Created attachment 489546 [details, diff] makes it build diff --git a/dev-util/kscope/kscope-1.9.4-r2.ebuild b/dev-util/kscope/kscope-1.9.4-r2.ebuild index a1d378341..bdd8e35f0 100644 --- a/dev-util/kscope/kscope-1.9.4-r2.ebuild +++ b/dev-util/kscope/kscope-1.9.4-r2.ebuild @@ -16,7 +16,7 @@ IUSE="" RDEPEND="dev-qt/qtcore:4 dev-qt/qtgui:4 - x11-libs/qscintilla:=[qt4(-)]" + >=x11-libs/qscintilla-2.10:=[qt4(-)]" DEPEND="${RDEPEND}" DOCS="ChangeLog" @@ -33,5 +33,7 @@ src_prepare() { # fix build failure with parallel make echo "CONFIG += ordered" >> kscope.pro - epatch "${FILESDIR}/${P}"-{actions,underlinking}.patch + epatch \ + "${FILESDIR}/${P}"-{actions,underlinking}.patch \ + "${FILESDIR}"/qscintilla-2.10.patch }
Thanks for your contribution, can you create a version of the patch that does not raise dependency on qscintilla and requires us to mask kscope as well?
Thanks for creating the tracker bug. Note that with a single exception (app-editors/qwriter; and dev-vcs/hgview to check qscintilla-python), I didn't make a runtime test of the packages, but I checked that if they had qscintilla*.so in any NEEDED, they have it still after updating qscintilla and rebuilding them. (This was to reduce the risk of a silent "functionality downgrade" like in the case of sci-mathematics/octave - bug 617024.) Plus I didn't make sure to handle all packages that depends on scintilla (but looks like only dev-db/tora is left to check). (In reply to Andreas Sturmlechner from comment #1) > Thanks for your contribution, can you create a version of the patch that > does not raise dependency on qscintilla and requires us to mask kscope as > well? OK. I don't know much about qmake, but from what I saw so far, it doesn't provide a clean way to do it. What do you think about 'has_version xxxx && epatch'?
Relies on Qt 4 and has self-declared dead upstream. Time to go?
Package removed.
