GRASS 6.1.0 was released this morning, please put in portage new ebuild.
Created attachment 94006 [details] grass directory in $PORTDIR/sci-geosciences, containing grass-6.1.0.ebuild and relative files Proposed ebuild for grass-6.1.0. Unpack the tarball in $PORTDIR/sci-geosciences.
http://allen.brooker.gb.net/misc/kitten-0day.jpg
(In reply to comment #2) > http://allen.brooker.gb.net/misc/kitten-0day.jpg Yeah, sorry for the killed kitten... If God killed one kitten for each working ebuild I've posted and 1 month old... GIS sws are quicker in development. And sorry if I try to lighten developer workload...
he?.. something strange.. I can't download the attachment from there :(
(In reply to comment #4) > he?.. something strange.. I can't download the attachment from there :( I just did it. Well, I uploaded the tarball also at this URL: http://syntaxerrormmm.altervista.org/dwls/grass-6.1.0.tar.bz2 Also, md5sum of the tarball: $ md5sum grass-6.1.0.tar.bz2 b9226d1242f9f2b3a5f76267d08a29d2 grass-6.1.0.tar.bz2
Wow, GRASS 6.2.0 (next stable) is nearly released and this bug isn't already fixed... Well, I will say Gentoo developers are less than Debian ones (I might not say so, I love Gentoo, but in some ways Bugzilla and developers group doesn't work as they should do)...
Unable to emerge grass 6.1.0 ebuild! kltgs1 grass # emerge -v grass >>> cfg-update-1.8.0-r3 : No new packages have been emerged, checksum index OK... Calculating dependencies... done! >>> Emerging (1 of 1) sci-geosciences/grass-6.1.0 to / >>> checking ebuild checksums ;-) >>> checking auxfile checksums ;-) >>> checking miscfile checksums ;-) >>> checking grass-6.1.0.tar.gz ;-) >>> Unpacking source... !!! ERROR: sci-geosciences/grass-6.1.0 failed. Call stack: ebuild.sh, line 1539: Called dyn_unpack ebuild.sh, line 711: Called src_unpack grass-6.1.0.ebuild, line 67: Called built_with_use 'dev-lang/tcl' 'thread' eutils.eclass, line 1614: Called die !!! dev-lang/tcl-8.4.9 does not actually support the thread USE flag! !!! If you need support, post the topmost build error, and the call stack if relevant. !!! This ebuild is from an overlay: '/usr/local/portage' kltgs1 grass # Does the TCL use tag need to be changed to threads as example below? kltgs1 grass # emerge -v tcl --pretend >>> cfg-update-1.8.0-r3 : Building checksum index... (takes a few seconds) done! These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] dev-lang/tcl-8.4.9 USE="-threads" 0 kB Total size of downloads: 0 kB kltgs1 grass #
(In reply to comment #8) > Does the TCL use tag need to be changed to threads as example below? Yes, and not only tcl but also tk; it's an issue that I've seen yesterday after posting here. Fix launching: sed -e -i 's/\(dev-lang\/tcl thread\)/\1s/; s/\(dev-lang\/tk thread\)/\1s/' $PORTDIR_OVERLAY/sci-geosciences/grass/grass-6.1.0.ebuild and re-digesting the ebuild. Thanks for the report.
(In reply to comment #9) > sed -e -i 's/\(dev-lang\/tcl thread\)/\1s/; s/\(dev-lang\/tk thread\)/\1s/' > $PORTDIR_OVERLAY/sci-geosciences/grass/grass-6.1.0.ebuild Well, sorry, without '-e' option.
Modified ebuild now in portage; it still has some missing math symbols if built without Blas/Lapack (gmath USE flag).
(In reply to comment #11) > Modified ebuild now in portage; it still has some missing math symbols if built > without Blas/Lapack (gmath USE flag). Well, it has some issues on dependency (proj listed as required dependency AND as optional one, if 'proj' useflag is used; ncurses *is needed* for a clean compile; man system dependency) and some others, which are best viewed in bug 153796. Thanks.
