Datasets module with the GDP hourly accessor (#188)

* Add datasets module

* Test that the dataset opens

* Add datasets module to the docs index

* Update PyPI dependencies

* Update conda-forge dependencies

* Bump version to 0.15.0

Author: milancurcic

clouddrift/__init__.py

@@ -5,5 +5,6 @@
 from clouddrift.raggedarray import RaggedArray
 import clouddrift.adapters
 import clouddrift.analysis
+import clouddrift.datasets
 import clouddrift.haversine
 import clouddrift.sphere

clouddrift/datasets.py

@@ -0,0 +1,18 @@
+"""
+This module provides functions to easily access ragged-array datasets.
+"""
+
+import xarray as xr
+
+
+def gdp1h() -> xr.Dataset:
+    """
+    Returns the hourly GDP dataset as an Xarray dataset.
+
+    Returns
+    -------
+    xarray.Dataset
+        GDP1h dataset
+    """
+    url = "https://noaa-oar-hourly-gdp-pds.s3.amazonaws.com/latest/gdp_v2.00.zarr"
+    return xr.open_dataset(url, engine="zarr")

docs/api.rst

@@ -33,6 +33,13 @@ Analysis
   :members:
   :undoc-members:
 
+Datasets
+--------
+
+.. automodule:: clouddrift.datasets
+  :members:
+  :undoc-members:
+
 Haversine
 ---------
 

environment.yml

@@ -4,13 +4,16 @@ channels:
 dependencies:
   - python>=3.9
   - numpy>=1.21.6
-  - xarray>=2022.6.0
-  - netcdf4>=1.6.1
+  - xarray>=2023.5.0
+  - netcdf4>=1.6.4
   - pyarrow>=9.0.0
   - tqdm>=4.64.1
   - fsspec>=2022.8.2
   - llvmlite>=0.38.1
   - awkward>=2.0.0
   - pip>=22.2.2
+  - aiohttp>=3.8.4
+  - requests>=2.31.0
+  - zarr>=2.14.2
   - pip:
     - git+https://github.com/Cloud-Drift/clouddrift.git

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "clouddrift"
-version = "0.14.0"
+version = "0.15.0"
 authors = [
   { name="Shane Elipot", email="[email protected]" },
   { name="Philippe Miron", email="[email protected]" },
@@ -20,13 +20,16 @@ classifiers = [
 ]
 
 dependencies = [
+    "aiohttp>=3.8.4",
+    "awkward>=2.0.0",
+    "fsspec>=2022.3.0",
+    "netcdf4>=1.6.4",
     "numpy>=1.22.4",
-    "xarray>=2022.3.0",
-    "netcdf4>=1.5.8",
     "pyarrow>=8.0.0",
     "tqdm>=4.64.0",
-    "fsspec>=2022.3.0",
-    "awkward>=2.0.0",
+    "requests>=2.31.0",
+    "xarray>=2023.5.0",
+    "zarr>=2.14.2",
 ]
 
 [project.urls]

tests/datasets_tests.py

@@ -0,0 +1,12 @@
+from clouddrift import datasets
+import unittest
+
+
+if __name__ == "__main__":
+    unittest.main()
+
+
+class datasets_tests(unittest.TestCase):
+    def test_gdp1h_opens(self):
+        ds = datasets.gdp1h()
+        self.assertTrue(ds)