improve export of test results to MS-Excel

data_analysis/02-3_oracle_workflow_test.py

@@ -14,6 +14,7 @@ from pprint import pprint
 
 import dopt_basics.datetime as dt
 import oracledb
+import pandas as pd
 import polars as pl
 import polars.selectors as cs
 import sqlalchemy as sql
@@ -1289,6 +1290,9 @@ def generate_test_result_data(
 
     test_results = test_results.sort(by=["WF_ID", "BEDP_MAN"], descending=False)
     test_results = test_results.select(pl.int_range(1, pl.len() + 1).alias("Index"), pl.all())
+    test_results = test_results.with_columns(
+        pl.lit(None, dtype=pl.String).alias("Anmerkungen/Feedback")
+    )
 
     return test_results
 
@@ -1296,25 +1300,61 @@ def generate_test_result_data(
 # %%
 test_results = generate_test_result_data(raw_data, pipe_res)
 test_results.head()
+
+
 # %%
-date_str = datetime.datetime.now().strftime("%Y-%m-%d")
-p_save = Path.cwd() / f"Testdatensatz_WF-100-200_{date_str}.xlsx"
-test_results.to_pandas().set_index("Index").to_excel(
-    p_save,
-    freeze_panes=(1, 1),
-    sheet_name=f"Ergebnisse_Testphase_{date_str}",
-)
+def write_test_results_excel(
+    data: pl.DataFrame,
+    base_filename: str,
+) -> None:
+    date_str = datetime.datetime.now().strftime("%Y-%m-%d")
+    p_save = Path.cwd() / f"{base_filename}_{date_str}.xlsx"
+    pd_df = data.to_pandas().set_index("Index")
+
+    with pd.ExcelWriter(p_save, engine="xlsxwriter") as writer:
+        sheet_name = f"Ergebnisse_Testphase_{date_str}"
+        pd_df.to_excel(
+            writer,
+            freeze_panes=(1, 1),
+            sheet_name=sheet_name,
+        )
+        worksheet = writer.sheets[sheet_name]
+
+        rows, cols = pd_df.shape
+        columns = ["Index"] + pd_df.columns.to_list()
+        worksheet.add_table(
+            0,
+            0,
+            rows,
+            cols,
+            {"columns": [{"header": c} for c in columns], "style": "Table Style Light 9"},
+        )
+        for i, col in enumerate(columns):
+            if i == 0:
+                worksheet.set_column(
+                    i, i, max(pd_df.index.astype(str).map(len).max(), len(col)) + 2
+                )
+                continue
+            worksheet.set_column(
+                i, i, max(pd_df[col].astype(str).map(len).max(), len(col)) + 2
+            )
+
+
+# %%
+write_test_results_excel(test_results, "Testdatensatz_WF-100-200")
+
 #####################################################################
 # %%
 # ** deviating titles where BEDP_MENGE_BEDARF_VM > MENGE_VORMERKER
 deviation_vm = test_results.with_columns(pl.col.MENGE_VORMERKER.fill_null(0)).filter(
     pl.col.BEDP_MENGE_BEDARF_VM > pl.col.MENGE_VORMERKER
 )
-deviation_vm = test_results.filter(pl.col.BEDP_TITELNR.is_in(dev["BEDP_TITELNR"].implode()))
-
-date_str = datetime.datetime.now().strftime("%Y-%m-%d")
-p_save = Path.cwd() / f"Abweichungen-VM_{date_str}.xlsx"
-deviation_vm.to_pandas().set_index("Index").to_excel(p_save, freeze_panes=(1, 1))
+deviation_vm = test_results.filter(
+    pl.col.BEDP_TITELNR.is_in(deviation_vm["BEDP_TITELNR"].implode())
+)
+deviation_vm
+# %%
+write_test_results_excel(deviation_vm, "Abweichungen-VM")
 # ** WF-200 potentially triggered
 raw_data.filter(pl.col.MELDENUMMER.is_in((17, 18))).filter(
     pl.col.BEDP_TITELNR.is_duplicated()