This is very helpful when a warehouse is splitted into multiple
stock points.

Also avoid returning no stock optimisations if no date could be
found in future.

Up to know, order builders were mostly used having in mind the idea of
deleting all previously created automated orders.

This way was mostly requesting to run generated delivery builder only
once per night, assuming there is no users working at that time.

So to allows running very often (many times a day) generated delivery
builders, allow them to update existing deliveries. This way :
- automated deliveries could be much more up to date
- there is no need to delete everything
- this generate few activities, there is no need of a long calculation
  to update everything

Also, the stock optimization was using dates with a precision of one
full day. Generic code must support much better precision.

So all newSimulation expected failures are removed. Make
generateMovementListForStockOptimisation looking min_flow and max_delay
on supply lines.

Introduce a getNextAlertInventoryDate in addition to
getNextNegativeInventoryDate. This allows to know when an inventory
will be below a reference quantity. This is particularly helpful
to know when an inventory is below the minimal admitted stock

So it's no longer a newSimulation expected failure. Review method
generateMovementListForStockOptimisation of builders to nicely
generated temporary movements depending on future negative stocks.

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@41628 20353a03-c40f-0410-a6d1-a30d3c3de9de

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@41557 20353a03-c40f-0410-a6d1-a30d3c3de9de

Clean up of activities is already handled by ERP5TypeTestCase.tearDown

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@39997 20353a03-c40f-0410-a6d1-a30d3c3de9de

(locating in the second line only works when the first line is shebang comment.)


git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@31447 20353a03-c40f-0410-a6d1-a30d3c3de9de

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@30693 20353a03-c40f-0410-a6d1-a30d3c3de9de

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@29606 20353a03-c40f-0410-a6d1-a30d3c3de9de

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@27956 20353a03-c40f-0410-a6d1-a30d3c3de9de

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@26926 20353a03-c40f-0410-a6d1-a30d3c3de9de

(DateTime().earliestTime() + gap).earliestTime(), since
DateTime().__add__() returns a different tz offset if daylight saving
changes.

For example (timezone is Europe/Paris):

>>> DateTime().earliestTime()
DateTime('2008/10/24')
>>> DateTime().earliestTime()._tz
'GMT+2'
>>> (DateTime().earliestTime()+10)
DateTime('2008/11/03')
>>> (DateTime().earliestTime()+10)._tz
'GMT+2'
>>> DateTime(str((DateTime().earliestTime()+10)))
DateTime('2008/11/03')
>>> DateTime(str((DateTime().earliestTime()+10)))._tz
'GMT+1'


git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@24312 20353a03-c40f-0410-a6d1-a30d3c3de9de

* Movement Group becomes ERP5 Document. See http://www.erp5.org/HowToConfigureBuilder for the detail.
* add _duplicate() in CopySupport.py. This method is similar to copy and paste, but it preserves workflow history.
* add a new target solver CopyAndPropagate.py that updates values according to passed divergence list and propagate them.


git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@23647 20353a03-c40f-0410-a6d1-a30d3c3de9de

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@23646 20353a03-c40f-0410-a6d1-a30d3c3de9de

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@22570 20353a03-c40f-0410-a6d1-a30d3c3de9de

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@22564 20353a03-c40f-0410-a6d1-a30d3c3de9de