Make datacube optional by moving imports to function level (#146)

* Move datacube imports to function level

* Fix missing import

* Make sure geometry remains a tuple

* Automatically update integration test validation results

---------

Co-authored-by: robbibt <robbibt@users.noreply.github.com>

Author: robbibt

intertidal/composites.py

@@ -2,11 +2,9 @@
 import sys
 
 import click
-import datacube
 import numpy as np
 import odc.geo.xr
 import xarray as xr
-from datacube.utils.aws import configure_s3_access
 from dea_tools.dask import create_local_dask_cluster
 from eo_tides.eo import pixel_tides
 from odc.algo import (
@@ -439,6 +437,18 @@ def tidal_composites_cli(
     aws_unsigned,
     overwrite,
 ):
+    # Attempt to import datacube and raise an error if not available
+    try:
+        import datacube
+        from datacube.utils.aws import configure_s3_access
+    except ImportError as e:
+        msg = (
+            "The DEA Tidal Composites CLI is configured for Australian applications, and "
+            "requires `datacube`. Please install DEA Intertidal with the "
+            "`[datacube]` extra, e.g.: `pip install dea-intertidal[datacube]`"
+        )
+        raise ImportError(msg) from e
+
     # Create sample filename to test if data exists on file system
     filename = f"{output_dir}ga_s2_tidal_composites_cyear_3/{output_version.replace('.', '-')}/{study_area[:4]}/{study_area[4:]}/{label_date}--P1Y/ga_s2_tidal_composites_cyear_3_{study_area}_{label_date}--P1Y_final.stac-item.json"
 

intertidal/elevation.py

@@ -3,12 +3,10 @@
 from itertools import repeat
 
 import click
-import datacube
 import matplotlib.pyplot as plt
 import numpy as np
 import seaborn as sns
 import xarray as xr
-from datacube.utils.aws import configure_s3_access
 from dea_tools.dask import create_local_dask_cluster
 from eo_tides.eo import pixel_tides
 from odc.algo import xr_quantile
@@ -1125,6 +1123,18 @@ def intertidal_cli(
     exposure_offsets,
     aws_unsigned,
 ):
+    # Attempt to import datacube and raise an error if not available
+    try:
+        import datacube
+        from datacube.utils.aws import configure_s3_access
+    except ImportError as e:
+        msg = (
+            "The DEA Intertidal CLI is configured for Australian applications, and "
+            "requires `datacube`. Please install DEA Intertidal with the "
+            "`[datacube]` extra, e.g.: `pip install dea-intertidal[datacube]`"
+        )
+        raise ImportError(msg) from e
+
     # Create a unique run ID for analysis based on input params and use
     # for logs
     input_params = locals()

intertidal/exposure.py

@@ -4,7 +4,6 @@
 import numpy as np
 import pandas as pd
 import pytz
-import sunriset
 import xarray as xr
 from eo_tides.eo import _pixel_tides_resample, pixel_tides
 
@@ -38,6 +37,9 @@ def temporal_filters(x, time_range, dem):
         timesteps.
 
     """
+    # Import sunriset here to avoid dependency issues if function not used
+    import sunriset
+
     if x == "dry":
         return time_range.drop(
             time_range[

intertidal/io.py

@@ -11,8 +11,6 @@
 import numpy as np
 import odc.geo.xr
 import xarray as xr
-from datacube.utils.geometry import Geometry as Geometry_datacube18
-from datacube.utils.masking import mask_invalid_data
 from dea_tools.coastal import glint_angle
 from eo_tides.stats import tide_stats
 from eodatasets3 import DatasetAssembler, serialise
@@ -116,8 +114,14 @@ def extract_geobox(
         the CRS, resolution, shape and extent of the study area).
 
     """
-    # List of valid input geometry types (from `odc-geo` or `datacube-core`)
-    GEOM_TYPES = (odc.geo.geom.Geometry, Geometry_datacube18)
+    # List of valid input geometry types (from `odc-geo` or `datacube`).
+    # If `datacube` is not installed, only support `odc-geo` geometries
+    try:
+        from datacube.utils.geometry import Geometry as Geometry_datacube18
+
+        geom_types = (odc.geo.geom.Geometry, Geometry_datacube18)
+    except ImportError:
+        geom_types = (odc.geo.geom.Geometry,)
 
     # Either `study_area` or `geom` must be provided
     if study_area is None and geom is None:
@@ -126,11 +130,11 @@ def extract_geobox(
         )
 
     # If custom geom is provided, verify it is a geometry
-    if geom is not None and not isinstance(geom, GEOM_TYPES):
+    if geom is not None and not isinstance(geom, geom_types):
         raise ValueError("Unsupported input type for `geom`; please provide a datacube Geometry object.")
 
