Jesse Keating <jkeating <at> redhat.com> writes:
> I think you're too focused on the scenario where the same update is
> being pushed to all the branches, when that isn't always what is
> happening.
Well, it is a quite common case, I at least tend to do this often, and I think
spamming maintainers with bogus EVR problem reports for this is really
unhelpful, and may lead to them ignoring the real problems.
In a test run (of course with a script hacked not to spam maintainers or
mailing lists, just me ;-) ) for a patch I have developed (see the end of this
mail) for the checker (using the current real Koji data), a total of 30 bogus
reports [1] were there (all removed with the fix)! That's out of a total of 76
reports, i.e. 39% of the reports are bogus!
> It's quite possible that different updates are being pushed,
> could be at the same time, could be at different times, and those
> updates could be completely unrelated to each other.
The case where there is no matching update in dist-f9-updates-testing isn't
affected by my patch at all, e.g. it still reports this one (where the F9
update hasn't hit testing yet, so of course the script can't know it's pending
for testing; but this is the exception rather than the norm, for example it
just affects 1 package in this run):
iptables:
dist-f8-updates-testing > dist-f9-updates (iptables-1.4.1.1-2.fc8
iptables-1.4.1.1-1.fc9)
> Since this is just an informative mailing, I think it's safe to let the
> maintainer decide if there is an actual problem or not, just like they
> will when they've requested an update that will fix any N-V-R issues,
> but it just hasn't been pushed yet.
I think they will be really confused about whether the reports are real issues
or not, and start ignoring the reports even if they complain about real issues.
> Also I'm trying to keep any "domain" knowledge out of the script so that
> it's more easily usable for other sites making use of koji. The only
> real domain knowledge in it is the variable definitions at the top of
> the script and the one place where we assume <package>-owner@ is a valid
> contact for the package.
With my patch, no special domain information is needed, you only have to
prepend a / to an updates tag following an updates-testing tag, i.e. invoke it
like this:
./check-upgrade-paths.py f8-final dist-f8-updates \
dist-f8-updates-testing /dist-f9-updates dist-f9-updates-testing dist-f10
The way this is handled is described in the code:
prepending a / is special: A /B C means A will not be checked against B, but
against the union of B and C
Only A is affected, everything preceding A will still be checked against B
normally, as will B against C.
But enough talking, here's the patch:
http://repo.calcforge.org/f10/check-upgrade-paths.py.diff
Apologies if the code is suboptimal, Python is not my preferred programming
language. ;-)
Kevin Kofler
[1]
abiword:
dist-f8-updates-testing > dist-f9-updates (1:abiword-2.6.4-2.fc8
1:abiword-2.6.4-1.fc9)
augeas:
dist-f8-updates-testing > dist-f9-updates (augeas-0.2.2-1.fc8
augeas-0.2.1-1.fc9)
bluez-libs:
dist-f8-updates-testing > dist-f9-updates (bluez-libs-3.35-1.fc8
