On Sat, Jan 25, 2025 at 03:56:02PM -0700, Simon Glass wrote:
Hi Tom,
On Sat, 25 Jan 2025 at 15:50, Tom Rini trini@konsulko.com wrote:
On Sat, Jan 25, 2025 at 03:47:26PM -0700, Simon Glass wrote:
Hi Tom,
On Sat, 25 Jan 2025 at 14:44, Tom Rini trini@konsulko.com wrote:
On Sat, Jan 25, 2025 at 02:31:39PM -0700, Simon Glass wrote:
This fails on samus_tpl as there is no 'net' command.
=> net list Unknown command 'net' - try 'help' !
Fix it by adding a condition for the test.
For this part, Reviewed-by: Tom Rini trini@konsulko.com
Add a blank line to keep pylint happy.
Can we silence that pylint? I don't look forward to the churn of adding more blanks everywhere especially since imo it doesn't improve readability either.
Yes, but we would want to do it globally. But it isn't really a good idea IMO. We are trying to use PEP8 which loads of tools expect...e.g. my IDE shows warnings in this case. It might be better to just fix the problems.
As to readability, I'm on the fence...in practice it doesn't matter much as few non-trivial programs have a lot of top-level functions. I would rather keep the tool happy and use the Python style that people expect when coming into the project.
Yes, we should do it globally. Can we not globally also tell tools the deviations from PEP8? I remember back when people would argue about if we should tell file editors what to expect for whitespace, etc :)
I'm sure we can do anything. But why deviate? Is it just that you want to avoid churn, or do you actually not like this part of PEP8?
I think it looks bad (personal preference) and we have that in almost none of our tests I suspect so adding blank lines to make PEP8 happy rather than adding hopefully a line to a common file normal tooling reads seems the wrong direction. But I don't care enough to not apply it if someone else does it. I'll add missing blank lines when needed assuming the pylint CI job complains once it's done.