feat!: simplify schema for coordinate dataframe. Use regular data typ…

…es (str instead of CategoricalDtype) and give coordinates as floats

pyproject.toml

@@ -8,7 +8,7 @@ authors = [{name="Statnett Datascience", email="Datascience.Drift@Statnett.no"}]
 requires-python = ">=3.11"
 dependencies = [
     "httpx>=0.27.2",
-    "numpy>=2.1.1",
+    "numpy>=1.9.0",
     "pandas>=2.2.3",
     "pandera>=0.20.4",
     "polyfactory>=2.17.0",

src/cimsparql/data_models.py

@@ -1,7 +1,6 @@
 import datetime as dt
 from typing import Self
 
-import pandas as pd
 import pandera as pa
 from pandera.typing import DataFrame, Index, Series
 
@@ -189,10 +188,10 @@ class TransfConToConverterSchema(NamedResourceSchema):
 
 class CoordinatesSchema(JsonSchemaOut):
     mrid: Series[str] = pa.Field()
-    x: Series[str] = pa.Field()
-    y: Series[str] = pa.Field()
-    epsg: Series[pd.CategoricalDtype] = pa.Field()
-    rdf_type: Series[pd.CategoricalDtype] = pa.Field()
+    x: Series[float] = pa.Field()
+    y: Series[float] = pa.Field()
+    epsg: Series[str] = pa.Field()
+    rdf_type: Series[str] = pa.Field()
 
 
 CoordinatesDataFrame = DataFrame[CoordinatesSchema]

tests/test_micro_t1_nl.py

@@ -3,7 +3,6 @@
 from string import Template
 from typing import Any
 
-import pandas as pd
 import pytest
 
 import tests.t_utils.common as t_common
@@ -268,11 +267,11 @@ def test_coordinates(test_model: t_common.ModelTest):
     assert test_model.model
     model = test_model.model
     crd = model.coordinates()
-    pd.testing.assert_index_equal(crd["epsg"].cat.categories, pd.Index(["4326"], dtype=str))
+    assert set(crd["epsg"]), {"4326"}
 
     cim = model.client.prefixes["cim"]
     categories = {f"{cim}ACLineSegment", f"{cim}Substation"}
-    assert crd["rdf_type"].cat.categories.difference(categories).empty
+    assert set(crd["rdf_type"]).difference(categories) == set()
 
     assert len(crd) == 49
     coordinates = crd.astype({"x": float, "y": float})