[PATCH 1/2] Convert host checking to pytest.

Gedare Bloom gedare at rtems.org
Mon Mar 16 03:18:21 UTC 2020


On Sun, Mar 15, 2020 at 9:09 PM Gedare Bloom <gedare at rtems.org> wrote:
>
> On Sun, Mar 15, 2020 at 5:54 PM Amar Takhar <amar at rtems.org> wrote:
> >
> > On 2020-03-15 17:26 -0600, Gedare Bloom wrote:
> >
> > > Make blank lines consistent, 1 or 2, between top-level definitions. (I
> > > know, we still haven't settled on a style, but let's aim for
> > > consistency.)
> >
> > Okay.
> >
> >
> > > I think all imports should go at the top?
> >
> > This is scoped it only runs under certain conditions extremely common in Python.
> > It's good practice as well as there's no point in loading a module that isn't
> > used all of the time.  Also, it makes for far clearer code as you know the
> > module is only used for that snippet.
> >
> Alright. I think this may need to be specially defined in style guide.
>
> I believe Google guide makes some restrictions on imports. I'm not
> enough Python expert to know, but I really think with the amount of
> Python code that is expected to come through, we need to decide on
> these things sooner than later.
>

I found what I was thinking about:
http://google.github.io/styleguide/pyguide.html#313-imports-formatting

PEP has similar guidance:
https://www.python.org/dev/peps/pep-0008/#imports

But there is another problem, which is the restriction not to import
specific classes/functions:
http://google.github.io/styleguide/pyguide.html#22-imports

I'm not sure, but I think that also is violated here. PEP doesn't
restrict on this.

As usual exceptions can be made for good reasons, which you both have
given and is in the current practice in RTEMS python code, but we do
need to be aware of these deviations from whatever style guide we end
up using as our baseline. The more deviations, the harder it is to
check code automatically using tools from other projects.

Gedare

> >
> > Amar.
> > _______________________________________________
> > devel mailing list
> > devel at rtems.org
> > http://lists.rtems.org/mailman/listinfo/devel


More information about the devel mailing list