Using object storage¶

obstore is a new, powerful Python library for getting from and putting to object storage. rustac can use anything that implements obstore.store.ObjectStore, and also provides its own zero-dependency version in rustac.store.

In [3]:

import contextily
import pandas
import rustac
from rustac.store import HTTPStore
from geopandas import GeoDataFrame

store = HTTPStore(
    "https://raw.githubusercontent.com/stac-utils/rustac-py/refs/heads/main/data"
)
items = await rustac.read("100-sentinel-2-items.parquet", store=store)
print(len(items["features"]))

data_frame = GeoDataFrame.from_features(items)
data_frame["datetime"] = pandas.to_datetime(data_frame["datetime"])
axis = data_frame.set_crs(epsg=4326).to_crs(epsg=3857).plot(alpha=0.5, edgecolor="k")
contextily.add_basemap(axis, source=contextily.providers.CartoDB.Positron)
axis.set_axis_off()

import contextily import pandas import rustac from rustac.store import HTTPStore from geopandas import GeoDataFrame store = HTTPStore( "https://raw.githubusercontent.com/stac-utils/rustac-py/refs/heads/main/data" ) items = await rustac.read("100-sentinel-2-items.parquet", store=store) print(len(items["features"])) data_frame = GeoDataFrame.from_features(items) data_frame["datetime"] = pandas.to_datetime(data_frame["datetime"]) axis = data_frame.set_crs(epsg=4326).to_crs(epsg=3857).plot(alpha=0.5, edgecolor="k") contextily.add_basemap(axis, source=contextily.providers.CartoDB.Positron) axis.set_axis_off()

100

There's a whole set of provided object storage backends. See https://developmentseed.org/obstore/latest/authentication/ for how you might configure authenticate with those backends.

Using obstore.store.ObjectStore.¶

If you're doing work with obstore directly, you can re-use the same ObjectStore.

In [4]:

from obstore.store import HTTPStore

store = HTTPStore(
    "https://raw.githubusercontent.com/stac-utils/rustac-py/refs/heads/main/data"
)
items = await rustac.read("100-sentinel-2-items.parquet", store=store)

from obstore.store import HTTPStore store = HTTPStore( "https://raw.githubusercontent.com/stac-utils/rustac-py/refs/heads/main/data" ) items = await rustac.read("100-sentinel-2-items.parquet", store=store)

/var/folders/yp/d6xvrkd943dgvqg5s9cpymc40000gn/T/ipykernel_65491/1946367820.py:4: RuntimeWarning: Successfully reconstructed a store defined in another Python module. Connection pooling will not be shared across store instances.
  items = await rustac.read("100-sentinel-2-items.parquet", store=store)

Notice how there's a warning? That's because we have to copy the ObjectStore when we pass it in to rustac. This means it can be a bit more efficient to use rustac.store if you're only working with rustac.