I've been working on getting PyChecker to run on anaconda in a way that the
output is usable, I'm making nice progress here. As a result I've a first patch
fixing various issues which I believe to be real issues.
As we are close to the F-10 preview release, and as I'm not at my best atm
(sick), I don't want to directly apply my fixes. So here is a patch fixing
this. Please review and apply it, I believe it fixes various real issues.
These all look fine to go ahead and commit, with the exception of the
couple comments below. I say go ahead and commit to master since they
all look like relevant potential bugzilla entries to me.
> diff --git a/iw/GroupSelector.py b/iw/GroupSelector.py
> index 4a60d78..b56833a 100644
> --- a/iw/GroupSelector.py
> +++ b/iw/GroupSelector.py
> @@ -97,7 +97,7 @@ def _deselectPackage(ayum, group, pkg):
> except mdErrors.PackageSackError:
> log = logging.getLogger("yum.verbose")
> log.debug("no such package %s from group %s" %(pkg,
> - self.group.groupid))
> + group.groupid))
> if pkgs:
> pkgs = ayum.bestPackagesFromList(pkgs)
> for po in pkgs:
This can just be groupid, not group.groupid. See slightly above.
Can drive be "/dev/whatever" here, or is it just going to be "whatever"?
The call to progressWindow just above makes me wonder what we're
expecting drive to be here. Might want to throw in a check. It'd be
nice if we were at all consistent about /dev/ vs. not in anaconda.
- Chris
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
10-28-2008, 03:18 PM
Jesse Keating
Patch: fix various issues caught by pychecker
On Tue, 2008-10-28 at 11:51 +0100, Hans de Goede wrote:
> - stdoutLog.critical(_("Error processing %%ksappend lines: %s")
> % e)
> + stdoutLog.critical(_("Error processing %%ksappend lines: %s")
> % msg)
I see a bunch of these, which could be a product of not having a project
standard for these try/except blocks. Would it be worthwhile to convert
everything to one standard and keep it that way in the future (as to
avoid silly head mistakes?)
--
Jesse Keating
Fedora -- Freedom˛ is a feature!
identi.ca: http://identi.ca/jkeating
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
10-29-2008, 10:54 AM
Hans de Goede
Patch: fix various issues caught by pychecker
Chris Lumens wrote:
These all look fine to go ahead and commit, with the exception of the
couple comments below. I say go ahead and commit to master since they
all look like relevant potential bugzilla entries to me.
Thanks for the review!
diff --git a/iw/GroupSelector.py b/iw/GroupSelector.py
index 4a60d78..b56833a 100644
--- a/iw/GroupSelector.py
+++ b/iw/GroupSelector.py
@@ -97,7 +97,7 @@ def _deselectPackage(ayum, group, pkg):
except mdErrors.PackageSackError:
log = logging.getLogger("yum.verbose")
log.debug("no such package %s from group %s" %(pkg,
- self.group.groupid))
+ group.groupid))
if pkgs:
pkgs = ayum.bestPackagesFromList(pkgs)
for po in pkgs:
This can just be groupid, not group.groupid. See slightly above.
Erm, that should be grpid not groupid, but yes using that is better, fixed.
Can drive be "/dev/whatever" here, or is it just going to be "whatever"?
Note this is s390 only code, so I've looked at what RHEL-5.3 does as there we
actually have testing on s390, and it seems that its always just "whatever". So
I'm keeping this as is in the patch. In any case its better then the old
reference a non existing variable code.
Applying now.
Regards,
Hans
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/anaconda-devel-list