On Mon, Mar 22, 2021 at 10:29:32AM +0100, Andrea Bolognani wrote: > On Mon, 2021-03-22 at 07:50 +0100, Erik Skultety wrote: > > On Fri, Mar 19, 2021 at 06:39:31PM +0100, Andrea Bolognani wrote: > > > sc_flake8: > > > @if [ -n "$(FLAKE8)" ]; then \ > > > - $(VC_LIST_EXCEPT) | $(GREP) '\.py$$' | xargs \ > > > - $(FLAKE8) --ignore $(FLAKE8_IGNORE) --show-source; \ > > > + DOT_PY=$$($(VC_LIST_EXCEPT) | $(GREP) '\.py$$'); \ > > > + BANG_PY=$$($(VC_LIST_EXCEPT) | xargs grep -l '^#!/usr/bin/env > > > python3$$'); \ > > > + ALL_PY=$$(printf "%s\n%s" "$$DOT_PY" "$$BANG_PY" | sort -u); \ > > > > Not that I'd be against ^this, but I think it might be worth (even for > > consistency reasons) to mandate that all Python scripts to use the '.py' > > extension explicitly instead and rename the ones that violate this. To > > support > > my argument, there are 34 scripts that use a suffix and 2 (one of which is > > the > > latest CI helper) that don't. > > It's a time-honored tradition to omit the suffix for scripts which > are called directly by the user, which is why you don't install > packages using dnf.py or build software using meson.py :)
Fair enough: Reviewed-by: Erik Skultety <eskul...@redhat.com>
