Test handling of duplicate defaultvalues

Author: dafeda

src/semeio/fmudesign/_excel2dict.py

@@ -3,6 +3,7 @@
 read by semeio.fmudesign.DesignMatrix.generate
 """
 
+import logging
 from collections import OrderedDict
 from collections.abc import Mapping
 from typing import Any
@@ -12,6 +13,8 @@
 import pandas as pd
 import yaml
 
+logger = logging.getLogger(__name__)
+
 
 def excel2dict_design(
     input_filename: str, sheetnames: Mapping[str, Any] | None = None
@@ -382,34 +385,34 @@ def _read_defaultvalues(filename: str, sheetname: str) -> OrderedDict[str, Any]:
     Returns:
         OrderedDict with defaultvalues (parameter, value)
     """
-    default_dict: OrderedDict[str, Any] = OrderedDict()
     default_df = pd.read_excel(
         filename, sheetname, header=0, index_col=0, engine="openpyxl"
     )
+
     default_df.dropna(axis=0, how="all", inplace=True)
     default_df = default_df.loc[
         :, ~default_df.columns.astype(str).str.contains("^Unnamed")
     ]
 
-    # Strip spaces before and after parameter names, if they are there
-    # it is probably invisible user errors in Excel.
-
-    default_df.index = pd.Index(
-        [
-            paramname.strip() if isinstance(paramname, str) else paramname
-            for paramname in default_df.index
-        ]
-    )
-    for row in default_df.itertuples():
-        if str(row[0]) in default_dict:
-            print(
-                f"WARNING: The default value '{row[0]}' "
-                f"is listed twice in the sheet '{sheetname}'. "
-                "Only the first entry will be used in output file"
+    if default_df.empty:
+        return OrderedDict()
+
+    # Strip leading/trailing spaces from parameter names such that
+    # for example "  PARAM" and "PARAM" are treated as duplicates.
+    default_df.index = default_df.index.str.strip()
+
+    # Check for duplicates and warn
+    duplicates = default_df.index.duplicated(keep="first")
+    if duplicates.any():
+        duplicate_names = default_df.index[duplicates].unique()
+        for dup_name in duplicate_names:
+            logger.warning(
+                f"The default value '{dup_name}' is listed twice in the sheet "
+                f"'{sheetname}'. Only the first entry will be used in output file"
             )
-        else:
-            default_dict[str(row[0])] = row[1]
-    return default_dict
+
+    default_df = default_df[~duplicates]
+    return OrderedDict(default_df.iloc[:, 0].to_dict())
 
 
 def _read_dependencies(

tests/fmudesign/data/config/design_input_mc_with_correls.xlsx

@@ -1,6 +1,7 @@
 """Testing code for generation of design matrices"""
 
 import json
+import logging
 import shutil
 from datetime import datetime
 from pathlib import Path
@@ -187,33 +188,37 @@ def test_generate_onebyone(tmpdir):
         pytest.fail("Timestamp in Metadata sheet is not in expected format")
 
 
-def test_generate_full_mc_snapshot(snapshot):
+def test_generate_full_mc_snapshot(snapshot, caplog):
     """Test that full monte carlo design matrix generation remains consistent.
-
     This is a snapshot test that verifies the entire output of the design matrix
     generation process, including both the design values and default values.
     """
     # Setup
     inputfile = TESTDATA / "config/design_input_mc_with_correls.xlsx"
-    input_dict = excel2dict_design(inputfile)
-    design = DesignMatrix()
 
-    # Generate the design matrix
-    design.generate(input_dict)
+    with caplog.at_level(logging.WARNING):
+        input_dict = excel2dict_design(inputfile)
+        design = DesignMatrix()
+        # Generate the design matrix
+        design.generate(input_dict)
+
+    # Check that the warning was logged
+    assert (
+        "The default value 'FAULTSEAL' is listed twice in the sheet 'defaultvalues'"
+        in caplog.text
+    )
 
     # Prepare data for snapshot comparison
     snapshot_dict = {
         "designvalues": design.designvalues.to_dict("records"),
         "defaultvalues": dict(design.defaultvalues),
     }
-
     # Serialize to string for snapshot comparison
     snapshot_str = json.dumps(
         snapshot_dict,
         indent=2,
         sort_keys=True,
     )
-
     # Verify against snapshot
     snapshot.assert_match(snapshot_str, "design_output_mc_with_correls.json")