Discussion:
[Erp5-dev] [Erp5-report] ERP5 r24898: 2308 Tests, 1 Errors, 12 Failures
Rafael Monnerat
2008-12-16 03:08:15 UTC
Permalink
Hi,

can someone take a look in the followed tests (and fix if it's possible)?
testSQLCachedWorklist (1 failures)
test_02_related_key (testSQLCachedWorklist.TestSQLCachedWorklist)
I think this comes after changes in SimulationTool.py
testConversionInSimulation (5 failures)
test_01_destination_total_asset_price_on_accounting_lines (testConversionInSimulation.TestConversionInSimulation)
test_01_diverged_purchase_packing_list_source_total_asset_price (testConversionInSimulation.TestConversionInSimulation)
test_01_diverged_sale_packing_list_destination_total_asset_price (testConversionInSimulation.TestConversionInSimulation)
test_01_simulation_movement_destination_asset_price (testConversionInSimulation.TestConversionInSimulation)
test_01_simulation_movement_source_asset_price (testConversionInSimulation.TestConversionInSimulation)
------------------------------------------------------------------------
_______________________________________________
Erp5-report mailing list
Erp5-report at erp5.org
http://mail.nexedi.com/mailman/listinfo/erp5-report
--
Rafael Manh?es Monnerat, Nexedi Developer - Tel. +55 (22) 8808 8500
ERP5 Enterprise: Free / Open Source ERP for Critical Applications
http://www.erp5.com
ERP5 Express: Hosted Open Source ERP for small companies
http://www.myerp5.com
Nexedi: Consulting and Development of Free / Open Source Software
http://www.nexedi.com
Robin Sébastien
2008-12-16 08:05:36 UTC
Permalink
Post by Rafael Monnerat
Hi,
can someone take a look in the followed tests (and fix if it's possible)?
testSQLCachedWorklist (1 failures)
test_02_related_key (testSQLCachedWorklist.TestSQLCachedWorklist)
This comes from a new feature and test added by Julien : support of related
keys in worklist, this was added for testWorklist, and by inheritance it is
also tested in testSQLCachedWorklist. Unfortunately, this is currently working
only for usual worklist, and it not implemented yet with optimisation on
worklists tested by testSQLCachedWorklist.

So there is no regression, but more work will be required in the future to
make sure to make it also working with cached worklist.

Regards.

S?bastien.
Post by Rafael Monnerat
I think this comes after changes in SimulationTool.py
testConversionInSimulation (5 failures)
test_01_destination_total_asset_price_on_accounting_lines
(testConversionInSimulation.TestConversionInSimulation)
test_01_diverged_purchase_packing_list_source_total_asset_price
(testConversionInSimulation.TestConversionInSimulation)
test_01_diverged_sale_packing_list_destination_total_asset_price
(testConversionInSimulation.TestConversionInSimulation)
test_01_simulation_movement_destination_asset_price
(testConversionInSimulation.TestConversionInSimulation)
test_01_simulation_movement_source_asset_price
(testConversionInSimulation.TestConversionInSimulation)
------------------------------------------------------------------------
_______________________________________________
Erp5-report mailing list
Erp5-report at erp5.org
http://mail.nexedi.com/mailman/listinfo/erp5-report
Vincent Pelletier
2008-12-16 08:50:53 UTC
Permalink
Post by Robin Sébastien
Post by Rafael Monnerat
Hi,
can someone take a look in the followed tests (and fix if it's possible)?
testSQLCachedWorklist (1 failures)
test_02_related_key (testSQLCachedWorklist.TestSQLCachedWorklist)
This comes from a new feature and test added by Julien : support of related
keys in worklist, this was added for testWorklist, and by inheritance it is
also tested in testSQLCachedWorklist. Unfortunately, this is currently
working only for usual worklist, and it not implemented yet with
optimisation on worklists tested by testSQLCachedWorklist.
So there is no regression, but more work will be required in the future to
make sure to make it also working with cached worklist.
One additional note: it should *not* be implemented without spending some time
to rething how columns are handled in cached worklists.
Currently, they are hardcoded in 2 places (a "create table" zsql method, and
the product). To implement related key support, we would have to extend
existing table column list with the list of all related keys (including
generated ones) to make it "work" in a generic instance.
What should be done is making those column lists computed so that table schema
changes dynamicaly to match worklist needs (or at the very least, provide a
central script for user to specify the list of column he needs).
--
Vincent Pelletier
Boris Kocherov
2008-12-16 13:29:25 UTC
Permalink
Post by Vincent Pelletier
One additional note: it should *not* be implemented without spending some time
to rething how columns are handled in cached worklists.
Currently, they are hardcoded in 2 places (a "create table" zsql method, and
the product). To implement related key support, we would have to extend
existing table column list with the list of all related keys (including
generated ones) to make it "work" in a generic instance.
What should be done is making those column lists computed so that table schema
changes dynamicaly to match worklist needs (or at the very least, provide a
central script for user to specify the list of column he needs).
What do you think about using of object relational mapping libraries
to resolve this problem? May be, http://www.sqlalchemy.org seems good
for this?
Sqlalchemy allows automatic schema migration
http://code.google.com/p/sqlalchemy-migrate/.
--
Always yours, Boris Kocherov
A specter is haunting Europe -- the specter of Communism.
Vincent Pelletier
2008-12-17 08:39:42 UTC
Permalink
Post by Boris Kocherov
What do you think about using of object relational mapping libraries
to resolve this problem? May be, http://www.sqlalchemy.org seems good
for this?
I'm not sure we want to add a dependancy on this just for worklist caching.
Post by Boris Kocherov
Sqlalchemy allows automatic schema migration
http://code.google.com/p/sqlalchemy-migrate/.
There is actualy no migration required: worklist cache table is dropped at
each refresh anyway (faster than "delete from ..."), so what is needed is
merely a way to retrieve what is needed to generate a "create table" query
dynamicaly.
--
Vincent Pelletier
Vincent Pelletier
2008-12-16 08:53:01 UTC
Permalink
Post by Rafael Monnerat
I think this comes after changes in SimulationTool.py
testConversionInSimulation (5 failures)
test_01_destination_total_asset_price_on_accounting_lines
(testConversionInSimulation.TestConversionInSimulation)
test_01_diverged_purchase_packing_list_source_total_asset_price
(testConversionInSimulation.TestConversionInSimulation)
test_01_diverged_sale_packing_list_destination_total_asset_price
(testConversionInSimulation.TestConversionInSimulation)
test_01_simulation_movement_destination_asset_price
(testConversionInSimulation.TestConversionInSimulation)
test_01_simulation_movement_source_asset_price
(testConversionInSimulation.TestConversionInSimulation)
I'll check my recent modifications.
--
Vincent Pelletier
Loading...