On Oct 6, 2008, at 3:02 AM, Martin Costabel wrote: > Alexander Hansen wrote: >> On Oct 5, 2008, at 3:04 PM, Gary K Olson wrote: >>> After installing the new version of clisp (2.46-2), the updated >>> maxima >>> would not install. I got the following message: >>> >>> Reading buildlock packages... >>> /sw/bin/dpkg-lockwait -i /sw/fink/dists/unstable/crypto/binary- >>> darwin- >>> powerpc/maxima_5.16.3-2_darwin-powerpc.deb >>> (Reading database ... 257364 files and directories currently >>> installed.) >>> Preparing to replace maxima 5.16.3-2 (using .../ >>> maxima_5.16.3-2_darwin- >>> powerpc.deb) ... >>> Unpacking replacement maxima ... >>> dpkg: dependency problems prevent configuration of maxima: >>> maxima depends on clisp (= 2.46-1); however: >>> Version of clisp on system is 2.46-2. > [] >> fink uses a cache for all of the .info and .patch files to speed >> up operations: but we've been getting a number of reports of late >> of this cache not getting updated properly under the normal >> indexing operation. >> "fink index -f" will forcibly flush and update the cache, and fink >> will then have the correct information to use. > > Maybe in Gary's case it really was a stale cache, but there is a > real obstacle to update clisp/maxima with the current logic: > > When clisp and maxima get new versions and you run `fink update-all` > or simply `fink update maxima`, it will fail, because maxima depends > on the new version clisp, so clisp is built first, but it cannot be > installed, because the old maxima is still installed and depends on > the old clisp. > > One first has to remove manually the old maxima, then install the > new clisp, and only then update maxima. During update-all, fink > talks about inconsistent dependencies and proposes to run > > fink scanpackages > sudo apt-get update > sudo apt-get install clisp=2.46-4 > > This has the same effect of removing the old version of maxima and > installing the new version of clisp. It does not, however, update > maxima. > > Alex, have you played with BuildConflicts (perhaps in clisp against > maxima)? The "Conflicts: maxima" in maxima does not seem to help here. > > -- > Martin > >
The Conflict on itself is a holdover from the prior varianted form of the packaging. I think it's a no-op in this case. I've not tried having clisp BuildConflict on maxima--that's worth looking into when I get some time. ------------------------------------------------------------------------- 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-users mailing list Fink-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-users
