On 3/30/25 5:29 PM, Frode Nordahl wrote:
> flake8 2.7.0 introduced the F824 check [0].
> 
> While the check sounds reasonable on the surface, we have code in
> the repository where the global variable is used only for access.
> 
> My conclusion from evaluating the alternative of adjusting the code
> to appease the check is that it would invite future mistakes where
> the global variable is unintentionally shadowed in addition to
> making the code less clear and explicit.
> 
> 0: https://github.com/PyCQA/flake8/pull/1974
> 1: https://github.com/PyCQA/pyflakes/pull/825
> Signed-off-by: Frode Nordahl <[email protected]>
> ---
>  Makefile.am | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/Makefile.am b/Makefile.am
> index a61a1cadf..cdcc4e9bd 100644
> --- a/Makefile.am
> +++ b/Makefile.am
> @@ -406,9 +406,10 @@ ALL_LOCAL += flake8-check
>  #   W504 line break after binary operator
>  # F*** -- warnings native to flake8
>  #   F811 redefinition of unused <name> from line <N> (only from flake8 v2.0)
> +#   F824 a `global` or `nonlocal` statement where the name is never 
> reassigned
>  # D*** -- warnings from flake8-docstrings plugin
>  # H*** -- warnings from flake8 hacking plugin (custom style checks beyond 
> PEP8)
> -FLAKE8_IGNORE = 
> E121,E123,E125,E126,E127,E128,E129,E131,E203,E722,W503,W504,F811,D,H,I
> +FLAKE8_IGNORE = 
> E121,E123,E125,E126,E127,E128,E129,E131,E203,E722,W503,W504,F811,F824,D,H,I
>  flake8-check: $(FLAKE8_PYFILES)
>       $(FLAKE8_WERROR)$(AM_V_GEN) \
>         src='$^' && \

Thanks, Frode!

I re-worded the subject line a bit to make it shorter and applied
the change to all supported branches down to 3.2.  That should
unbreak our CI.

Best regards, Ilya Maximets.
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Reply via email to