File Format Decision

[John Ralls]

> On Sep 21, 2016, at 7:57 PM, Fross, Michael <michael at fross.org> wrote:
> 
> On Wed, Sep 21, 2016 at 11:58 AM, Geert Janssens <geert.gnucash at kobaltwit.be
>> wrote:
> 
>> don't know of any other strong benefits of the sql ba
> 
> 
> Thanks Geert. That is very helpful.  I expected SQL to be faster with
> larger datasets, but if it's all just read into memory, then I can see why
> that is not a benefit.
> 
> I'll stick with XML and look forward to the data-centric release that will
> come out in a bit over a year (if the interview I read remains the case) as
> I expect this will change the landscape considerably.
> 
> What a great community.  Thank you.

I think you mean database-centric, and I don't think it's going to make it for the next major release. I'm getting close to finishing the backend C++ rewrite; next is to rewrite the query code to wrap SQL so that it queries the database instead of the in-memory objects and modify the XML backend so that it creates an in-memory SQL database that the new query module can work with. Then all of the code that expects everything to be in memory has to change to query for the objects it needs and change the GUI to work with that model. We're less than a year away from feature freeze for 2.8 and I'm the only one working on this at present.

Regards,
John Ralls