fix case when columns are in different order

jobs/generic/compare_pandas_job.py

@@ -16,8 +16,9 @@ def transform(self, tableA, tableB):
             print('The deltas in columns are: ', diff_columns)
             print('The columns in common are: ', common_columns)
             print('Rest of the comparison will be based on columns in common')
-            tableA = tableA[common_columns]
-            tableB = tableB[common_columns]
+
+        tableA = tableA[common_columns]  # applied even if columns are all the same to make sure they are in the same order (impt for checks below)
+        tableB = tableB[common_columns]
 
         # Comparing datasets length
         diff_row_count = len(tableA) != len(tableB)
@@ -52,7 +53,6 @@ def transform(self, tableA, tableB):
         df_out = compare_dfs(tableA, pks1, compare1, tableB, pks2, compare2, strip, filter_deltas, threshold=0.01)
         print('Finishing compare, column by column.')
 
-        # import ipdb; ipdb.set_trace()
         return df_out
 
 

yaetos/libs/analysis_toolkit/query_helper.py

@@ -191,7 +191,6 @@ def check_delta(row):
         if item1 == item2:  # check necessary for next step. See comment below.
             item1 = item1 + '_1'
             item2 = item2 + '_2'
-            # raise Exception(f'Error. item1: {item1}, item2: {item2}')
         is_numeric_1 = pd.api.types.is_numeric_dtype(df_joined[item1])
         is_numeric_2 = pd.api.types.is_numeric_dtype(df_joined[item2])
         is_str_1 = pd.api.types.is_string_dtype(df_joined[item1])