Re: GNU Parallel Bug Reports Unexpected encoding behavior in 20180622

[Manuel Salvadores]

I think I got it. This works.

❯ parallel 'LC_ALL="en_US.UTF-8" python -c "import locale; print(locale.getpreferredencoding())"' ::: {1..2}

UTF-8

UTF-8

I also saw this line in the release "Deal better with multibyte chars by forcing LC_ALL=C." which I imagine is what changed the behavior.

Manuel

On Thu, Jul 5, 2018 at 4:34 PM, Manuel Salvadores <address@hidden> wrote:

Hi,

I have noticed that the last version of parallel is behaving differently in terms of setting up preferred encoding. The logs below show how with version 20180522 I get a UTF-8 encoding and 20180622 defaults to US-ASCII. Is this a bug or expected new behavior? Is there any setting to keep UTF-8 as the preferred encoding?

Thanks!

Manuel

(this is test was done on OSX High Sierra 10.13.5)

venv ❯ python

Python 3.6.5 (default, Jun 17 2018, 12:13:06)

Parallel 20180522

venv ❯ parallel 'python -c "import locale; print(locale.getpreferredencoding())"' ::: {1..2}

UTF-8

UTF-8

Parallel 20180622

venv ❯ parallel 'python -c "import locale; print(locale.getpreferredencoding())"' ::: {1..2}

US-ASCII

US-ASCII