     # Otherwise, extract GeoBox from geometry
-    if geom is not None and isinstance(geom, GEOM_TYPES):
+    if geom is not None and isinstance(geom, geom_types):
         geobox = GeoBox.from_geopolygon(geom, crs=crs, resolution=resolution)
 
     # If no custom geom provided, load tile from GridSpec tile grid
@@ -253,6 +257,18 @@ def load_data(
         to generate ODC lineage metadata for DEA Intertidal)
 
     """
+    # Attempt to import datacube and raise an error if not available
+    try:
+        from datacube.utils.masking import mask_invalid_data
+    except ImportError as e:
+        msg = (
+            "The `load_data` function requires `datacube` to be installed. "
+            "Please consider loading data with `odc-stac` instead, or install "
+            "DEA Intertidal with the `[datacube]` extra, e.g.: `pip install "
+            "dea-intertidal[datacube]`"
+        )
+        raise ImportError(msg) from e
+
     # Set spectral bands to load
     s2_spectral_bands = [
         "nbart_blue",
@@ -516,6 +532,18 @@ def load_topobathy_mask(
         following analysis.
 
     """
+    # Attempt to import datacube and raise an error if not available
+    try:
+        from datacube.utils.masking import mask_invalid_data
+    except ImportError as e:
+        msg = (
+            "The `load_topobathy_mask` function requires `datacube` to be installed. "
+            "Please consider loading data with `odc-stac` instead, or install "
+            "DEA Intertidal with the `[datacube]` extra, e.g.: `pip install "
+            "dea-intertidal[datacube]`"
+        )
+        raise ImportError(msg) from e
+
     # Load from datacube, reprojecting to GeoBox of input satellite data
     topobathy_ds = dc.load(product=product, like=geobox, resampling=resampling).squeeze("time")
 
@@ -572,6 +600,18 @@ def load_aclum_mask(
         False equals all other classes.
 
     """
+    # Attempt to import datacube and raise an error if not available
+    try:
+        from datacube.utils.masking import mask_invalid_data
+    except ImportError as e:
+        msg = (
+            "The `load_aclum_mask` function requires `datacube` to be installed. "
+            "Please consider loading data with `odc-stac` instead, or install "
+            "DEA Intertidal with the `[datacube]` extra, e.g.: `pip install "
+            "dea-intertidal[datacube]`"
+        )
+        raise ImportError(msg) from e
+
     try:
         # Load from datacube, reprojecting to GeoBox of input satellite data
         aclum_ds = dc.load(product=product, like=geobox, resampling=resampling).squeeze("time")

tests/README.md

@@ -10,7 +10,7 @@ Integration tests
 
 This directory contains tests that are run to verify that DEA Intertidal code runs correctly. The ``test_intertidal.py`` file runs a small-scale full workflow analysis over an intertidal flat in the Gulf of Carpentaria using the DEA Intertidal [Command Line Interface (CLI) tools](../notebooks/Intertidal_CLI.ipynb), and compares these results against a LiDAR validation DEM to produce some simple accuracy metrics.
 
-The latest integration test completed at **2025-09-24 16:17**. Compared to the previous run, it had an:
+The latest integration test completed at **2025-09-24 17:19**. Compared to the previous run, it had an:
 - RMSE accuracy of **0.14 m ( :heavy_minus_sign: no change)**
 - MAE accuracy of **0.12 m ( :heavy_minus_sign: no change)**
 - Bias of **0.12 m ( :heavy_minus_sign: no change)**

tests/validation.csv

@@ -131,3 +131,4 @@ time,Correlation,RMSE,MAE,R-squared,Bias,Regression slope
 2025-09-24 05:21:37.998155+00:00,0.975,0.145,0.122,0.95,0.116,1.121
 2025-09-24 06:00:07.113646+00:00,0.975,0.145,0.122,0.95,0.116,1.121
 2025-09-24 06:17:15.044540+00:00,0.975,0.145,0.122,0.95,0.116,1.121
+2025-09-24 07:19:12.704574+00:00,0.975,0.145,0.122,0.95,0.116,1.121