With sys-apps/portage-2.1_rc2-r3 i get wrong versionnumber (the to be installed instead of the currently installed) on blocking packages. Did not try with older versions of portage to check if this is a new issue or has been around for a while. Example: # emerge nx-x11 -vp Calculating dependencies... done! [blocks B ] <net-misc/nx-x11-1.5.0-r8 (is blocking net-misc/nx-x11-1.5.0-r8) [ebuild N ] net-misc/nxcomp-1.5.0-r2 513 kB [ebuild U ] net-misc/nx-x11-1.5.0-r8 [1.4.0-r4] USE="-rdesktop% -vnc%" 32,574 kB # emerge -C nx-x11 net-misc/nx-x11 selected: 1.4.0-r4 protected: none omitted: none # emerge nx-x11 -vp Calculating dependencies... done! [ebuild N ] net-misc/nxcomp-1.5.0-r2 513 kB [ebuild N ] net-misc/nx-x11-1.5.0-r8 USE="-rdesktop -vnc" 32,574 kB Notice how the blocking version (1.5.0-r8) is the same as the one to be installed even though the actually installed is 1.4.0-r4
Take a closer look and you'll notice the < operator.
I might also point out that it appears only sometimes (when an older version of a package blocks a newer version?) These example eg appear to be correct [blocks B ] <net-print/foomatic-db-20050910 (is blocking net-print/foomatic-filters-3.0.2-r1) [blocks B ] <media-libs/faad2-2.0-r9 (is blocking media-libs/libmp4v2-1.4.1)
Ahh ok, sorry. Easy to miss/realise that that small < . Wouldnt it be more logical/intuitive to actually state the current version instead of the < sign? That way you would imediately understand what is going on. :) Reopening as request for enhancment
Not really IMHO. By using the exact dep statement people can see what range of packages are blocking, if you just list a single version you loose that info. Leaving open for others to comment.
> By using the exact dep statement people can see what range of packages are blocking, if you just list a single version you loose that info. Yes, I guess that is true, but does it really work correctly? Eg, would nx-x11-1.5.0-r7 really also be blocking? I cant test since unfortunately r7 (or any other 1.5.0 for that matter) is not the portage tree currently. And if it really is <-r8 , it would still be more intuitive to print <=nx-x11-1.5.0-r7 is blocking, but I guess that would be needed to be implemented as a "best way to write an ebuild"-convention.
(In reply to comment #5) > And if it really is <-r8 , it would still be more intuitive to print > <=nx-x11-1.5.0-r7 is blocking, but I guess that would be needed to be > implemented as a "best way to write an ebuild"-convention. It takes the blocking version from the ebuild... DEPEND="!<net-misc/nx-x11-1.5.0-r8 <--- here! ~net-misc/nxcomp-1.5.0 !net-misc/nx-x11-bin" I don't see why <=net-misc/nx-x11-1.5.0-r7 would be more intuitive. Lets focus on Bug 79606 instead of trying to solve cosmetic, completely unimportant stuff.
Apparently noone else wants to comment here. Not a bug.
