Fix issues from August 2025 data (#104)

sid-kap · web-flow · commit 189a635822e8 · 2026-01-03T01:04:07.000-05:00
diff --git a/python/housing_data/build_metros.py b/python/housing_data/build_metros.py
@@ -5,7 +5,7 @@
 
 
 def load_crosswalk_df(data_repo_path: Path) -> pd.DataFrame:
-    crosswalk_df = pd.read_csv(data_repo_path / "data/crosswalk/cbsa2fipsxw.csv")
+    crosswalk_df = pd.read_csv(data_repo_path / "data/crosswalk/cbsa2fipsxw_2023.csv")
 
     # Drop the μSAs, no one cares about them.
     # Most of them are just one county anyway, so showing the combined metro stats doesn't
diff --git a/python/housing_data/california_hcd_data.py b/python/housing_data/california_hcd_data.py
@@ -76,7 +76,9 @@ def load_california_hcd_data(
         ],
         None,
     )
-    assert df["building_type"].isnull().sum() == 0
+
+    assert df["building_type"].isnull().sum() < 50
+    df = df[df["building_type"].notnull()]
 
     df = df.rename(columns={"YEAR": "year"}).astype({"year": str})
 

Original file line number	Diff line number	Diff line change
`@@ -76,7 +76,9 @@ def load_california_hcd_data(`
`76`	`76`	`],`
`77`	`77`	`None,`
`78`	`78`	`)`
`79`		`- assert df["building_type"].isnull().sum() == 0`
	`79`	`+`
	`80`	`+ assert df["building_type"].isnull().sum() < 50`
	`81`	`+ df = df[df["building_type"].notnull()]`
`80`	`82`
`81`	`83`	`df = df.rename(columns={"YEAR": "year"}).astype({"year": str})`
`82`	`84`