exist
because the contract, the implementation and the service have different
consumers and different dependencies.
- ores.storage.api — the contract: the bucket and key path helpers, and the modelled operation surface both interfaces are generated from.
- ores.storage.core — the implementation: the streaming HTTP client, the
tar.gzarchiver, and the atomic and composite transfer helpers. - ores.storage.service — the process: it binds the operation set's subjects, authenticates the caller and answers them.
Storage holds no bucket name of its own, no registry and no allow-list: a caller names its bucket, and which buckets exist is the caller's concern. The target is stated in Object Storage.
2. Parts
| Part | What it is for |
|---|---|
| ores.storage.api | The public contract a consumer links, and no dependency it does not need. |
| ores.storage.core | Where the bytes move: HTTP, archive and the transfer helpers. |
| ores.storage.service | The NATS entry point, and the process shell around it. |
3. See also
- Object Storage – the target state for the contract these parts implement.
- Data storage – where object storage sits among the other stores.
- Compute job lifecycle – the largest consumer, and the bucket and key scheme it owns.