GnuCash AqBanking 3 port - Re: [PATCH] AqBanking version checks

[Andreas Köhler]

Hi Manfred,

I am sorry for answering that late and am going to be rather responsive
the following days :-)

Am Donnerstag, den 20.03.2008, 08:33 +0100 schrieb Manfred Usselmann:
> Hi,
> 
> Christian Stimming <stimming at tuhh.de> schrieb am Wed, 19 Mar 2008
> 22:16:26 +0100:
> 
> > Am Dienstag, 18. März 2008 08:21 schrieb Manfred Usselmann:
> > >
> > > I just wanted to ask if somebody is already working on the
> > > AqBanking 3 port when I read your mail.
> > 
> > Nobody is working on it yet.
> > 
> > > Since I have never done any development work for GnuGash and
> > > AqBanking and also have only limited time it will definitely take
> > > some time until it will be done. But I had the impression that
> > > until now nobody is working on it and it needs to be done sooner or
> > > later...
> > >
> > > But if you accept the challenge that would probably lead to an
> > > earlierer result since you have at least experience with GnuCash
> > > development in contrast to me.
> > >
> > > Would it make sense if I nevertheless start to familiarised myself
> > > with the sources? Would it be possible to share the workload?
> > 
> > It would make sense in any case. Even if Andreas implemented the
> > majority of source code changes, we would still need lots of testing:
> > All the different security media have to be tested 1. when being
> > upgraded from aqbanking2, and also 2. when being newly set up. 
> 
> OK, makes sense.

That is really cool!  Any help is greatly appreciated.  I guess my
motivation was like yours: "someone has to do that someday".

> > I'd suggest to keep the wiki page as up to date as possible, i.e.
> > adding more tasks once they have been identified, and adding
> > descriptions of what has been found out and maybe solved so far.
> 
> Could you please send me the link to this wiki page again? 

http://wiki.gnucash.org/wiki/AqBanking3_Porting

Currently I wonder about two things I cannot decide yet, namely whether
it will make sense to create a branch to work on the port and, if we do,
to create an import-export module "aqbanking".  From my naive point of
view, this will

* avoid a lot of #if's
* has a better chance not to break aqbanking2
* might be easier to backport
* finally renames hbci which is not hbci only anyway
but also
* copy a lot of code and
* maybe be harder to maintain afterwards

I suppose this simply depends on how big the changes between aq2/gwen2
and aq3/gwen3 programming actually are.

Maybe Christian has an idea :-)

-- andi5

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Dies ist ein digital signierter Nachrichtenteil
Url : http://lists.gnucash.org/pipermail/gnucash-devel/attachments/20080320/c2ed3fe2/attachment.bin