added to_netcdf() tests

.github/workflows/pytest.yml

@@ -15,7 +15,7 @@ jobs:
       fail-fast: false
       matrix:
         # we assume it also works for inbetween versions
-        python-version: ["3.9", "3.12", "3.14"]
+        python-version: ["3.9", "3.10", "3.12", "3.14"]
         os: [ubuntu-latest, windows-latest, macos-latest]
     runs-on: ${{ matrix.os }}
 

pyproject.toml

@@ -49,7 +49,7 @@ Issues = "https://github.com/deltares/ddlpy/issues"
 [project.optional-dependencies]
 dev = [
 	"bump2version>=0.5.11",
-	"flake8",
+	"black",
 	"pytest>=3.8.2",
 	"pytest-cov",
 	"twine",
@@ -72,7 +72,8 @@ examples = [
 ]
 netcdf = [
 	"xarray",
-	"h5netcdf",
+	"h5netcdf[h5py]",
+	"netcdf4",
 ]
 
 [project.scripts]
@@ -86,6 +87,9 @@ testpaths = ["tests"]
 addopts = "--durations=0"
 filterwarnings = [
 	"error",
+	# temporarily ignore numpy incompatibility: https://github.com/Deltares/ddlpy/issues/192
+	# this warning broke github tests for python 3.11, 3.12 and 3.13
+	"ignore:numpy.ndarray size changed:RuntimeWarning",
 	]
 
 [tool.flake8]

tests/conftest.py

@@ -0,0 +1,49 @@
+# -*- coding: utf-8 -*-
+"""
+Created on Wed Apr 22 09:56:57 2026
+
+@author: veenstra
+"""
+
+import pytest
+import ddlpy
+import datetime as dt
+
+
+@pytest.fixture(scope="session")
+def endpoints():
+    """
+    Get the endpoints from the api
+    """
+    endpoints = ddlpy.ddlpy.ENDPOINTS
+    return endpoints
+
+
+@pytest.fixture(scope="session")
+def locations():
+    """return all locations"""
+    locations = ddlpy.locations()
+    return locations
+
+
+@pytest.fixture(scope="session")
+def location(locations):
+    """return sample location"""
+    bool_grootheid = locations["Grootheid.Code"] == "WATHTE"
+    bool_groepering = locations["Groepering.Code"] == ""
+    bool_procestype = locations["ProcesType"] == "meting"
+    location = locations[bool_grootheid & bool_groepering & bool_procestype].loc[
+        "denhelder.marsdiep"
+    ]
+    return location
+
+
+@pytest.fixture(scope="session")
+def measurements(location):
+    """measurements for a location"""
+    start_date = dt.datetime(1953, 1, 1)
+    end_date = dt.datetime(1953, 4, 1)
+    measurements = ddlpy.measurements(
+        location, start_date=start_date, end_date=end_date
+    )
+    return measurements

