OpenPyXL - Formatted date columns set themselves as 'Custom' in output Excel file

I need to insert a dataframe into an Excel file, with every date column in format DD/MM/YYYY.

I use the dataframe.to_excel function to send the dataframe into the output file, using an ExcelWriter with correctly set date_format and datetime_format parameters.

The format is correctly inserted into the excel file for datetime objects, but my output file decides the object is Custom and puts it in YYYY-MM-DD format.

Oddly enough, when I edit the corresponding cells, they format correctly, and they stay as such when set to be Short Date objects. How can I make this the default behavior, so that it's the first thing I see when opening the file, without having to set the column as a Short Date ?

Please note that I need this to work with columns, not cells. I handle big chunks of data using this program, and iterating through each cell of each column that is of type datetime is not desired.

Here is a very simplified version of my code.

from pandas import DataFrame, ExcelWriter, to_datetime
from pandas._libs.tslibs.timestamps import Timestamp

def Send_Dataframe_To_Excel(dataframe):
		columns_of_type_datetime = list(dataframe.select_dtypes(include=["datetime", "datetime64", "datetime64[ns]", "datetimetz"]).columns)
			#find all columns of type datetime
			
		for column in columns_of_type_datetime :
			dataframe[column] = to_datetime(dataframe[column], format='%Y-%m-%d %H:%M:%S', errors='coerce').dt.date
				#convert them to pandas datetime
				
		with ExcelWriter("output_file.xlsx", engine="openpyxl", date_format='DD/MM/YYYY', datetime_format='DD/MM/YYYY') as writer:
			dataframe.to_excel(writer, sheet_name="Sheet1", index=False)

dataframe = DataFrame({'ID': [1, 2, 3, 4, 5, 6, 7],
			'text_data': ['text1', 'text2', 'text3', 'text4', 'text5', 'text6', 'text7'],
			'number_data': [11, 12, 13, 14, 15, 16, 17,],
			'date_data': [Timestamp('2011-01-01 00:20:00'), Timestamp('2012-02-02 00:00:00'), Timestamp('2013-03-03 00:00:00'), Timestamp('2014-04-04 00:00:00'), Timestamp('2015-05-05 00:00:00'), Timestamp('2016-06-06 00:00:00'), Timestamp('2017-07-07 00:00:00')]})
			
			
Send_Dataframe_To_Excel(dataframe)

Here is what I expected in the output file :

Here is what the output file actually ends up looking like (as you can see, I can set the cell type to Short Date permanently, "fixing" the problem, but it's not the default behavior. Also, "Personnalisé" means "Custom" and "Date courte" means "Short date"):

That's an open issue (see, GH44284).

For now, you need to switch (if possible) to the default engine (which is xlxwriter if installed) :

with ExcelWriter(
    'output_file.xlsx',
    engine='xlsxwriter', # <- here
    date_format='DD/MM/YYYY',
    datetime_format='DD/MM/YYYY') \
as writer:
    dataframe.to_excel(writer, sheet_name="Sheet1", index=False)

Output :