On Wed, Aug 13, 2008 at 12:07:55AM -0400, Jack Howarth wrote:
>    Here are my two cents on this issue. In the past for the
> python variant packages I maintain, I have added conflicts
> with the view that it is undesirable to allow users to
> install multiple python variants of executable packages.
> The user should have the option to build the package against
> the python of his choice but it is highly unlikely that they
> really want multiple python variants installed at the same
> time. By allowing that, you just impose the user the responsibility
> of manually deinstalling the older python variant if they
> upgrade the package to a newer python variant. Also, currently
> fink tries to force the developer to resort to profile scripts
> to gain access to the desired binaries (since it doesn't allow
> symlinks in %p/bin to the access the variant. The problem with
> this approach is that it forces the developer to resort to
> adding the variant binary directory to the path. I think this
> is a rather poor design decision since it tends to create
> unnecessarily long path environmentals. In relax-py, I decided
> to avoid this issue by creating the desired symlink in the
> postinstall script and deleting it in the preremove script.

So essentially you subvert the whole variant-orthogonality issue
instead of just using update-alternatives to create automatic
fall-over among whatever variant(s) a user may choose to install?

dan

-- 
Daniel Macks
[EMAIL PROTECTED]
http://www.netspace.org/~dmacks


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Fink-devel mailing list
[email protected]
http://news.gmane.org/gmane.os.apple.fink.devel

Reply via email to