Fix: add missing import of cast_and_arrow function to sp2xr_pipeline

scripts/sp2xr_pipeline.py

@@ -27,7 +27,12 @@ from sp2xr.distribution import (
     make_hist_meta,
 )
 from sp2xr.concentrations import add_concentrations
-from sp2xr.schema import CANONICAL_DTYPES, DEFAULT_FLOAT, enforce_schema
+from sp2xr.schema import (
+    cast_and_arrow,
+    CANONICAL_DTYPES,
+    DEFAULT_FLOAT,
+    enforce_schema,
+)
 
 
 def main():
@@ -334,7 +339,7 @@ def main():
 
             ddf_conc = ddf_pbp_hk_dt.map_partitions(
                 add_concentrations, dt=run_config["dt"], meta=meta_conc
-            ).map_partitions(_cast_and_arrow, meta=meta_conc)
+            ).map_partitions(cast_and_arrow, meta=meta_conc)
 
             idx_target = "datetime64[ns]"
             ddf_conc = ddf_conc.map_partitions(
@@ -481,7 +486,7 @@ def main():
                     BC_type=run_config["BC_type"],
                     t=1,
                     meta=meta_hist,
-                ).map_partitions(_cast_and_arrow, meta=meta_hist)
+                ).map_partitions(cast_and_arrow, meta=meta_hist)
                 results.append(ddf_out)
 
             # --- Scattering histogram
@@ -512,7 +517,7 @@ def main():
                 BC_type=None,
                 t=1,
                 meta=meta_hist,
-            ).map_partitions(_cast_and_arrow, meta=meta_hist)
+            ).map_partitions(cast_and_arrow, meta=meta_hist)
             results.append(ddf_scatt)
 
             # --- Timelag histogram

src/sp2xr/schema.py

@@ -176,7 +176,7 @@ def enforce_schema(
     return ddf
 
 
-def _cast_and_arrow(pdf: pd.DataFrame) -> pd.DataFrame:
+def cast_and_arrow(pdf: pd.DataFrame) -> pd.DataFrame:
     """
     Cast every column in *this* partition to the canonical dtype
     (or DEFAULT_FLOAT), then switch the frame to pyarrow-backed dtypes.