Refactor util in back-end

backend_py/primary/primary/services/summary_delta_vectors.py

@@ -4,8 +4,7 @@
 import pyarrow.compute as pc
 import numpy as np
 
-
-from primary.services.service_exceptions import InvalidDataError, Service
+from .utils.summary_vector_table_helpers import validate_summary_vector_table
 
 
 @dataclass
@@ -17,18 +16,6 @@ class RealizationDeltaVector:
     unit: str
 
 
-def _is_valid_vector_table(vector_table: pa.Table, vector_name: str) -> bool:
-    """
-    Check if the vector table is valid.
-
-    Expect the table to contain the following columns: DATE, REAL, vector_name.
-    """
-    expected_columns = {"DATE", "REAL", vector_name}
-    if set(vector_table.column_names) != expected_columns:
-        unexpected_columns = set(vector_table.column_names) - expected_columns
-        raise InvalidDataError(f"Unexpected columns in table {unexpected_columns}", Service.GENERAL)
-
-
 def create_delta_vector_table(
     first_vector_table: pa.Table, second_vector_table: pa.Table, vector_name: str
 ) -> pa.Table:
@@ -42,8 +29,8 @@ def create_delta_vector_table(
 
     `Note`: Pre-processing of DATE-columns, e.g. resampling, should be done before calling this function.
     """
-    _is_valid_vector_table(first_vector_table, vector_name)
-    _is_valid_vector_table(second_vector_table, vector_name)
+    validate_summary_vector_table(first_vector_table, vector_name)
+    validate_summary_vector_table(second_vector_table, vector_name)
 
     joined_vector_table = first_vector_table.join(
         second_vector_table, keys=["DATE", "REAL"], join_type="inner", right_suffix="_second"
@@ -52,7 +39,6 @@ def create_delta_vector_table(
         joined_vector_table.column(vector_name), joined_vector_table.column(f"{vector_name}_second")
     )
 
-    # TODO: Should a schema be defined for the delta vector?
     delta_table = pa.table(
         {
             "DATE": joined_vector_table.column("DATE"),
@@ -70,7 +56,7 @@ def create_realization_delta_vector_list(
     """
     Create a list of RealizationDeltaVector from the delta vector table.
     """
-    _is_valid_vector_table(delta_vector_table, vector_name)
+    validate_summary_vector_table(delta_vector_table, vector_name)
 
     real_arr_np = delta_vector_table.column("REAL").to_numpy()
     unique_reals, first_occurrence_idx, real_counts = np.unique(real_arr_np, return_index=True, return_counts=True)

backend_py/primary/primary/services/sumo_access/summary_access.py

@@ -11,8 +11,12 @@
 from fmu.sumo.explorer.objects import TableCollection, Table
 from webviz_pkg.core_utils.perf_timer import PerfTimer
 
-from primary.services.utils.arrow_helpers import sort_table_on_real_then_date, is_date_column_monotonically_increasing
-from primary.services.utils.arrow_helpers import find_first_non_increasing_date_pair
+from primary.services.utils.arrow_helpers import (
+    find_first_non_increasing_date_pair,
+    sort_table_on_real_then_date,
+    is_date_column_monotonically_increasing,
+)
+from primary.services.utils.summary_vector_table_helpers import validate_summary_vector_table
 from primary.services.service_exceptions import (
     Service,
     NoDataError,
@@ -160,10 +164,7 @@ async def get_vector_async(
         table, vector_metadata = await self.get_vector_table_async(vector_name, resampling_frequency, realizations)
 
         # Verify that columns are as we expect
-        expected_columns = {"DATE", "REAL", vector_name}
-        if set(table.column_names) != expected_columns:
-            unexpected_columns = set(table.column_names) - expected_columns
-            raise InvalidDataError(f"Unexpected columns in table {unexpected_columns}", Service.SUMO)
+        validate_summary_vector_table(table, vector_name, Service.SUMO)
 
         real_arr_np = table.column("REAL").to_numpy()
         unique_reals, first_occurrence_idx, real_counts = np.unique(real_arr_np, return_index=True, return_counts=True)

backend_py/primary/primary/services/utils/summary_vector_table_helpers.py

@@ -0,0 +1,17 @@
+import pyarrow as pa
+
+from primary.services.service_exceptions import InvalidDataError, Service
+
+
+def validate_summary_vector_table(vector_table: pa.Table, vector_name: str, service: Service = Service.GENERAL) -> None:
+    """
+    Check if the vector table is valid - single vector table should contain columns DATE, REAL, vector_name.
+
+    Expect the pyarrow table to contain the following columns: DATE, REAL, vector_name.
+
+    Raises InvalidDataError if the table does not contain the expected columns.
+    """
+    expected_columns = {"DATE", "REAL", vector_name}
+    if set(vector_table.column_names) != expected_columns:
+        unexpected_columns = set(vector_table.column_names) - expected_columns
+        raise InvalidDataError(f"Unexpected columns in table {unexpected_columns}", service)