John Haiducek

Also, there is an unrelated failure in the TestMultispecies function of test_pybats; I've been bypassing that test for now.

The python 3 failures occur in test_pybats. However, there is no test for the 'u' prefixes that show up in python 2.x; I've just been doing print statements to check for those.

'str is not bytes' did not cause an exception directly; the exceptions I got occurred elsewhere and seemed to indicate that the string was not being decoded.

I tried that and got exceptions in python 3.6. Maybe there is some subtlety I missed, but this version at least works.

Decode strings in BATS IDL files as utf-8, but only if the python version is >= 3

Change string encoding in SWMF IDL files from utf-8 to ascii to reduce verbosity when things are printed

Dan saw a lot of strings prefixed with 'u' in the output from his scripts and found it distracting. High characters in one of these files is something we've yet to encounter, and probably would only result from some other error (like trying to process binary data as ascii).

Change string encoding in SWMF IDL files from utf-8 to ascii to reduce verbosity when things are printed