Your comments open lot of light for me. May be I could help in some bugs and the trivial ones, in order to have 0.8 ASAP. With this in mind I consider that depend on 0.8 is good enough for 5.2 release, unless GDA maintainer consider to keep Vala extension on its branch until 0.8 is close or is released.
2011/12/8 Maciej Marcin Piechotka <[email protected]>: > On Thu, 2011-12-08 at 16:12 -0600, Daniel Espinosa wrote: >> I've recently implemented some libgee-0.7 interfaces for libgda[1], >> and I want to merge to master, but may be 0.8 will not be ready for >> libgda-5.2 release? >> >> What are your planed milestones and release plan? >> > > Right now unfortunately it's partially 'when it's done'. The main > problem is the amount of time I had available during this term. I hope > that during next term it will be better. > > The main parts which have not been done (and been planned): > > - Adapter for Set in terms of Map interface. Should be very simple (I > was considering posting Gnome Love bug for that). > - Adding signals. Should be very simple. > - Implementing ConcurrentSortedSet. The changes in interfaces could be > quite simple and done quickly while the implementation may be moved > outside scope of 0.8 release (in any case main part of concurrent data > structures was already implemented). > - Removing the _impl functions. For this I need to have bug #640330[1] > solved. From what I understand Jürg have some plans for semantic > analyser in Vala etc. Releasing the 0.8 before this bug is solved will > result in very ugly API which I would prefer to avoid. > - Finally finishing SortedMap implementation. The solution requires to > solve bug #596549[2]. > - Polishing subset implementation (can be offseted till later). I was > about to post some question about that. > > Assuming lack of #640330 the changes could be done before New Year > making the release before libgda-5.2 (I assume it corresponds to Gnome > 3.4). > > Those points were also considered: > > - Reverse iterators/collections. While they should be reasonable simple > to write someone needs to write them ;) I had patches but they cannot be > used after introduction of concurrent data structures. > - Asynchronous iterators. It requires probably either making all > iterators async (as the API/ABI is broken it should not be a big deal) > or the Vala extension[3]. The original intention was to iterate over > file in nice way (possibly introducing helper for it later on) but I > guess they would fit the iteration over database records nice as well. > Probably for full use would require to have support for #604827[4] > > Both should be also relatively easy to implement. > > I consider bug #640330 especially 'annoying' as it both prevent from > API/ABI cleanness I would like to have in 0.8 and it prevents to have > short conditional compilation between 0.6 and 0.8. > > I would say the API/ABI stable version is max(bug #596549 solved + Δt, > bug #640330 solved + Δt, end of January). > > Unfortunately both Jürg (from what I heard on #vala) and I were busy so > I am a bit out of touch and I cannot say any exact date or Vala version > (some of Vala enhancements are not necessary trivial). People from folks > considered conditional compilation as described in bug #661102[5] > (depending on #640330). > > I am sorry that I cannot give exact time. > > Best regards > > [1] https://bugzilla.gnome.org/show_bug.cgi?id=640330 > [2] https://bugzilla.gnome.org/show_bug.cgi?id=596549 > [3] Previously I believed that it requires extension but I see a way to > workaround it. > [4] https://bugzilla.gnome.org/show_bug.cgi?id=604827 > [5] https://bugzilla.gnome.org/show_bug.cgi?id=661102 > -- Trabajar, la mejor arma para tu superación "de grano en grano, se hace la arena" (R) (en trámite, pero para los cuates: LIBRE) _______________________________________________ gnome-db-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/gnome-db-list
