read_stata failing with URL

[jseabold]

I'm not entirely sure the root cause yet. It's something to do with the default encoding I think. I'm pretty sure Stata uses iso-8859-1 not cp1252. Changing this inside pandas.io.maybe_read_encoded_stream works, but changing the encoding in StataReader breaks something else that I don't have time to debug.

Can someone sanity check me. Does this work on Python 2?

dta = pd.read_stata("http://www.stata-press.com/data/r13/fullauto.dta")

[TomAugspurger]

I get a

UnicodeDecodeError: 'charmap' codec can't decode byte
0x9d in position 2644: character maps to <undefined>

in python 2.

[jreback]

the default encoding is passed (even to a url), so prob should be a bit more inference here.

[bashtage]

Is there someone permanent to place a dta file at a URL for testing?

Or is there a reasonable method to mock a URL locally using a file?

[bashtage]

Fixed, waiting for final green.