[Erp5-dev] Chaning values on packing list and its propagation into simulation - bug or feature?

[Romain Courteaud]

* ??ukasz Nowak [2007-11-02 13:42:55 +0100]:

> Hello,

Hello,

> I've played a lot with simulations related to accounting - simple ones
> (if invoice is causality of parent applied rule) and more complicated -
> when packing list or order are parents of applied rule.
> 
> I've spotted on bizzare problem.
> 
> Create Sale Packing List, add line to it. Confirm - applied rules and
> simulations will be created.
> 
> Check if _all_ related applied rules and simulation movements are
> correct.
> 
> Modify quantity on Sale Packing List Line, then packing list will
> become diverge, accept decision, solved.
> 
> Check if _all_ related applied rules and simulation movements are
> correct (==updated).
> 
> Now change start date on Sale Packing List. It will become diverged.
> Accept decision.
> 
> Check if _all_ related applied rules and simulation movements are
> correct (==updated). And this check is not passed - start date on
> simulation movement, which parent's specialise is
> 'default_invoicing_rule' is not modified.
> 
> And implication - invoice will be created with "wrong" date - old one,
> not which is on packing list (same applies to accounting transaction
> lines).
> 
> Attached unit test (hope enough Nexedish) is following above scenario
> (with addition, that packing list is generated by order).
> 
> My questions:
>  - even if I run this unit test on revision 17359 *please* check me
> twice - I'm famous with simple mistakes ;)
>  - is such behaviour correct?
>  - is such situation unit tested anywhere? (in previous email there
> was another problem - modification of dates on invoices)
>  - is it a bug or feature?
> 
> If it is a bug (and I didn't make another simple mistake) I'd be glad
> if it could be confirmed.

Seems to be a bug for me.
I think I fix it in http://svn.erp5.org/?view=rev&revision=17391.

Can you confirm that if fixes your bug ?

Regards,
Romain