Fix num features extracted

compass/services/threaded.py

@@ -613,11 +613,12 @@ def _dump_jurisdiction_info(
         "subdivision": jurisdiction.subdivision_name,
         "jurisdiction_type": jurisdiction.type,
         "FIPS": jurisdiction.code,
-        "found": False,
         "total_time": seconds_elapsed,
         "total_time_string": str(timedelta(seconds=seconds_elapsed)),
         "jurisdiction_website": None,
         "cost": None,
+        "found": False,
+        "num_features_extracted": None,
         "documents": None,
     }
 
@@ -630,6 +631,9 @@ def _dump_jurisdiction_info(
         new_info["documents"] = [
             _compile_doc_info(doc) for doc in extraction_context.data_docs
         ]
+        new_info["num_features_extracted"] = extraction_context.attrs.get(
+            "num_features_extracted"
+        )
         new_info["jurisdiction_website"] = extraction_context.attrs.get(
             "jurisdiction_website"
         )
@@ -645,7 +649,6 @@ def _compile_doc_info(doc):
     out_fp = doc.attrs.get("source_fp", doc.attrs.get("out_fp"))
     return {
         "source": doc.attrs.get("source"),
-        "num_features_extracted": doc.attrs.get("num_features_extracted"),
         "effective_year": year if year is not None and year > 0 else None,
         "effective_month": month if month is not None and month > 0 else None,
         "effective_day": day if day is not None and day > 0 else None,

compass/utilities/parsing.py

@@ -197,7 +197,7 @@ def num_ordinances_dataframe(data, exclude_features=None):
         mask = ~data["feature"].str.casefold().isin(exclude_features)
         data = data[mask].copy()
 
-    return ordinances_bool_index(data).sum()
+    return int(ordinances_bool_index(data).sum())
 
 
 def ordinances_bool_index(data):

tests/python/unit/services/test_services_threaded.py

@@ -542,6 +542,7 @@ async def test_jurisdiction_updater_process(tmp_path):
         doc,
         attrs={
             "jurisdiction_website": "http://jurisdiction.gov",
+            "num_features_extracted": 1,
         },
     )
     context.data_docs = [doc]
@@ -572,6 +573,7 @@ async def test_jurisdiction_updater_process(tmp_path):
     second = data["jurisdictions"][1]
     assert second["found"] is True
     assert second["jurisdiction_website"] == "http://jurisdiction.gov"
+    assert second["num_features_extracted"] == 1
     assert pytest.approx(second["cost"]) == 22.5
     assert second["documents"][0]["ord_filename"] == "doc.pdf"
     assert second["documents"][0]["effective_year"] == 2023

tests/python/unit/utilities/test_utilities_parsing.py

@@ -150,6 +150,23 @@ def test_num_ordinances_dataframe_with_values():
     assert num_ordinances_dataframe(data) == 3
 
 
+def test_num_ordinances_dataframe_returns_int_with_missing_values():
+    """Test ordinance count type with NaN and None values"""
+
+    data = pd.DataFrame(
+        {
+            "feature": ["setback", "height", "noise"],
+            "value": [100, np.nan, None],
+            "summary": [None, "test", None],
+        }
+    )
+
+    result = num_ordinances_dataframe(data)
+
+    assert result == 2
+    assert type(result) is int
+
+
 def test_num_ordinances_dataframe_with_exclude():
     """Test `num_ordinances_dataframe` with excluded features"""
 
@@ -163,6 +180,25 @@ def test_num_ordinances_dataframe_with_exclude():
     assert num_ordinances_dataframe(data, exclude_features=["height"]) == 2
 
 
+def test_num_ordinances_dataframe_with_exclude_all_features():
+    """Test ordinance count when exclusions remove all features"""
+
+    data = pd.DataFrame(
+        {
+            "feature": ["setback", "HEIGHT", "noise"],
+            "value": [100, 200, 300],
+            "summary": ["test", "test", "test"],
+        }
+    )
+
+    result = num_ordinances_dataframe(
+        data, exclude_features=["setback", "height", "noise"]
+    )
+
+    assert result == 0
+    assert type(result) is int
+
+
 def test_ordinances_bool_index_none():
     """Test `ordinances_bool_index` with None data"""