mrd-storage-client

A Python client for the MRD Storage Server

Start an MRD Storage Server using Docker:

docker run -p 3333:3333 ghcr.io/ismrmrd/mrd-storage-server:latest

Simple usage:

from mrd_storage_client import Storage

storage = Storage("localhost", 3333)
example_dict = {"key": "value"}
storage.store(example_dict)
fetched_dict = storage.fetch_latest()
assert example_dict == fetched_dict
storage.close()

A more advanced example:

import numpy as np
from mrd_storage_client import Storage

storage = Storage("localhost", 3333, subject="my_test_subject")

# Store five random arrays with 1 minute TTL and a custom tag "array_idx"
arrays = (np.random.rand(32, 32) for _ in range(5))
for idx, my_array in enumerate(arrays):
    storage.store(my_array, name="my_arrays", ttl="1m", custom_tags={"array_idx": idx})

# Fetch blobs as an iterator
for blob in storage.fetch_blobs(name="my_arrays"):
    print(f"Got array_idx = {blob.get('array_idx')}")
storage.close()

For a complete API documentation, see the mrd-storage-server repo.