simple-zarr-server

A minimal server for sharing zarr over HTTP.

Installation

You can install simple-zarr-server via pip:

pip install simple-zarr-server

Usage

CLI:

$ simple-zarr-server --help

# Usage: simple-zarr-server [OPTIONS] PATH
#
# Options:
#  --cors TEXT                     Origin to allow CORS. Use wildcard '*' to allow all.
#  -w, --allow-write
#  --host TEXT                     Bind socket to this host.  [default: 127.0.0.1]
#
#  --port INTEGER                  Bind socket to this port.  [default: 8000]
#  --reload                        Enable auto-reload.
#  --loop [auto|asyncio|uvloop]    Event loop implementation.  [default: auto]
#  --http [auto|h11|httptools]     HTTP protocol implementation.  [default: auto]
#
#  --ws [auto|none|websockets|wsproto]
#                                  WebSocket protocol implementation. [default: auto]
#
#  --use-colors / --no-use-colors  Enable/Disable colorized logging.
#  --proxy-headers / --no-proxy-headers
#                                  Enable/Disable X-Forwarded-Proto,
#                                  X-Forwarded-For, X-Forwarded-Port to
#                                  populate remote address info.
#
#  --forwarded-allow-ips TEXT      Comma seperated list of IPs to trust with
#                                  proxy headers. Defaults to the
#                                  $FORWARDED_ALLOW_IPS environment variable if
#                                  available, or '127.0.0.1'.

$ simple-zarr-server /dataset.zarr # or /dataset.n5, or /dataset.zip
# INFO:     Uvicorn running on http://127.0.0.1:8000 (Press CTRL+C to quit)

Python API:

The python API is more flexible than the CLI, and can serve any zarr.Array, zarr.Group or np.ndarray.

Server

from simple_zarr_server import serve
import numpy as np
arr = np.random.rand(1024, 1024)
serve(arr) # creates an in-memory store if not zarr.Array or zarr.Group

Client

`zarr-python`

import zarr
from fsspec import get_mapper
store = get_mapper("http://localhost:8000")
arr = zarr.open(store, mode='r')
# or 
import dask.array as da
arr = da.from_zarr("http://localhost:8000")

`zarr.js`

import { openArray } from 'zarr';
arr = await openArray({ store: 'http://localhost:8000' });

Advanced: Serving a remote pyramidal tiff as Zarr

Server

from napari_lazy_openslide import OpenSlideStore
from simple_zarr_server import serve
import zarr

store = OpenSlideStore('tumor_004.tif') # custom zarr store
grp = zarr.open(store)
serve(grp)

Client

import napari
import dask.array as da
import zarr
from fsspec import get_mapper

store = get_mapper("http://localhost:8000")
z_grp = zarr.open(store)
datasets = z_grp.attrs["multiscales"][0]["datasets"]
pyramid = [
    da.from_zarr(store, component=d["path"]) for d in datasets
]
with napari.gui_qt():
    napari.view_image(pyramid)

Note

This package is experimental. It wraps any zarr-python store as a REST API, enabling remote access over HTTP. It is similar to xpublish, but is more minimal and does not provide special endpoints that are specific to Xarray datasets. If your data are Xarray dataset, please use xpublish! simple-zarr-server was designed with imaging data in mind, and when combined with a tool like ngrok provides an interesting way to share local images with collaborators.

Some non-standard zarr stores that might be of interest include: