python - Why does pandas read_csv not support multiple comments (#,@,...)? -

i found pandas read_csv method faster numpy loadtxt. unfortunatly find myself in situation have go numpy because loadtxt has option of setting comments=['#','@']. pandas read_csv method can take 1 comment string comment='#' far can tell site. suggestions or workarounds make life easier , make me not pivot numpy? why pandas not support multiple comment indicators?

# save in test.dat @ bla # bla 1 2 3 4 

minimal example:

# work, 1 type of comment accounted df = pd.read_csv('test.dat', index_col=0, header=none, comment='#')  # not work (not suprising reading help) df = pd.read_csv('test.dat', index_col=0, header=none, comment=['#','@'])  # work slow df = np.loadtxt('test.dat', comments=['#','@']) 

the short answer nobody has implemented in pandas yet. looking through github issues, looks else has suggested , maintainers open patch implements it: https://github.com/pandas-dev/pandas/issues/13948

could opportunity contribute pandas project if feel comfortable that, or keep eye on issue if else it. part of codebase handles comments looks around here in _check_comments: https://github.com/pandas-dev/pandas/blob/master/pandas/io/parsers.py#l2348