BUG: pd.to_datetime fails to identify actual date format about pandas HOT 4 CLOSED

Thanks for the report! This is a duplicate of #58328

But, your analysis is correct and if you have an idea for a fix feel free to submit a PR!

from pandas.

Not sure if it is right, but my idea for the fix would be as below

1. checking if the first element has 'May' in the date string
2. if so then get the date format as usual. Suppose the date format we got is "%B-%d-%Y"
3. then try converting all dates with "%B-%d-%Y" and "%b-%d-%Y" (replace capital B with small b) . If one of them doesn't work, continue with the other format.

@Aloqeely what do you think of this approach?

Edit:
step 3 could be like below

format = get_format(data[0])  # "%B-%d-%Y"
formats_to_try = [format, format.replace('%B', '%b')]

for fmt in formats_to_try:
    try:
        return pd.to_datetime(*same_args, format=fmt)
    except ValueError:
        continue

raise ValueError("Mixed date formats")

from pandas.

That would work, but 'May' is not the only month name where this happens, for example 'Mai' in French has the same problem.

Could you please discuss in the original issue and get input from pandas devs?

from pandas.

Sure closing this issue

from pandas.