BUG: Fix bug in pd.read_clipboard on windows with PY3 (GH4561); not decoding properly

[jreback]

closes #4561

[cpcloud]

should this maybe be get_option('display.encoding')?

[cpcloud]

you could even do

encoding = kwargs.get('encoding', get_option('display.encoding'))

[jreback]

ok

[jreback]

this is a windows encoding, could use sys.getsystemencoding()?

on side note:
cc @jtratner

in pandas/compat.py

    def bytes_to_str(b, encoding='utf-8'):
        return b.decode(encoding)

shouldn't this be:

    def bytes_to_str(b, encoding=None):
        if encoding is None:
             encoding='utf-8'
        return b.decode(encoding)

[cpcloud]

why? strings are immutable so even if it's appended to in the function it won't be seen in the caller

[cpcloud]

similar to if you increment an int parameter

def f(x):
    x += 1
y = 1
f(y)
print(y) # prints 1

[cpcloud]

also i think it's sys.getfilesystemencoding()

[jtratner]

@jreback how about:

    def bytes_to_str(b, encoding='utf-8'):
        return b.decode(encoding or 'utf-8')

For the other one, you can just wrap in an or instead - clearer and doesn't execute get_option if not needed

encoding = kwargs.get('encoding') or get_option('display.encoding')

[cpcloud]

What's with all of you green icon folk? 😄

[jreback]

they must have changed the default, I don't have a pic up

[jreback]

https://travis-ci.org/jreback/pandas/jobs/10216546

I guess pyperclip acts differently on windows and linux under py3.......

[cpcloud]

it's annoying that all string types in python2 have decode and encode methods while only bytes types in python3 have decode and str types have only encode

[jreback]

you could monkey-patch strings at runtime to make it a noop! but pretty I guess the error means you are not doing something right