On 2013-11-27 13:02:08, Lisandro Damián Nicanor Pérez Meyer wrote:
CCing the release.d.o bug.
Marcin: As you already stated, Calligra won't switch to libpqxx4 but to
libpq.
But that will happen (hopefully) early next year.
We really don't want to drop psql support. Is there any chance to have
libpqxx3 back with a bigger epoch until that release? The dev package is
libpqxx3-dev, so I think there will be no problem in having both versions 3
and 4 in the archive (you will have to drop the dummy package).
Lisandro,
Dropping versioning from *-dev *-doc and *-dbg was a conclusion to my decision
of not going with support for more then one libpqxx version in Debian. Having
above in mind I'm not saying that there is now way back for libpqxx3-dev etc..
But we have to consider at least 2 below points:
1. Calligra is switching to libpq according to you 'early next year' so even if
early mean end of March it's only 4 months away. This is rather short time
from now.
2. On 2012-06-03 I sent an email to debian-qt-...@lists.debian.org with
information that libpqxx4 is in experimental. In the same email I attached
calligra build log against libpqxx4 and an information that it FTBFS. None of
debian-qt-kde team members act on this email (at least I'm not aware of it)
or ever replied to me for example with the request you just made.
Ok, there is partially my fault, as I didn't filed bug against calligra at
this time but still it was 16 months ago.
So, for now my position on having 2 libpqxx versions in Debian is as follow
(unless you convince me otherwise):
I'm going to maintain only libpqxx4 in Debian for now, as for me uploading
libpqxx3 for 3-4 months and then removing it is just additional work for for my
sponsor, ftp and release teams and me of course and I prefer to avoid this as
for now much as I can.
I'm happy to reconsider re-uploading libpqxx3 but no earlier then 2 months
before freeze, if there will be risk that calligra with libpq support won't make
to Jessie.
All above are my thoughts about this subject but they are not set in stone, so
if you have anything to add which may make me change my mind just please let me
know, I'm really opened for other solutions if they exist.
I'm not sure if I am in the position to suggest you what to do now but, if I
were you I would drop postgres support for this 3-4 months, inform users about
it and try push upstream to release new version with libpq support as fast as
they can or at least declare when are they going to do so.
--
|_|0|_| |
|_|_|0| Heghlu'Meh QaQ jajVam |
|0|0|0| kuLa - |
gpg --keyserver pgp.mit.edu --recv-keys 0x58C338B3
3DF1 A4DF C732 4688 38BC F121 6869 30DD 58C3 38B3
signature.asc
Description: Digital signature
