Re: [arch-haskell] collaborative HABS maintenance

On 16/01/11 05:58, Leif Warner wrote:

I thought it was simply a matter of running cabal2arch on what's on hackage?

In general this is true, but it isn't enough if: - Upstream has put in incomplete dependency information in the CABAL file - You need a pkgrel > 1

I haven't kept up with everything on this list, but I was under the impression there was a file somewhere that told cabal2arch which package names not to prepend with "haskell-". Would anyone know where to configure that? Looking back in the mailing list I found a mention of "Other thoughts: * The translation table of Hackage names to Arch names (currently hardcoded into cabal2arch) should be easily accessible programmatically for use in scripts and other tools (e.g. bauerbill)." but I couldn't find such a list in the cabal2arch source. -Leif

The algorithm is - If the package contains a library, then name it haskell-X - If the package contains only executables, then name it X See 'cbl2pkg' in Distribution.Archlinux. /M

On Sat, Jan 15, 2011 at 3:55 AM, Magnus Therning mailto:magnus@therning.org> wrote:

On 15/01/11 11:41, Peter Simons wrote: > Hi Leif, > >> There's an awful lot of packages in that dropped list that I use. I could >> maintain PKGBUILDs for them myself, either on HABS or AUR directly. > > that would be really great. Yes, please help us maintain those packages! > > The contents of the HABS tree determines what goes into the binary > repository. If you'd like to see your packages included in there, then > HABS is the way to go. Please note that we upload the HABS packages to > AUR, too. If you prefer to maintain the packages on AUR, though, then > just let us know; we'll orphan them so that you can take them over. > > Generally speaking, we need to figure out how to maintain the HABS tree > collaboratively, but an obvious way to do it would be the following > procedure: > > (1) Fork 'archhaskell/habs' on Github. > > (2) Modify the PKGBUILDs as you see fit. > > (3) Ensure that everything compiles by running 'makeworld'. > > (4) Create a pull request to have your changes merged back. > > Magnus and I would then pull the changes, run the necessary re-builds, > and copy the binary packages to the server so that everyone else gets > them, too. > > The beauty of this scheme is that basically *everyone* can contribute > patches and updates in an efficient and transparent manner, and everyone > else can access those changes without having to wait for Magnus or me to > do anything. If we are too slow for your taste, then just set up your > own binary repository! :-)

I have just one thing to add to this. We are more than happy to hand out commit access straight to the archhaskell/habs repo in the future. But please follow Peter's suggestion first for a few updates.

/M

-- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org mailto:magnus@therning.org jabber: magnus@therning.org mailto:magnus@therning.org twitter: magthe http://therning.org/magnus

_______________________________________________ arch-haskell mailing list arch-haskell@haskell.org mailto:arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

_______________________________________________ arch-haskell mailing list arch-haskell@haskell.org http://www.haskell.org/mailman/listinfo/arch-haskell

-- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus@therning.org jabber: magnus@therning.org twitter: magthe http://therning.org/magnus