Datetime type fixes

Author: rickybassom

src/gmn_python_api/trajectory_summary_reader.py

@@ -6,8 +6,8 @@
 import os.path
 from io import StringIO
 from typing import Any
+from typing import Optional
 
-import numpy as np
 import numpy.typing as npt
 import pandas as pd  # type: ignore
 from pandas._typing import FilePathOrBuffer  # type: ignore
@@ -20,8 +20,9 @@
 
 def read_trajectory_summary_as_dataframe(
     filepath_or_buffer: FilePathOrBuffer,
-    camel_case_column_names: bool = False,
-    avro_compatible: bool = False,
+    camel_case_column_names: Optional[bool] = False,
+    avro_compatible: Optional[bool] = False,
+    avro_long_beginning_utc_time: Optional[bool] = True,
 ) -> pd.DataFrame:
     """
     Reads a trajectory summary file into a Pandas DataFrame.
@@ -30,6 +31,9 @@ def read_trajectory_summary_as_dataframe(
     :param camel_case_column_names: If True, column names will be camel cased e.g. m_deg
     :param avro_compatible: If True, the rows in the dataframe will match the avsc
      schema with row.to_dict().
+    :param avro_long_beginning_utc_time: If True, the time column will be converted from
+     a datetime object to an int64 epoch time which is compatible with the long
+     timestamp-micros avro type.
 
     :return: Pandas DataFrame of the trajectory summary file.
     """
@@ -78,12 +82,16 @@ def read_trajectory_summary_as_dataframe(
 
     if avro_compatible:
         camel_case_column_names = True
-        trajectory_df["Beginning (UTC Time)"] = trajectory_df[
-            "Beginning (UTC Time)"
-        ].astype(np.int64) / int(1e6)
-        trajectory_df["Beginning (UTC Time)"] = trajectory_df[
-            "Beginning (UTC Time)"
-        ].astype(np.int64)
+
+        if avro_long_beginning_utc_time:
+            # convert datetime nano to micro epoch and round to int
+            trajectory_df["Beginning (UTC Time)"] = (
+                trajectory_df["Beginning (UTC Time)"].astype("int64") / 1e3
+            )
+            trajectory_df["Beginning (UTC Time)"] = (
+                trajectory_df["Beginning (UTC Time)"].round(0).astype("int64")
+            )
+
         trajectory_df["IAU (code)"] = trajectory_df["IAU (code)"].astype("unicode")
         trajectory_df["Schema (version)"] = trajectory_df["Schema (version)"].astype(
             "unicode"

src/gmn_python_api/trajectory_summary_schema.py

@@ -15,7 +15,7 @@
 
 
 SCHEMA_VERSION = "2.0"
-""""""
+"""The supported trajectory summary data format version."""
 
 _MODEL_TRAJECTORY_SUMMARY_FILE_PATH = os.path.join(
     os.path.dirname(__file__),
@@ -25,12 +25,12 @@
 """Model v2.0 trajectory summary file."""
 
 _AVRO_PATH = os.path.join(tempfile.gettempdir(), "trajectory_summary.avro")
-""""""
+"""The path for the temporary Avro file used to create the .avsc schema"""
 
 _AVSC_PATH = os.path.join(
     tempfile.gettempdir(), f"trajectory_summary_schema_{SCHEMA_VERSION}.avsc"
 )
-""""""
+"""The path for the temporary Avro schema file that will be returned as a string."""
 
 
 def get_trajectory_summary_avro_schema() -> Dict[str, Dict[str, Any]]:
@@ -39,7 +39,9 @@ def get_trajectory_summary_avro_schema() -> Dict[str, Dict[str, Any]]:
     :return: The Avro schema in .avsc format.
     """
     data_frame = gmn_python_api.read_trajectory_summary_as_dataframe(  # type: ignore
-        _MODEL_TRAJECTORY_SUMMARY_FILE_PATH, avro_compatible=True
+        _MODEL_TRAJECTORY_SUMMARY_FILE_PATH,
+        avro_compatible=True,
+        avro_long_beginning_utc_time=False,
     )
 
     pdx.to_avro(_AVRO_PATH, data_frame)