Bug#907718: python-dateutil breaks python-vobject autopkgtest
On Wed, Jan 09, 2019 at 05:55:47PM +0200, Adrian Bunk wrote:
> On Thu, Dec 13, 2018 at 11:21:35PM +0100, Jonas Smedegaard wrote:
> > Control: tags -1 help
> >
> > Quoting Guido Günther (2018-12-13 19:21:45)
> > > On Thu, Dec 13, 2018 at 05:53:03PM +0100, Jonas Smedegaard wrote:
> > > > @Guido, could you please have a fresh look at this bug to see if it
> > > > can either be lowered or fully solved, now that (I believe) all
> > > > blocking bugs are addressed?
> > >
> > > If you have some spare cycles it'd trust your judgement here. I have
> > > this on my todo list since some time and it keeps slipping.
> >
> > I would have acted already if I could: I don't really understand what
> > the remaining bug _is_ here (as also indicated by how I earlier closed
> > it in error).
>
> The only potential issue here is adding Breaks against the older
> versions of packages that were broken (not just build/test breakage).
>
> This would help preventing problems during stretch->buster upgrades
> (or if python-dateutil gets backported to stretch).
The attached patch adds the necessary Breaks: against the versions from
the affected packages before they got fixed.
We could probably upload that and move on with dateutil in buster.
diff --git a/debian/control b/debian/control
index 7cef506..5bc440a 100644
--- a/debian/control
+++ b/debian/control
@@ -22,6 +22,8 @@ Package: python-dateutil
Architecture: all
Depends: ${python:Depends}, ${misc:Depends}, tzdata
XB-Python-Version: ${python:Version}
+Breaks: python-icalendar (<< 4.0.3-2~),
+python-vobject (<< 0.9.6.1-0.1~),
Description: powerful extensions to the standard Python datetime module
It features:
.
@@ -45,6 +47,9 @@ Package: python3-dateutil
Architecture: all
Depends: ${python3:Depends}, ${misc:Depends}, tzdata
XB-Python-Version: ${python3:Version}
+Breaks: khal (<< 1:0.9.10-0.1~),
+python3-icalendar (<< 4.0.3-2~),
+python3-vobject (<< 0.9.6.1-0.1~),
Description: powerful extensions to the standard Python 3 datetime module
It features:
.
signature.asc
Description: PGP signature
Bug#907718: python-dateutil breaks python-vobject autopkgtest
On Thu, Dec 13, 2018 at 11:21:35PM +0100, Jonas Smedegaard wrote: > Control: tags -1 help > > Quoting Guido Günther (2018-12-13 19:21:45) > > On Thu, Dec 13, 2018 at 05:53:03PM +0100, Jonas Smedegaard wrote: > > > @Guido, could you please have a fresh look at this bug to see if it > > > can either be lowered or fully solved, now that (I believe) all > > > blocking bugs are addressed? > > > > If you have some spare cycles it'd trust your judgement here. I have > > this on my todo list since some time and it keeps slipping. > > I would have acted already if I could: I don't really understand what > the remaining bug _is_ here (as also indicated by how I earlier closed > it in error). The only potential issue here is adding Breaks against the older versions of packages that were broken (not just build/test breakage). This would help preventing problems during stretch->buster upgrades (or if python-dateutil gets backported to stretch). > - Jonas cu Adrian -- "Is there not promise of rain?" Ling Tan asked suddenly out of the darkness. There had been need of rain for many days. "Only a promise," Lao Er said. Pearl S. Buck - Dragon Seed
Processed: Re: Bug#907718: python-dateutil breaks python-vobject autopkgtest
Processing control commands: > tags -1 help Bug #907718 [src:python-dateutil] python-dateutil: RRULE UNTIL values must be specified in UTC when DTSTART is timezone-aware Added tag(s) help. -- 907718: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=907718 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems
Bug#907718: python-dateutil breaks python-vobject autopkgtest
Control: tags -1 help Quoting Guido Günther (2018-12-13 19:21:45) > On Thu, Dec 13, 2018 at 05:53:03PM +0100, Jonas Smedegaard wrote: > > @Guido, could you please have a fresh look at this bug to see if it > > can either be lowered or fully solved, now that (I believe) all > > blocking bugs are addressed? > > If you have some spare cycles it'd trust your judgement here. I have > this on my todo list since some time and it keeps slipping. I would have acted already if I could: I don't really understand what the remaining bug _is_ here (as also indicated by how I earlier closed it in error). - Jonas -- * Jonas Smedegaard - idealist & Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private signature.asc Description: signature
Bug#907718: python-dateutil breaks python-vobject autopkgtest
Hi, On Thu, Dec 13, 2018 at 05:53:03PM +0100, Jonas Smedegaard wrote: > Hi, > > @Guido, could you please have a fresh look at this bug to see if it can > either be lowered or fully solved, now that (I believe) all blocking > bugs are addressed? If you have some spare cycles it'd trust your judgement here. I have this on my todo list since some time and it keeps slipping. -- Guido
Bug#907718: python-dateutil breaks python-vobject autopkgtest
Hi, @Guido, could you please have a fresh look at this bug to see if it can either be lowered or fully solved, now that (I believe) all blocking bugs are addressed? - Jonas -- * Jonas Smedegaard - idealist & Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private signature.asc Description: signature
Bug#907718: python-dateutil breaks python-vobject autopkgtest
Hi Adrian, On 02-09-18 10:04, Adrian Bunk wrote: >>> This also causes FTBFS, the first two seen are: >>> https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/python-icalendar.html >>> https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/python-vobject.html >> This was broken by efba12d3 due to >> >> https://github.com/dateutil/dateutil/issues/620 >> >> so we likely want reverse deps to catch up and fix this. > > Adding khal as affected package: > https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/khal.html Any reason why you don't also create FTBFS bug reports for these reverse dependencies while you're at it (as Guido said this change is here to stay)? They are now not notified that they need to fix anything IIUC. (Merely having one's package being affected by an other bug isn't enough to trigger action in my mind). Those bugs should be blocking this bug, like what we do with other bugs that need versioned Breaks. Paul signature.asc Description: OpenPGP digital signature
Bug#907718: python-dateutil breaks python-vobject autopkgtest
Control: affects -1 src:khal
On Sat, Sep 01, 2018 at 03:47:58PM +0200, Guido Günther wrote:
>...
