[Koha-devel] Issues with DBIx and nested transactions

[Kyle Hall]

Agreed! The next step I think is to move ( add ) caching from C4::Context
into Koha::Config::SysPrefs thus removing all dependency on those
C4::Context subs. After that we begin replacing the use of
C4::Context->preference with Koha::Config::SysPrefs. Does that make sense?
I wonder if we can implement Koha::Cache for syspref caching now. I'm not
up to date with it's state.

Kyle

http://www.kylehall.info
ByWater Solutions ( http://bywatersolutions.com )
Meadville Public Library ( http://www.meadvillelibrary.org )
Crawford County Federated Library System ( http://www.ccfls.org )
Mill Run Technology Solutions ( http://millruntech.com )

On Wed, Jun 10, 2015 at 10:23 AM, Tomas Cohen Arazi <tomascohen at gmail.com>
wrote:

> I think the natural next step is to just remove set_preference and
> preference from C4::Context.
>
> 2015-06-10 11:17 GMT-03:00 Fridolin SOMERS <fridolin.somers at biblibre.com>:
>
>> I noticed the use of Koha::Config::SysPrefs in C4::Context introduces
>> circual dependences :(
>> C4::Context > Koha::Config::SysPrefs > Koha::Database > C4::Context
>>
>> You confirme ?
>>
>> Maybe we should create a new module C4::Config with all methods not
>> directly linked to the context.
>>
>> Regards
>>
>> Le 04/06/2015 19:13, Tomas Cohen Arazi a écrit :
>>
>>> I've just pushed bug 13967 [1]. It introduces a new package
>>> (Koha::Config::SysPrefs) that heavily relies on DBIx::Class to deal with
>>> the required CRUD for system preferences.
>>>
>>> It should be no surprise that the first widespread use of DBIx on Koha's
>>> codebase raises problems. Widespread means every call to anything is
>>> using
>>> it :-D.
>>>
>>> The point: it noticeably broke lots of our t/db_dependent/ tests. A first
>>> guess would point to the package itself. But such doesn't seem to be the
>>> case. It is highlighting an underlaying problem in how we deal with the
>>> DB:
>>> the errors are related to nested transactions taking place, so I would
>>> point to our (abuse?) of $dbh->{AutoCommit} = 0 on the tests.
>>>
>>> Actually, if you pick any db_dependent test that is currently failing on
>>> master and set the AutoCommit = 1 you will notice the tests work as
>>> expected (with the side effect of altering the DB of course).
>>>
>>> So, this is a call for all of you, whenever you find some time, to try to
>>> figure what's going on and possible solutions to this. I ain't no DBIx
>>> expert and will be doing my best but really need your help.
>>>
>>> Before someone says it: reverting the commits will put us back to the
>>> previous state, of course. But it will just hide the problem for a while.
>>>
>>> [1] http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=13967
>>>
>>>
>>>
>>> _______________________________________________
>>> Koha-devel mailing list
>>> Koha-devel at lists.koha-community.org
>>> http://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-devel
>>> website : http://www.koha-community.org/
>>> git : http://git.koha-community.org/
>>> bugs : http://bugs.koha-community.org/
>>>
>>>
>> --
>> Fridolin SOMERS
>> Biblibre - Pôles support et système
>> fridolin.somers at biblibre.com
>> _______________________________________________
>> Koha-devel mailing list
>> Koha-devel at lists.koha-community.org
>> http://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-devel
>> website : http://www.koha-community.org/
>> git : http://git.koha-community.org/
>> bugs : http://bugs.koha-community.org/
>>
>
>
>
> --
> Tomás Cohen Arazi
> Prosecretaría de Informática
> Universidad Nacional de Córdoba
> ✆ +54 351 5353750 ext 13168
> GPG: B76C 6E7C 2D80 551A C765  E225 0A27 2EA1 B2F3 C15F
>
> _______________________________________________
> Koha-devel mailing list
> Koha-devel at lists.koha-community.org
> http://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-devel
> website : http://www.koha-community.org/
> git : http://git.koha-community.org/
> bugs : http://bugs.koha-community.org/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.koha-community.org/pipermail/koha-devel/attachments/20150612/4cc445dc/attachment.html>