tests/test_ddlpy.py

@@ -20,45 +20,6 @@
 }
 
 
-@pytest.fixture(scope="session")
-def endpoints():
-    """
-    Get the endpoints from the api
-    """
-    endpoints = ddlpy.ddlpy.ENDPOINTS
-    return endpoints
-
-
-@pytest.fixture(scope="session")
-def locations():
-    """return all locations"""
-    locations = ddlpy.locations()
-    return locations
-
-
-@pytest.fixture(scope="session")
-def location(locations):
-    """return sample location"""
-    bool_grootheid = locations["Grootheid.Code"] == "WATHTE"
-    bool_groepering = locations["Groepering.Code"] == ""
-    bool_procestype = locations["ProcesType"] == "meting"
-    location = locations[bool_grootheid & bool_groepering & bool_procestype].loc[
-        "denhelder.marsdiep"
-    ]
-    return location
-
-
-@pytest.fixture(scope="session")
-def measurements(location):
-    """measurements for a location"""
-    start_date = dt.datetime(1953, 1, 1)
-    end_date = dt.datetime(1953, 4, 1)
-    measurements = ddlpy.measurements(
-        location, start_date=start_date, end_date=end_date
-    )
-    return measurements
-
-
 def test_send_post_request_errors_wrongapi():
     url = "https://ddapi20-waterwebservices.rijkswaterstaat.nl/ONLINEWAARNEMINGENSERVICES/OphalenCatalogus"
     with pytest.raises(IOError) as e:
@@ -612,161 +573,3 @@ def test_check_convert_wrongorder():
     # assert output
     with pytest.raises(ValueError):
         _, _ = ddlpy.ddlpy._check_convert_dates(end_date, start_date)
-
-
-def test_simplify_dataframe(measurements):
-    """
-    should be in test_utils.py
-    """
-    assert len(measurements.columns) == 48
-    meas_simple = ddlpy.simplify_dataframe(measurements)
-    assert hasattr(meas_simple, "attrs")
-    # TODO: the below should be 47 and 1, but there are still RIKZ_WAT instances in
-    # OpdrachtgevendeInstantie column, which is different from RIKZMON_WAT
-    # this also probably partly causes the 96 duplicated timestamps
-    # https://github.com/Rijkswaterstaat/WaterWebservices/issues/16
-    assert len(meas_simple.attrs) == 46
-    assert len(meas_simple.columns) == 2
-    expected_columns = [
-        "WaarnemingMetadata.OpdrachtgevendeInstantie",
-        "Meetwaarde.Waarde_Numeriek",
-    ]
-    assert set(meas_simple.columns) == set(expected_columns)
-
-
-def test_simplify_dataframe_always_preserve(measurements):
-    """
-    should be in test_utils.py
-    """
-    assert len(measurements.columns) == 48
-    always_preserve = [
-        "WaarnemingMetadata.Statuswaarde",
-        "WaarnemingMetadata.OpdrachtgevendeInstantie",
-        "WaarnemingMetadata.Kwaliteitswaardecode",
-        "Groepering.Code",
-        "BemonsteringsApparaat.Code",
-        "Meetwaarde.Waarde_Numeriek",
-    ]
-    meas_simple = ddlpy.simplify_dataframe(
-        measurements, always_preserve=always_preserve
-    )
-    assert hasattr(meas_simple, "attrs")
-    assert len(meas_simple.attrs) == 42
-    assert len(meas_simple.columns) == 6
-    expected_columns = [
-        "WaarnemingMetadata.Statuswaarde",
-        "WaarnemingMetadata.OpdrachtgevendeInstantie",
-        "WaarnemingMetadata.Kwaliteitswaardecode",
-        "Groepering.Code",
-        "BemonsteringsApparaat.Code",
-        "Meetwaarde.Waarde_Numeriek",
-    ]
-    assert set(meas_simple.columns) == set(expected_columns)
-
-
-def test_simplify_dataframe_always_preserve_invalid_key(measurements):
-    """
-    should be in test_utils.py
-    """
-    assert len(measurements.columns) == 48
-    always_preserve = ["invalid_key"]
-    with pytest.raises(ValueError) as e:
-        _ = ddlpy.simplify_dataframe(measurements, always_preserve=always_preserve)
-    assert "column 'invalid_key' not present in dataframe" in str(e.value)
-
-
-def test_simplify_dataframe_alfanumeriek_with_nan_dropped(locations):
-    bool_grootheid = locations["Grootheid.Code"] == "WATHTE"
-    bool_groepering = locations["Groepering.Code"] == ""
-    bool_procestype = locations["ProcesType"] == "meting"
-    location = locations[bool_grootheid & bool_groepering & bool_procestype].loc["a12"]
-
-    start_date = dt.datetime(2009, 1, 1)
-    end_date = dt.datetime(2009, 4, 1)
-    measurements = ddlpy.measurements(
-        location, start_date=start_date, end_date=end_date
-    )
-    meas_simple = ddlpy.simplify_dataframe(df=measurements)
-    expected_columns = [
-        "WaarnemingMetadata.Kwaliteitswaardecode",
-        "Meetwaarde.Waarde_Numeriek",
-    ]
-    assert set(meas_simple.columns) == set(expected_columns)
-
-
-def test_dataframe_to_xarray(measurements):
-    """
-    should be in test_utils.py
-    """
-    always_preserve = [
-        "WaarnemingMetadata.Statuswaarde",
-        "WaarnemingMetadata.Kwaliteitswaardecode",
-        "MeetApparaat.Code",
-        "WaardeBepalingsMethode.Code",
-        "Meetwaarde.Waarde_Numeriek",
-    ]
-    ds_clean = ddlpy.dataframe_to_xarray(
-        df=measurements,
-        always_preserve=always_preserve,
-    )
-
-    non_constant_columns = [
-        "WaarnemingMetadata.OpdrachtgevendeInstantie",
-        "Meetwaarde.Waarde_Numeriek",
-    ]
-
-    preserved = always_preserve + non_constant_columns
-
-    for varname in measurements.columns:
-        # check if all varnames in always_preserve and non-constant columns are indeed preserved as variables
-        if varname in preserved:
-            assert varname in ds_clean.data_vars
-            assert varname not in ds_clean.attrs.keys()
-        else:
-            assert varname not in ds_clean.data_vars
-            assert varname in ds_clean.attrs.keys()
-            varname_oms = varname.replace(".Code", ".Omschrijving")
-            assert varname_oms in ds_clean.attrs.keys()
-
-    # check if times and timezone are correct
-    refdate_utc = measurements.tz_convert(None).index[0]
-    ds_firsttime = ds_clean.time.to_pandas().iloc[0]
-    assert refdate_utc == ds_firsttime
-    assert ds_firsttime.tz is None
-
-
-def test_dataframe_to_xarray_drop_omschrijving(measurements):
-    """
-    in case of non-unique Code/Omschrijving pairs, the Omschrijving variable should be
-    dropped also. The information it contains is added as attrs to the Code value.
-    """
-    # make MeetApparaat non-unique
-    measurements.loc["1953-01-01 02:40:00+01:00", "MeetApparaat.Code"] = "newcode"
-    measurements.loc["1953-01-01 02:40:00+01:00", "MeetApparaat.Omschrijving"] = (
-        "newoms"
-    )
-
-    always_preserve = [
-        "WaarnemingMetadata.Statuswaarde",
-        "WaarnemingMetadata.Kwaliteitswaardecode",
-        "WaardeBepalingsMethode.Code",
-        "Meetwaarde.Waarde_Numeriek",
-    ]
-
-    ds = ddlpy.dataframe_to_xarray(measurements, always_preserve=always_preserve)
-    for varn in ds.data_vars:
-        assert not varn.endswith(".Omschrijving")
-
-    expected_attrs = {"newcode": "newoms", "10272": "other:Vlotterniveaumeter"}
-    assert ds["MeetApparaat.Code"].attrs == expected_attrs
-
-
-def test_code_description_attrs_from_dataframe_prevent_empty(measurements):
-    """
-    should be in test_utils.py
-    https://github.com/Deltares/ddlpy/issues/156
-    """
-    assert "" in measurements["Groepering.Code"].unique()
-    attr_dict = ddlpy.utils.code_description_attrs_from_dataframe(measurements)
-    for attr_key_value_pairs in attr_dict.values():
-        assert "" not in attr_key_value_pairs.keys()