Cloud-optimized GeoTIFF ... Parallel I/O
Yet another attempt at creating a GeoTIFF reader, in Rust, with Python bindings.
cargo add --git https://github.com/weiji14/cog3pio.git
pip install git+https://github.com/weiji14/cog3pio.git
Tip
The API for this crate/library is still unstable and subject to change, so you may want to pin to a specific git commit using either:
cargo add --git https://github.com/weiji14/cog3pio.git --rev <sha>
pip install git+https://github.com/weiji14/cog3pio.git@<sha>
where <sha>
is a commit hashsum obtained from
https://github.com/weiji14/cog3pio/commits/main
use std::io::Cursor;
use bytes::Bytes;
use cog3pio::io::geotiff::read_geotiff;
use ndarray::Array3;
use object_store::path::Path;
use object_store::{parse_url, GetResult, ObjectStore};
use tokio;
use url::Url;
#[tokio::main]
async fn main() {
let cog_url: &str =
"https://github.com/cogeotiff/rio-tiler/raw/6.4.0/tests/fixtures/cog_nodata_nan.tif";
let tif_url: Url = Url::parse(cog_url).unwrap();
let (store, location): (Box<dyn ObjectStore>, Path) = parse_url(&tif_url).unwrap();
let stream: Cursor<Bytes> = {
let result: GetResult = store.get(&location).await.unwrap();
let bytes: Bytes = result.bytes().await.unwrap();
Cursor::new(bytes)
};
// Read GeoTIFF into an ndarray::Array
let arr: Array3<f32> = read_geotiff(stream).unwrap();
assert_eq!(arr.dim(), (1, 549, 549));
assert_eq!(arr[[0, 500, 500]], 0.13482364);
}
import numpy as np
from cog3pio import read_geotiff
# Read GeoTIFF into a numpy array
array: np.ndarray = read_geotiff(
path="https://github.com/cogeotiff/rio-tiler/raw/6.4.0/tests/fixtures/cog_nodata_nan.tif"
)
assert array.shape == (1, 549, 549) # bands, height, width
assert array.dtype == "float32"
import xarray as xr
# Read GeoTIFF into an xarray.DataArray
dataarray: xr.DataArray = xr.open_dataarray(
filename_or_obj="https://github.com/cogeotiff/rio-tiler/raw/6.4.1/tests/fixtures/cog_nodata_nan.tif",
engine="cog3pio",
)
assert dataarray.sizes == {'band': 1, 'y': 549, 'x': 549}
assert dataarray.dtype == "float32"
Note
Currently, this crate/library only supports reading single or multi-band float32 GeoTIFF files, i.e. other dtypes (e.g. uint16) don't work yet. See roadmap below on future plans.
Short term (Q1 2024):
- Multi-band reader to
ndarray
(relying onimage-tiff
) - Read from HTTP remote storage (using
object-store
)
Medium term (Q2 2024):
- Integration with
xarray
as aBackendEntrypoint
- Implement single-band GeoTIFF reader for multiple dtypes (uint/int/float) (relying
on
geotiff
crate)
Longer term (Q3-Q4 2024):
- Parallel reader (TBD on multi-threaded or asynchronous)
- Direct-to-GPU loading