As of August 8th 2012, poppler's flag xpdf-headers has been removed, but some ebuilds haven't been fixed for this. #snip# > grep "xpdf-headers" **/*.ebuild app-office/libreoffice/libreoffice-3.5.5.3.ebuild: pdfimport? ( >=app-text/poppler-0.16[xpdf-headers(+),cxx] ) app-office/libreoffice/libreoffice-3.5.6.2.ebuild: pdfimport? ( >=app-text/poppler-0.16[xpdf-headers(+),cxx] ) app-office/libreoffice/libreoffice-3.5.9999.ebuild: pdfimport? ( >=app-text/poppler-0.16[xpdf-headers(+),cxx] ) app-office/libreoffice/libreoffice-3.6.0.4.ebuild: libreoffice_extensions_pdfimport? ( >=app-text/poppler-0.16[xpdf-headers(+),cxx] ) app-office/libreoffice/libreoffice-3.6.9999.ebuild: libreoffice_extensions_pdfimport? ( >=app-text/poppler-0.16[xpdf-headers(+),cxx] ) app-office/libreoffice/libreoffice-9999-r2.ebuild: libreoffice_extensions_pdfimport? ( >=app-text/poppler-0.16[xpdf-headers(+),cxx] ) app-text/apvlv/apvlv-0.1.2-r1.ebuild: >=app-text/poppler-0.18[cairo,xpdf-headers(+)] app-text/apvlv/apvlv-0.1.4.ebuild: >=app-text/poppler-0.18[cairo,xpdf-headers(+)] app-text/calibre/calibre-0.8.62.ebuild: >=app-text/poppler-0.12.3-r3[qt4,xpdf-headers(+)] app-text/calibre/calibre-0.8.64.ebuild: >=app-text/poppler-0.12.3-r3[qt4,xpdf-headers(+)] app-text/calibre/calibre-0.8.65.ebuild: >=app-text/poppler-0.12.3-r3[qt4,xpdf-headers(+)] app-text/pdf2djvu/pdf2djvu-0.7.12.ebuild: >=app-text/poppler-0.16.7[xpdf-headers(+)] app-text/pdf2djvu/pdf2djvu-0.7.13.ebuild: >=app-text/poppler-0.16.7[xpdf-headers(+)] dev-tex/luatex/luatex-0.70.1.ebuild: >=app-text/poppler-0.12.3-r3[xpdf-headers(+)] dev-tex/luatex/luatex-0.70.1-r1.ebuild: >=app-text/poppler-0.12.3-r3[xpdf-headers(+)] dev-tex/pdftex/pdftex-1.40.11.ebuild:RDEPEND=">=app-text/poppler-0.12.3-r3[xpdf-headers(+)] media-gfx/inkscape/inkscape-0.48.2.ebuild: >=app-text/poppler-0.12.3-r3[cairo,xpdf-headers(+)] media-gfx/inkscape/inkscape-0.48.3.1.ebuild: >=app-text/poppler-0.12.3-r3[cairo,xpdf-headers(+)] net-print/cups-filters/cups-filters-1.0.23-r1.ebuild: app-text/poppler[cxx,jpeg?,lcms,tiff?,xpdf-headers(+)] net-print/cups-filters/cups-filters-9999.ebuild: app-text/poppler[cxx,jpeg?,lcms,tiff?,xpdf-headers(+)] #snap# Reproducible: Always Actual Results: (re-)emerges of affected packages fail Expected Results: should have been done
Actually, 'man 5 ebuild' - 'xpdf-headers(+)' construct means it's handled properly.
Minor correction - yes, that useflag dependency is now redundant, but as it's handled correctly anyway, there's no need to rush people to update the old ebuilds. This can be done at next convenient opportunity.
How is this invalid, if it's not currently solved by synching portage to the latest revision and packages fail because of it? I'm not having the problem, just reporting it, but something obviously is wrong here.
(In reply to comment #3) > How is this invalid, if it's not currently solved by synching portage to the > latest revision and packages fail because of it? I'm not having the problem, > just reporting it, but something obviously is wrong here. In such case, what exactly is the failure ?
Just look in the thread linked above or to make it short > emerge: there are no ebuilds built with USE flags to satisfy "app-text/poppler > [cxx,jpeg,lcms,tiff,xpdf-headers]". > !!! One of the following packages is required to complete your request: > - app-text/poppler-0.20.3::gentoo (Missing IUSE: xpdf-headers) > #more here# Again, just reporting that for another user. Don't know what machine he has and if he's running ~arch, but still, even for ~arch this would be such an easy fix that it's quite strange no one did it.
(In reply to comment #5) > Just look in the thread linked above or to make it short > > Again, just reporting that for another user. Don't know what machine he has > and if he's running ~arch, but still, even for ~arch this would be such an > easy fix that it's quite strange no one did it. Well, looking at that thread, I'd say that there might be a bug, bug elsewhere. - Still there's too little info to say for sure. - In that thread not even full emerge command isn't mentioned, neither is timestamp of the tree nor portage version. The problem *might* sort of lie in portage: the change in cups-fiters useflag deps was made without a revbump, so it seems portage isn't calculating the value of installed cups-filters version correctly (quite likely adding '-D' to emerge line would help).
I hit this issue this morning during a routine emerge @world. Calculating dependencies... done! !!! All ebuilds that could satisfy "app-text/poppler[cxx,jpeg,lcms,tiff,xpdf-headers]" have been masked. !!! One of the following masked packages is required to complete your request: - app-text/poppler-0.20.3::gentoo (masked by: ~amd64 keyword) - app-text/poppler-0.20.2-r2::gentoo (masked by: ~amd64 keyword) - app-text/poppler-0.18.4-r2::gentoo (masked by: exclude option) (dependency required by "net-print/cups-filters-1.0.20" [installed]) (dependency required by "net-print/cups-1.6.1" [installed]) (dependency required by "app-text/ghostscript-gpl-9.05-r1[cups]" [installed]) (dependency required by "net-print/foomatic-filters-4.0.16" [installed]) I had net-print/cups ~amd64 in my package.keywords file for some now-forgotten reason; commenting that out allowed the emerge @world to continue.
I'd say comment 7 confirms what I wrote earlier. Chances are that 'emerge -upvD @world' (perhaps with '--with-bdeps y', but not necessarily) would have worked.
*** This bug has been marked as a duplicate of bug 430582 ***
