This is a demonstration of a bug in the python standard library.
To run it yourself:
$ pip install argparse_min_example
$ argparse-min-example --help
You should get output something like the following. As you can see, python calculates the width of 'བོད་སྐད་ལ་' incorrectly, because of the vowel markers.
usage: argparse-min-example [-h] [--language1 བོད་སྐད་ལ་]
[--language2 LANGUAGE]
Process some integers.
optional arguments:
-h, --help show this help message and exit
--language1 བོད་སྐད་ལ་
Lanugage for output
--language2 LANGUAGE Lanugage for output