From 89b0cc745e3f577d6c822fc27907beee0d422298 Mon Sep 17 00:00:00 2001
From: dvp <d.portnov@iterrf.ru>
Date: Thu, 5 Sep 2024 14:38:48 +0300
Subject: [PATCH] style: add typing

---
 src/xpypact/collector.py                 | 20 +++++++++++++-------
 src/xpypact/dao/duckdb/implementation.py |  5 ++++-
 2 files changed, 17 insertions(+), 8 deletions(-)

diff --git a/src/xpypact/collector.py b/src/xpypact/collector.py
index 04b618d5..76d2aea5 100644
--- a/src/xpypact/collector.py
+++ b/src/xpypact/collector.py
@@ -143,7 +143,7 @@ def append(self, inventory: Inventory, material_id: int, case_id: int) -> FullDa
 
         return self
 
-    def _append_rundata(self, inventory, material_id, case_id):
+    def _append_rundata(self, inventory: Inventory, material_id: int, case_id: int) -> None:
         rundata = inventory.meta_info
         st = strptime(rundata.timestamp, "%H:%M:%S %d %B %Y")
         ts = dt.datetime(  # noqa: DTZ001 - no tzinfo is available from the FISPACT output
@@ -167,12 +167,12 @@ def _append_rundata(self, inventory, material_id, case_id):
                     rundata.dose_rate_distance,
                 ),
             ],
-            schema=RunDataSchema,
+            schema=RunDataSchema,  # type: ignore[arg-type]
             orient="row",
         ).with_columns(pl.col("dose_rate_distance").round(5))
         self.rundata = pl.concat([self.rundata, rundata_df], how="vertical", rechunk=False)
 
-    def _append_timesteps(self, inventory, material_id, case_id):
+    def _append_timesteps(self, inventory: Inventory, material_id: int, case_id: int) -> None:
         timesteps_df = pl.DataFrame(
             (
                 (
@@ -204,7 +204,9 @@ def _append_timesteps(self, inventory, material_id, case_id):
         )
         self.timesteps = pl.concat([self.timesteps, timesteps_df], how="vertical", rechunk=False)
 
-    def _append_timestep_nuclides(self, inventory, material_id, case_id):
+    def _append_timestep_nuclides(
+        self, inventory: Inventory, material_id: int, case_id: int
+    ) -> None:
         timesteps_nuclides_df = pl.DataFrame(
             (
                 (
@@ -222,7 +224,7 @@ def _append_timestep_nuclides(self, inventory, material_id, case_id):
             rechunk=False,
         )
 
-    def _append_timestep_gamma(self, inventory, material_id, case_id):
+    def _append_timestep_gamma(self, inventory: Inventory, material_id: int, case_id: int) -> None:
         gs = inventory.inventory_data[-1].gamma_spectrum
 
         if not gs:
@@ -251,7 +253,7 @@ def _append_timestep_gamma(self, inventory, material_id, case_id):
             rechunk=False,
         )
 
-    def _get_timestep_times(self):
+    def _get_timestep_times(self) -> pl.DataFrame:
         first_case = self.timesteps.lazy().select("material_id", "case_id").limit(1)
         return (
             self.timesteps.lazy()
@@ -337,7 +339,11 @@ def get_timestep_gamma_as_spectrum(self) -> pl.DataFrame | None:
         return ql.collect()
 
     class Result(ms.Struct):
-        """Finished collected data."""
+        """Finished collected data.
+
+        The only function of this class is to save
+        the collected data to parquet files.
+        """
 
         rundata: pl.DataFrame
         time_step_times: pl.DataFrame  # use time_step_ prefix here
diff --git a/src/xpypact/dao/duckdb/implementation.py b/src/xpypact/dao/duckdb/implementation.py
index 3371b314..5bc1a3b0 100644
--- a/src/xpypact/dao/duckdb/implementation.py
+++ b/src/xpypact/dao/duckdb/implementation.py
@@ -155,7 +155,10 @@ def create_indices(con: db.DuckDBPyConnection) -> db.DuckDBPyConnection:
     """Create primary key like indices on tables after loading.
 
     Note:
-        use only for debugging
+        indexes are not used and, even more, are harmful in DuckDB.
+        This is provided to use the indexes only for testing and debugging the database.
+        The test is, that in a valid database with the loaded content
+        the indexes should be created successfully.
     """
     return con.execute(
         """