Description of problem: Could not parse header: 'Nils Philippsen <nils>', expected: 'Firstname Lastname <email> - version-release' Version-Release number of selected component (if applicable): gnome-packagekit-2.32.0-2.fc14.x86_64 How reproducible: didn't try Steps to Reproduce: 1. Open gpk-update-viewer with sane-backends-1.0.21-5.fc14 as an available update. Actual results: sane-backends-1.0.21-5.fc14 This update will fix bugs and other non-critical problems. This notification was issued on 11/23/2010. For more information about this update please visit this website: • https://admin.fedoraproject.org/updates/F14/FEDORA-2010-18115. For more information about bugs fixed by this update please visit this website: • https://bugzilla.redhat.com/show_bug.cgi?id=603321. This is a test update, and is not designed for normal use. Please report any problems or regressions you encounter. The developer logs will be shown as no description is available for this update: 2010-11-23 Nils Philippsen <nils> - 1.0.21-5 • build mustek_usb2 backend again, enable use of libpthread (#603321) Could not parse header: 'Nils Philippsen <nils>', expected: 'Firstname Lastname <email> - version-release' Expected results: Should be able to parse changelog, or at least show the unparsed input if there is an error in parsing. Additional info: The newest changelog entries: * Tue Nov 23 2010 Nils Philippsen <nils> - 1.0.21-5 - build mustek_usb2 backend again, enable use of libpthread (#603321) * Mon Nov 08 2010 Nils Philippsen <nils> - let sane-backends require arch-specific version/release of -libs (#621217) * Wed Nov 03 2010 Nils Philippsen <nils> - 1.0.21-4 - xerox_mfp: correct color mode malfunction (#614949) - xerox_mfp: add USB id for SCX-4500W (#614948) * Fri Jun 25 2010 Nils Philippsen <nils> - 1.0.21-3 - build with -fno-strict-aliasing - use PIC/PIE because SANE-enabled software is likely to deal with data coming from untrusted sources (client <-> saned via network) * Mon Jun 07 2010 Nils Philippsen <nils> - rectify devel subpackage description * Wed Jun 02 2010 Nils Philippsen <nils> - 1.0.21-2 - fix pkgconfig file (#598401) * Wed May 05 2010 Nils Philippsen <nils> - 1.0.21-1 - version 1.0.21 - remove obsolete rpath, hal, genesys-gl841-registers patches - update pkgconfig, udev, man-utf8->man-encoding, epson-expression800, docs-utf8 patches - remove hal conditional - package man pages * Fri Feb 26 2010 Nils Philippsen <nils> - 1.0.20-12 - convert some documentation files to UTF-8 - fix permissions of pkgconfig file * Tue Dec 29 2009 Nils Philippsen <nils> - 1.0.20-11 - genesys_gl841: always send registers before trying to acquire a line (#527935) * Mon Dec 28 2009 Nils Philippsen <nils> - build v4l backend (#550119) - don't use lockdir, fix make install * Thu Oct 22 2009 Nils Philippsen <nils> - 1.0.20-10 - don't set owner, group or mode as this may interfere with setting ACLs * Thu Oct 22 2009 Nils Philippsen <nils> - 1.0.20-9 - fix device file ownership and mode
If Nils doesn't put the version number in the changelog then there's no way PackageKit can filter the results to something sane. Not putting description text in bohdi is bad, not putting description text in bodhi and then not providing a parseable changelog is worse. If we want to give our users a non-laughable update experience, we have to start caring a bit more about the update details we give. Sorry to be blunt.
Given that package changelogs were never designed to be parsed the way PackageKit is doing, and yet it is attempting to do it anyway, I believe this principle applies even moreso: Be conservative in what you send; be liberal in what you accept[1] [1] http://en.wikipedia.org/wiki/Robustness_principle So yes, perhaps Nils should put version numbers in changelog entries. But also, PackageKit should accept changelog entries without version numbers. Just spitting out an error message and stopping the parse there doesn't contribute to providing a "non-laughable update experience" either. What exactly is the insurmountable problem of parsing that PackageKit can't overcome? Why can't it treat the version number in the changelog as optional, and gracefully proceed if it is missing? Should I work up a patch that does this? Or, alternatively, why don't we change all our tools to enforce the requirement of having a version in the changelog entries if that is the One True Way(tm)? rpmlint, rpmbuild, koji, bodhi are possible enforcement points for this.
I agree that some of the details in the changelog are probably not that interesting to users, I've filled in the update details in bodhi accordingly. However, gnome-packagekit shouldn't barf on version-less changelog entries as e.g. the reason for this one was that there was no package built from that change, i.e. it was queued up for the subsequent -5 release. I'd suggest that gnome-packagekit simply continues looking for the last released version in the following changelog entries in this situation instead. Mind that arbitrarily formatted, grandfathered changelog entries may appear in packages imported into Fedora from elsewhere.
This message is a notice that Fedora 14 is now at end of life. Fedora has stopped maintaining and issuing updates for Fedora 14. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At this time, all open bugs with a Fedora 'version' of '14' have been closed as WONTFIX. (Please note: Our normal process is to give advanced warning of this occurring, but we forgot to do that. A thousand apologies.) Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, feel free to reopen this bug and simply change the 'version' to a later Fedora version. Bug Reporter: Thank you for reporting this issue and we are sorry that we were unable to fix it before Fedora 14 reached end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged to click on "Clone This Bug" (top right of this page) and open it against that version of Fedora. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete. The process we are following is described here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping