datalakefileclient.upload_data without overwrite

[kasuteru]

It seems that I cannot use upload_data without setting the overwrite flag to true. Maybe I am doing something wrong:

file1 = my_file_system_client.get_file_client("path/to/NONEXISTING_file.txt")
file1.upload_data(data) # Trying to upload to a non-existing file
# Raises: azure.core.exceptions.ResourceNotFoundError: (PathNotFound) The specified path does not exist.

file1 = my_file_system_client.get_file_client("path/to/NONEXISTING_file.txt")
file1.upload_data(data, overwrite=True) # Trying to upload to a non-existing file, with overwrite
# Works!

file1 = my_file_system_client.get_file_client("path/to/existing.txt")
file1.upload_data(data) # Trying to upload to a existing file, without overwrite
# Raises: azure.storage.filedatalake._generated.models._models_py3.StorageErrorException: Operation returned an invalid status 'The condition specified using HTTP conditional header(s) is not met.'

file1 = my_file_system_client.get_file_client("path/to/existing.txt")
file1.upload_data(data, overwrite=True) # Trying to upload to a existing file, with overwrite
# Works!
 

If this is the case for everyone, I am questioning the use of the overwrite flag... or at least its default value of False?