[Kde-pim] Review Request 112975: Set conforming string for postgreSQL

Dan Vrátil dvratil at redhat.com
Mon Sep 30 14:42:05 BST 2013



> On Sept. 29, 2013, 5:27 p.m., Cédric  Villemain wrote:
> > server/src/storage/dbconfigpostgresql.cpp, line 164
> > <http://git.reviewboard.kde.org/r/112975/diff/3/?file=192961#file192961line164>
> >
> >     locale should be based on the current sytem setting.
> >     
> >     --text-search-config=CFG might be set accordingly if akonadi expect to benefit from PostgreSQL full text (text vector, stemmers, ispell, ...)

We don't use fulltext, all string matches are only index-based, so that should not be a problem.

Since only Akonadi server is talking to the database, and AS is using UTF8 too, it makes sense to ignore system locale and enforce UTF8 I think.


> On Sept. 29, 2013, 5:27 p.m., Cédric  Villemain wrote:
> > server/src/storage/dbconfigpostgresql.cpp, line 197
> > <http://git.reviewboard.kde.org/r/112975/diff/3/?file=192961#file192961line197>
> >
> >     Extra argument can be added here to provide unix_socket_directory and maybe listen_adresses if it is preferable to keep setting it.

I think listen_address has been disabled for "security reasons". Yes, settings it as command line argument instead of changing config file would be preferable. 


> On Sept. 29, 2013, 5:27 p.m., Cédric  Villemain wrote:
> > server/src/storage/dbconfigpostgresql.cpp, line 283
> > <http://git.reviewboard.kde.org/r/112975/diff/3/?file=192961#file192961line283>
> >
> >     it is way better to try a 
> >     "stop -m immediate"
> >     before having to go to SIGABRT (and it works on windows too, while I'm unsure about 'kill')
> >

Ok, please use "stop -m immediate" instead of kill.


> On Sept. 29, 2013, 5:27 p.m., Cédric  Villemain wrote:
> > server/src/storage/dbconfigpostgresql.cpp, line 194
> > <http://git.reviewboard.kde.org/r/112975/diff/3/?file=192961#file192961line194>
> >
> >     -t10 is ok.
> >     *but* if startup takes more time it will abort. It can take more time if there are pending changes in the WriteAheadLog that must be applyed before opening services. This can be many IO and fsync() and may take longer (but will be surprising)

-t20 would probably still be fine.


- Dan


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/112975/#review41000
-----------------------------------------------------------


On Sept. 28, 2013, 4:40 p.m., Cédric  Villemain wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/112975/
> -----------------------------------------------------------
> 
> (Updated Sept. 28, 2013, 4:40 p.m.)
> 
> 
> Review request for Akonadi.
> 
> 
> Repository: akonadi
> 
> 
> Description
> -------
> 
> So the input/output are as expected.
> 
> 
> Diffs
> -----
> 
>   server/src/storage/dbconfigpostgresql.cpp acc65af149d0d4849dca535e9d0599e234befab2 
> 
> Diff: http://git.reviewboard.kde.org/r/112975/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Cédric  Villemain
> 
>

_______________________________________________
KDE PIM mailing list kde-pim at kde.org
https://mail.kde.org/mailman/listinfo/kde-pim
KDE PIM home page at http://pim.kde.org/


More information about the kde-pim mailing list