Unicode error in printer.py about honcho HOT 4 CLOSED

I had the same issue. Setting LANG=en_US.UTF-8 as mentioned in #51 didn't fix the problem.
I could fix it by setting the environment variable PYTHONIOENCODING=utf8.

This problem seems to appear if the output of a python script is piped into another program.
(http://stackoverflow.com/questions/492483/setting-the-correct-encoding-when-piping-stdout-in-python)
Without knowing too much about the internals of heroku, I guess they pipe stdout/stderr to their log mechanism, which probably leads to this error.

from honcho.

Interesting. I'd really like to see some tests added which use some test UTF-8 files -- both valid and invalid, to replicate these bugs so that we can fix them. See here for some good test files.

from honcho.

@nickstenning I unfortunately won't have time to work on this.

from honcho.

So I'm still not entirely clear on what the "best" solution to this is. In Python 2 a sensible option is to only write bytes (AKA str) to sys.stdout, thus bypassing any encoding issues. In Python 3 one can write bytes to sys.stdout.buffer, but this all seems like avoiding the issue.

As of d07ac1e, Honcho ensures that all data it receives from processes is normalised to UTF-8 (invalid UTF-8 is replaced with U+FFFD). It then tries to write only unicode (or str in Python 3) to sys.stdout.

I think this should be fixed, but I have a feeling I've not heard the end of these issues.

from honcho.