* Cédric Krier: " Re: [tryton-dev] Test and dependecies" (Thu, 15 May 2014 13:06:29 +0200):
> On 15 May 12:50, Mathias Behrle wrote:
> > * Cédric Krier: " Re: [tryton-dev] Test and dependecies" (Wed, 14 May 2014
> > 16:06:49 +0200):
> >
> > > On 14 May 11:26, Mathias Behrle wrote:
> > > > The second issue is, that sooner or later there *can* be the case,
> > > > where a dependent module will add constraints breaking a test in the
> > > > former module. Just not implementing necessary constraints, because
> > > > they could break tests, can not be the goal.
> > >
> > > But such module will be just like account_stock_continental, a bad
> > > designed module.
> >
> > So you mean basically, that adding a requires on a field of a supered
> > model is generally bad design?
>
> Yes constraints are always bad.
I agree, that they should be avoided as much as possible, but the consequence
of your statement is to not use constraints at all. Is this correct?
--
Mathias Behrle
MBSolutions
Gilgenmatten 10 A
D-79114 Freiburg
Tel: +49(761)471023
Fax: +49(761)4770816
http://m9s.biz
UStIdNr: DE 142009020
PGP/GnuPG key availabable from any keyserver, ID: 0x8405BBF6
signature.asc
Description: PGP signature
