Problems with ".if HAVE_GNOME" tests because of installation order

[Alexander Nedotsukov]

Alexander,

This is not gnome ports specific problem. Any port wich tries to guess 
it configuration in way
".if exists() xx_DEPENDS+=" have such flaw. Situation can be resolved 
sometimes by adding
WITHOUT_xx option if they supported by port (unfortunately this is not 
sysutils/cdrdao case).
The better way seems to use info stored in /var/db/pkg rather than do 
recursive make wailk through
ports tree but this I guess need more than my pure speculation on 
subject :-)

All the best,
Alexander.

Alexander Leidinger wrote:

>Hi,
>
>I haven't digged into this, it may be a portupgrade problem, not a
>problem of the ports collection.
>
>PR 61840 describes a scenario which seems to be a general problem with
>conditional gnome dependencies.
>
>Without having gnome installed install a port (A) which depends upon
>gnome only if it is installed. Now install gnome and after that a port
>which depends on A, but not on gnome.
>
>pkgdb -F will now report an error if port A sucks in some more
>dependencies in the gnome case, e.g. cdrdao checks for gnomelibs and
>adds gtk-- and gnome-- if gnomelibs is installed. If you install cdrdao,
>gnomelibs and then k3b, pkgdb -F will report a missing dependency
>(gnome--) for k3b (see PR 61840).
>
>Any ideas/thoughts about this problem?
>
>Bye,
>Alexander.
>
>  
>