On Fri, 2013-12-06 at 13:23 -0500, Anne Mulhern wrote:
----- Original Message -----
> From: "David Lehman" <dlehman(a)redhat.com>
> To: "anaconda patch review"
<anaconda-patches(a)lists.fedorahosted.org>
> Sent: Friday, December 6, 2013 1:11:01 PM
> Subject: Re: [blivet 04/13] StorageDevice.minSize returns 0 if format's minSize
is 0.
>
> On Fri, 2013-12-06 at 12:54 -0500, Anne Mulhern wrote:
> >
> >
> >
> > ----- Original Message -----
> > > From: "David Lehman" <dlehman(a)redhat.com>
> > > To: "anaconda patch review"
<anaconda-patches(a)lists.fedorahosted.org>
> > > Sent: Friday, December 6, 2013 12:06:29 PM
> > > Subject: Re: [blivet 04/13] StorageDevice.minSize returns 0 if
format's
> > > minSize is 0.
> > >
> > > On Fri, 2013-12-06 at 11:50 -0500, mulhern wrote:
> > > > Previously, if format's minSize was 0, then it returned
self.size
> > > > instead.
> > > > I think this was an error, and the intention was to fall back on
> > > > self.size
> > > > only if format.minSize was not set.
> > >
> > > Many formats return 0 for minSize, meaning there is no known minimum
> > > size. It should be treated the same as None, since no formatting
> > > metadata can possibly exist without some space to store it in.
> > >
> > > David
> > >
> >
> > How about having those formats return None? I don't know if this is good
or
> > bad...but most arith operators, except +, -, etc. are defined on None, so
> > min(None, 1) is None and so forth.
>
> I'd be concerned about fallout from such a change, nor do I see any
> value in changing that now.
>
> If flags.installer_mode is False we do not automatically call
> self.updateSizeInfo() from the FS constructor. Because of that, all of
> the minSize methods in fs.py need to add a check for
> self._minInstanceSize of None and instead return self._minSize if that
> is the case. Once this is done you will never get a value of None from a
> format's minSize attribute. This is my fault -- I did not take this into
> account when I made the change to not update size info outside of
> anaconda.
>
> David
OK. In working copy, I
* remove the code changes
* improve the comment change
* and add a comment to minSize and maxSize property in DeviceFormat indicating that 0
means unknown.
Does that sound good?
Yes. I'll post a patch at some point to ensure we're not giving minSize
a value of None.
David
- mulhern
>
> >
> > - mulhern
> >
> > >
> > > _______________________________________________
> > > anaconda-patches mailing list
> > > anaconda-patches(a)lists.fedorahosted.org
> > >
https://lists.fedorahosted.org/mailman/listinfo/anaconda-patches
> > >
> > _______________________________________________
> > anaconda-patches mailing list
> > anaconda-patches(a)lists.fedorahosted.org
> >
https://lists.fedorahosted.org/mailman/listinfo/anaconda-patches
>
>
> _______________________________________________
> anaconda-patches mailing list
> anaconda-patches(a)lists.fedorahosted.org
>
https://lists.fedorahosted.org/mailman/listinfo/anaconda-patches
>
_______________________________________________
anaconda-patches mailing list
anaconda-patches(a)lists.fedorahosted.org
https://lists.fedorahosted.org/mailman/listinfo/anaconda-patches