> > On Fri, Aug 31, 2018 at 09:41:32PM +0200, Paul Gevers wrote:
> > > Source: python-dateutil, python-vobject
> > > Control: found -1 python-dateutil/2.7.3-1
> > > Control: found -1 python-vobject/0.9.5-2
> > > X-Debbugs-CC: debian...@lists.debian.org
> > > User: debian...@lists.debian.org
> > > Usertags: breaks needs-update
> > >
> > > Dear maintainers,
> > >
> > > With a recent upload of python-dateutil the autopkgtest of
> > > python-vobject started to fail in testing. I copied some of the output
> > > below.
> > >...
> >
> > This also causes FTBFS, the first two seen are:
> > https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/python-icalendar.html
> > https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/python-vobject.html
> >
> > I strongly suspect that there will also be runtime issues with other
> > packages using python-dateutil currently in stretch and buster, so even
> > in case this is considered a bug in reverse dependencies there will
> > still be Breaks required in python{,3}-dateutil against packages in
> > stretch and buster that got broken (unless it was only a broken test).
>
> This was broken by efba12d3 due to
>
> https://github.com/dateutil/dateutil/issues/620
>
> so we likely want reverse deps to catch up and fix this.
Adding khal as affected package:
https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/khal.html
> Cheers,
> -- Guido
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
Processed: Re: Bug#907718: python-dateutil breaks python-vobject autopkgtest
Processing control commands: > affects -1 src:khal Bug #907718 [src:python-dateutil] python-dateutil: RRULE UNTIL values must be specified in UTC when DTSTART is timezone-aware Added indication that 907718 affects src:khal -- 907718: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=907718 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems
Bug#907718: python-dateutil breaks python-vobject autopkgtest
Hi,
On Sat, Sep 01, 2018 at 02:54:25PM +0300, Adrian Bunk wrote:
> Control: reassign -1 src:python-dateutil 2.7.3-1
> Control: retitle -1 python-dateutil: RRULE UNTIL values must be specified in
> UTC when DTSTART is timezone-aware
> Control: severity -1 serious
> Control: affects -1 python-dateutil python3-dateutil src:python-vobject
> src:python-icalendar
Thanks!
>
> On Fri, Aug 31, 2018 at 09:41:32PM +0200, Paul Gevers wrote:
> > Source: python-dateutil, python-vobject
> > Control: found -1 python-dateutil/2.7.3-1
> > Control: found -1 python-vobject/0.9.5-2
> > X-Debbugs-CC: debian...@lists.debian.org
> > User: debian...@lists.debian.org
> > Usertags: breaks needs-update
> >
> > Dear maintainers,
> >
> > With a recent upload of python-dateutil the autopkgtest of
> > python-vobject started to fail in testing. I copied some of the output
> > below.
> >...
>
> This also causes FTBFS, the first two seen are:
> https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/python-icalendar.html
> https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/python-vobject.html
>
> I strongly suspect that there will also be runtime issues with other
> packages using python-dateutil currently in stretch and buster, so even
> in case this is considered a bug in reverse dependencies there will
> still be Breaks required in python{,3}-dateutil against packages in
> stretch and buster that got broken (unless it was only a broken test).
This was broken by efba12d3 due to
https://github.com/dateutil/dateutil/issues/620
so we likely want reverse deps to catch up and fix this.
Cheers,
-- Guido
Processed: Re: Bug#907718: python-dateutil breaks python-vobject autopkgtest
Processing control commands: > reassign -1 src:python-dateutil 2.7.3-1 Bug #907718 [src:python-dateutil, src:python-vobject] python-dateutil breaks python-vobject autopkgtest Bug reassigned from package 'src:python-dateutil, src:python-vobject' to 'src:python-dateutil'. No longer marked as found in versions python-dateutil/2.7.3-1 and python-vobject/0.9.5-2. Ignoring request to alter fixed versions of bug #907718 to the same values previously set Bug #907718 [src:python-dateutil] python-dateutil breaks python-vobject autopkgtest Marked as found in versions python-dateutil/2.7.3-1. > retitle -1 python-dateutil: RRULE UNTIL values must be specified in UTC when > DTSTART is timezone-aware Bug #907718 [src:python-dateutil] python-dateutil breaks python-vobject autopkgtest Changed Bug title to 'python-dateutil: RRULE UNTIL values must be specified in UTC when DTSTART is timezone-aware' from 'python-dateutil breaks python-vobject autopkgtest'. > severity -1 serious Bug #907718 [src:python-dateutil] python-dateutil: RRULE UNTIL values must be specified in UTC when DTSTART is timezone-aware Severity set to 'serious' from 'normal' > affects -1 python-dateutil python3-dateutil src:python-vobject > src:python-icalendar Bug #907718 [src:python-dateutil] python-dateutil: RRULE UNTIL values must be specified in UTC when DTSTART is timezone-aware Added indication that 907718 affects python-dateutil, python3-dateutil, src:python-vobject, and src:python-icalendar -- 907718: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=907718 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems
