| Summary: | x11-libs/libqxt-0.6.0-r1: Multiple QA issues | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Markos Chandras (RETIRED) <hwoarang> |
| Component: | New packages | Assignee: | Qt Bug Alias <qt> |
| Status: | RESOLVED FIXED | ||
| Severity: | QA | Keywords: | InOverlay |
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | [qting-edge] | ||
| Package list: | Runtime testing required: | --- | |
| Bug Depends on: | |||
| Bug Blocks: | 331933 | ||
|
Description
Markos Chandras (RETIRED)
2010-08-08 00:15:32 UTC
This is a first-time stabilization at least for x86. Please fix the following QA issues: * Messages for package x11-libs/libqxt-0.6.0-r1: * QA Notice: Pre-stripped files found: * /usr/lib/libQxtGui.so.0.6.0 * /usr/lib/libQxtNetwork.so.0.6.0 * /usr/lib/libQxtCore.so.0.6.0 * /usr/lib/libQxtBerkeley.so.0.6.0 * /usr/lib/qt4/plugins/designer/libQxtDesignerPlugins.so * QA Notice: The following files contain insecure RUNPATHs * Please file a bug about this at http://bugs.gentoo.org/ * with the maintaining herd of the package. * /usr/lib/qt4:/var/tmp/portage/x11-libs/libqxt-0.6.0-r1/work/libqxt/lib usr/lib/libQxtGui.so.0.6.0 * /usr/lib/qt4:/var/tmp/portage/x11-libs/libqxt-0.6.0-r1/work/libqxt/lib usr/lib/libQxtNetwork.so.0.6.0 * /usr/lib/qt4:/var/tmp/portage/x11-libs/libqxt-0.6.0-r1/work/libqxt/lib usr/lib/libQxtCore.so.0.6.0 * /usr/lib/qt4:/var/tmp/portage/x11-libs/libqxt-0.6.0-r1/work/libqxt/lib usr/lib/libQxtBerkeley.so.0.6.0 * /usr/lib/qt4:/var/tmp/portage/x11-libs/libqxt-0.6.0-r1/work/libqxt/lib usr/lib/qt4/plugins/designer/libQxtDesignerPlugins.so Just from a quick look at the ebuild, shouldn't eqmake4 come *after* ./configure ? Also, the ebuild should inherit multilib instead of eutils. UnCC arches till we fix the ebuild
Davide, configure script runs qmake libqxt.pro -recursive
However, if I use
eqmake4 ${PN}.pro \
QXT_INSTALL_DOCS="/usr/share/doc/${PF}" \
QXT_INSTALL_FEATURES="/usr/share/qt4/mkspecs/features " \
-recursive
it seems like -recursive is not respected at all. Any thoughts?
Try swapping the order of arguments so that -recursive comes before the two variables assignments, but after ${PN}.pro
Nope Ah! It relies on the cache! The configure writes various config values to .qmake.cache which is usually read by qmake at the beginning of the parsing. However eqmake4 always disables the cache when calling qmake, thus breaking the above procedure! Unfortunately -nocache seems to have priority over -cache, so this issue is not fixable in the ebuild :( Nevermind, I've found a solution. I'll push an updated ebuild to qting-edge later. (In reply to comment #7) > Nevermind, I've found a solution. I'll push an updated ebuild to qting-edge > later. > Excellent. Please attach your patch here when you are ready so we can fix the ebuild on tree Sorry for the delay, I noticed some other issues with the current 0.6.0-r1 ebuild in portage that I wanted to fix, and I've been very busy in the last few days so it took quite some time to properly fix everything. I just committed a rewritten live ebuild to qting-edge and subsequently revbumped 0.6.0 applying the same improvements there too. Please read the git commit message for details. Don't forget the new patch and the updated metadata.xml when moving 0.6.0-r2 to portage! :) Thank you very much Davide I will try to review the new ebuild and commit it to portage as soon as possible +*libqxt-0.6.0-r2 (20 Aug 2010) + + 20 Aug 2010; Markos Chandras <hwoarang@gentoo.org> + +libqxt-0.6.0-r2.ebuild, metadata.xml: + Revbump to fix issues reported on bug #331585. New use flags, Adjusted + qt-assistant dependency, switched to qt4-r2 and EAPI=3, use systems' + qdoc3, fix arguments when crypt use flag was enabled. Many thanks to + Davide Pesavento <davidepesa@gmail.com> for his work, patches,review and + testing + |