bluez-libs-3.32-1.fc9)
bluez-utils:
dist-f8-updates-testing > dist-f9-updates (bluez-utils-3.35-3.fc8
bluez-utils-3.32-1.fc9)
bzr-gtk:
dist-f8-updates-testing > dist-f9-updates (bzr-gtk-0.94.0-5.fc8
bzr-gtk-0.94.0-2.fc9)
collectl:
dist-f8-updates-testing > dist-f9-updates (collectl-3.0.0-1.fc8
collectl-2.6.4-1.fc9)
decibel-audio-player:
dist-f8-updates-testing > dist-f9-updates (decibel-audio-player-0.10-2.fc8
decibel-audio-player-0.10-1.fc9)
ext3grep:
dist-f8-updates-testing > dist-f9-updates (ext3grep-0.7.0-1.fc8
ext3grep-0.6.0-1.fc9)
ez-ipupdate:
dist-f8-updates-testing > dist-f9-updates (ez-ipupdate-3.0.11-0.19.b8.fc8
ez-ipupdate-3.0.11-0.18.b8.fc9)
flashrom:
dist-f8-updates-testing > dist-f9-updates
(flashrom-0-0.11.20080607svn3418.fc8 flashrom-0-0.9.20080517svn3332.fc9)
gamazons:
dist-f8-updates-testing > dist-f9-updates (gamazons-0.83-3.fc8
gamazons-0.83-2.fc9)
gyachi:
dist-f8-updates-testing > dist-f9-updates (gyachi-1.1.35-16.fc8
gyachi-1.1.35-6.fc9)
nautilus-sendto:
dist-f8-updates-testing > dist-f9-updates (nautilus-sendto-1.0.1-1.fc8
nautilus-sendto-1.0.0-1.fc9)
notification-daemon-engine-nodoka:
dist-f8-updates-testing > dist-f9-updates
(notification-daemon-engine-nodoka-0.1.0-3.fc8
notification-daemon-engine-nodoka-0.1.0-2.fc9)
ocaml-json-static:
dist-f8-updates-testing > dist-f9-updates (ocaml-json-static-0.9.6-5.fc8
ocaml-json-static-0.9.6-4.fc9)
ocaml-openin:
dist-f8-updates-testing > dist-f9-updates (ocaml-openin-20070524-4.fc8
ocaml-openin-20070524-3.fc9)
ocaml-pa-monad:
dist-f8-updates-testing > dist-f9-updates (ocaml-pa-monad-1.2.0-5.fc8
ocaml-pa-monad-1.2.0-4.fc9)
ocaml-pgocaml:
dist-f8-updates-testing > dist-f9-updates (ocaml-pgocaml-1.1-3.fc8
ocaml-pgocaml-1.1-2.fc9)
pcmanfm:
dist-f8-updates-testing > dist-f9-updates (pcmanfm-0.5-1.fc8
pcmanfm-0.4.6.2-1.fc9)
pinot:
dist-f8-updates-testing > dist-f9-updates (pinot-0.87-1.fc8
pinot-0.86-1.fc9)
python-fedora:
dist-f8-updates-testing > dist-f9-updates (python-fedora-0.3.3-1.fc8
python-fedora-0.2.99.11.1-1.fc9)
qlandkarte:
dist-f8-updates-testing > dist-f9-updates (qlandkarte-0.7.3-1.fc8
qlandkarte-0.7.2-1.fc9)
rubygem-activeldap:
dist-f8-updates-testing > dist-f9-updates (rubygem-activeldap-1.0.1-1.fc8
rubygem-activeldap-0.10.0-10.fc9)
rubygem-hoe:
dist-f8-updates-testing > dist-f9-updates (rubygem-hoe-1.7.0-1.fc8
rubygem-hoe-1.5.1-5.fc9)
strigi:
dist-f8-updates-testing > dist-f9-updates (strigi-0.5.11-1.fc8
strigi-0.5.9-2.fc9)
tellico:
dist-f8-updates-testing > dist-f9-updates (tellico-1.3.3-1.fc8
tellico-1.3.2.1-1.fc9)
thunar-shares:
dist-f8-updates-testing > dist-f9-updates (thunar-shares-0.16-1.fc8
thunar-shares-0.12-1.fc9)
vala:
dist-f8-updates-testing > dist-f9-updates (vala-0.3.4-2.fc8
vala-0.3.4-1.fc9)
xenner:
dist-f8-updates-testing > dist-f9-updates (xenner-0.41-1.fc8
xenner-0.37-1.fc9)
xfce4-screenshooter-plugin:
dist-f8-updates-testing > dist-f9-updates
(xfce4-screenshooter-plugin-1.3.1-1.fc8 xfce4-screenshooter-plugin-1.2.0-1.fc9)
--
fedora-devel-list mailing list
fedora-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/fedora-devel-list
