Re: [PATCH] iotests: fix source directory location

From: John Snow <jsnow@redhat.com>
To: Kevin Wolf <kwolf@redhat.com>
Cc: "Daniel P. Berrangé" <berrange@redhat.com>,
	qemu-devel <qemu-devel@nongnu.org>,
	Qemu-block <qemu-block@nongnu.org>,
	"Hanna Reitz" <hreitz@redhat.com>,
	"Paolo Bonzini" <pbonzini@redhat.com>
Subject: Re: [PATCH] iotests: fix source directory location
Date: Mon, 11 Jul 2022 17:30:01 -0400	[thread overview]
Message-ID: <CAFn=p-Z0pqYhGXkrTPq9W8FePxbycKD8wAVztJNatxJjWTarhw@mail.gmail.com> (raw)
In-Reply-To: <YpD8aeAUxXokKy4n@redhat.com>

[-- Attachment #1: Type: text/plain, Size: 2773 bytes --]

On Fri, May 27, 2022, 12:29 PM Kevin Wolf <kwolf@redhat.com> wrote:

> Am 26.05.2022 um 16:21 hat John Snow geschrieben:
> > On Thu, May 26, 2022, 3:54 AM Daniel P. Berrangé <berrange@redhat.com>
> > wrote:
> >
> > > On Wed, May 25, 2022 at 08:25:12PM -0400, John Snow wrote:
> > > > If you invoke the check script from outside of the tests/qemu-iotests
> > > > directory, the directories initialized as source_iotests and
> > > > build_iotests will be incorrect.
> > > >
> > > > We can use the location of the source file itself to be more
> accurate.
> > > >
> > > > Signed-off-by: John Snow <jsnow@redhat.com>
> > > > Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
> > > > ---
> > > >  tests/qemu-iotests/testenv.py | 4 ++--
> > > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/tests/qemu-iotests/testenv.py
> > > b/tests/qemu-iotests/testenv.py
> > > > index a864c74b123..9b0f01e84db 100644
> > > > --- a/tests/qemu-iotests/testenv.py
> > > > +++ b/tests/qemu-iotests/testenv.py
> > > > @@ -217,10 +217,10 @@ def __init__(self, imgfmt: str, imgproto: str,
> > > aiomode: str,
> > > >              self.build_iotests =
> > > os.path.dirname(os.path.abspath(sys.argv[0]))
> > > >          else:
> > > >              # called from the source tree
> > > > -            self.source_iotests = os.getcwd()
> > > > +            self.source_iotests = str(Path(__file__,
> '..').resolve())
> > >
> > > Path(__file__).parent
> > >
> > > >              self.build_iotests = self.source_iotests
> > > >
> > > > -        self.build_root = os.path.join(self.build_iotests, '..',
> '..')
> > > > +        self.build_root = str(Path(self.build_iotests,
> > > '../..').resolve())
> > >
> > > Path(self.build_iotests).parent.parent
> > >
> > > to be portable
> > >
> >
> > With windows? I think Path() is meant to be a fully portable class as-is,
> > but I'll double-check my assumption. I use ".." elsewhere in code already
> > checked in, so if it's a problem I ought to fix it everywhere.
>
> I don't see any potential problem with the second hunk because we're
> dealing with the path of a directory there, but "regular_file.py/.."
> looks a bit fishy to me and doesn't work if you ask the kernel. Is this
> guaranteed to work in Python or is it an implementation detail of Path
> that may change?
>

... I apparently never hit send on this draft reply:

Good question, I don't know. I just know that when starting from __file__,
it seems to work in that manner. I used that trick when I added the
PYTHONPATH stuff directly into testenv a while back and never thought more
of it.

If it makes people uneasy to look at, I can just use .parent for the
Principle of Least Surprise.

[-- Attachment #2: Type: text/html, Size: 4263 bytes --]