Text encoding breaks rpi-source

[melissaboiko]

I'm getting this on a freshly installed script per README:

 *** Linux source commit: df4d584e66a660195422ca2b7ce1d68d1181f2a2
Traceback (most recent call last):
  File "/usr/local/bin/rpi-source", line 366, in <module>
    check_diskspace(args.dest)
  File "/usr/local/bin/rpi-source", line 178, in check_diskspace
    df = sh_out("df %s" % dir)
         ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/bin/rpi-source", line 101, in sh_out
    return out.decode('ascii')
           ^^^^^^^^^^^^^^^^^^^
UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 20: ordinal not in range(128)

If I experimentally change the line from 'ascii' to 'utf-8' it works, so it's probably a matter of properly setting up the console to the current running locale.

[pelwell]

Out of interest, what might the value of dir have been?

[pelwell]

Since UTF-8 is backwards compatible with ASCII for values 0-127, I can't think of a reason not to change all occurrences of ascii to utf-8. What do you think, @popcornmix?

[popcornmix]

I believe utf-8 is the default for strings with python 3, so yes, I don't think we should use ascii without a good reason.

[pelwell]

Fixed by #21.