Issue

wr.catalog.sanitize_dataframe_columns_names does not correctly sanitize CammelCase to snake_case. It also allows duplicate columns which would be problematic in Athena.

Example

import pandas as pd import awswrangler as wr print(wr.__version__) # 2.13.0 df = pd.DataFrame({'SomeLongStringWithMultipleWords': [1, 2, 3], 'sinceEpoch': [4, 5, 6], 'SinceEpoch': [7, 8, 9], 'since_epoch': [7, 8, 9]}) df_clean = wr.catalog.sanitize_dataframe_columns_names(df) print(df_clean) # somelongstringwithmultiplewords sinceepoch sinceepoch since_epoch # 0 1 4 7 7 # 1 2 5 8 8 # 2 3 6 9 9 print(df_clean["sinceepoch"]) # sinceepoch sinceepoch # 0 4 7 # 1 5 8 # 2 6 9 

First issue

Based on the documentation for this function, I would have expected each column name to be snake_case.

Second issue

This is more of a judgement call and I could see reasons for not adding it, however, to live up to the function name, if duplicate columns would either throw an error or have a suffix added to them. I'd appreciate it if this was at least optional.