Re: [Fink-devel] Policy question

[Daniel H. <jad...@nc...>]

It was a new package submission where I told the author that they needed
to change the /sw's to %p in some manner, and suggested looking at
dpkg.info as an example of "patching the patch", but the author then said
there was no need to do that and that they added a DescPackaging line to
indicate that it needed to be changed by hand.  Just wanted to make sure I
was right about /sw being forbidden before saying it.  ;)  Thanks!

Daniel

-- 
"The most addictive drug in the world is music."
     - The Lost Boyz

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Daniel Henninger wrote:
> | Is this not against fink policy?
> | DescPackaging: <<
> | Don't mind the following result:
> | $ fink -v validate xinitrc.info
> | Warning: Patch file appears to contain a hardcoded /sw. (xinitrc.patch)
> |
> | You must *manually* ensure the prefix cleanness.
> | <<
> |
> | There are always sed-based methods to automate setting up the proper
> | paths, are there not?
>
> I'm not sure of the context here, care to elaborate?
>
> Ususally the method for dealing with patches which require a path is to
> modify the patch so that it contains something like @PREFIX@ instead of
> /sw
> and then use a patchscript
> PatchScript: sed -e 's|@PREFIX@|%p|g' < %a/%n.patch |patch -p1
>
> So, yes, you are correct, it is against fink policy to hardcode the
> prefix.
>
> Peter
> - --
> Peter O'Gorman - http://www.pogma.com
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.6 (Darwin)
>
> iQCVAwUBQfzpLLiDAg3OZTLPAQKFdgP/TsA6BsIrBxouXkbQx89oAHBRncbRl6AV
> 9Psx7GIFTuEgNn2/rFmIDAYn9FusXYtqEZEuBNM1M19S1+GKC5taasmHUnsriT+m
> r4Ea6q8QpVV97pa2FAWUxFFKPVEEIP5piV4cP60hxsmqz5ZnYXllAuj4JeZEMx+V
> qztg5DvknIk=
> =+V6t
> -----END PGP SIGNATURE-----
>