Half a year ago I noticed that the cdf package is badly maintained and wrote a new ebuild for v.3.0. While I was at it I also noticed that it seems like upstream names the tarballs "cdfXY-*" no matter to which x.y.Z version it has been updated. So either upstream should be convinced to include the full version in the tarball or we need to rename the tarballs and mirror locally.
Created attachment 81650 [details] cdf-3.0.ebuild In case someone picks this up and finds the ebuild useful. I really don't want to maintain this package, though.
It is certainly useful to me on ppc. The 2.7 version in portage has text relocations problems when I used it with labplot. Your ebuild didn't suffer from this problem. May be it should be forwarded to the science overlay?
Yet another year gone without anyone caring. Latest release is 3.1.1.
(In reply to comment #4) > Yet another year gone without anyone caring. Latest release is 3.1.1. > Hum... I will have to check the science overlay. If you care, may be you should become proxy-maintainer for the package? Or may be I should fill a bug for labplot and put this one as blocker for it :)
Feel free to fix whatever's bugging you, just don't break anything =)
carlo, can you please take care of this yourself? You likely know the package better than QA.
No, I'm not going to touch this package. I barely have the time to care for stuff I do already. If no one from the sci team wants to take care I propose removing the two optional dependencies on cdf and removing the package from the tree.
I updated cdf-3.1 in cvs. I don't know the package, but seems to be used. If anyone could give it a shot, please do. Closing for now.
