Quoting column names make the case-sensitive in PostgreSQL,
and this is the default strategy we are using so far. It is
important to be consistent there.
We might want to do the same to table names too.
This allows create fields with mixed cases names for example.
bzr revid: odo@openerp.com-20110919201845-heer0rttcouvtc9x
In SQL, not in (2, 7, 9) will return only non-null records.
In OpenERP, a domain with not in [2, 7, 9] should return
every record whose value is not 2 or 7 or 9, including
the one with a null value.
This means that the union of (not in XXX) and (in XXX)
will return the whole set.
bzr revid: vmt@openerp.com-20110810130646-2y822a276igz2f1g
The newly added tests were wrong before the fix.
The change might break existing code if it was relying
on this broken behavior.
This also starts a better separation of concern between
parse() and __leaf_to_sql.
bzr revid: vmt@openerp.com-20110810085604-1f6ahwzuzfklj1b7
- it does two unrelated things
- splitting it makes visible that ids2 and operator are no more used in some calls
- _rec_convert now behaves differently but it probably cannot be called with an empty ids arg anyway.
bzr revid: vmt@openerp.com-20110726142244-54ty1gn7jxxz3cwi
This adds a normalize_domain() method to osv.expression, which ir.rule calls
before combining expressions due to multiple rules or groups being applicable
to a certain user.
YAML tests are also added with a trivial unit test of normalize_domain() and
some additional tests for ir_rule in order to verify that unnormalized domains
are properly normalized before being combined by ir.rule.
bzr revid: odo@openerp.com-20110324111757-uwuoqvm3lxkipr08
- Some logging code moved from netsvc.py to loglevels.py
- Changed imports to use the new openerp module
- config and netsvc initialization calls move to openerp-server.py
- Moved openerp-server.py outside the old bin directory
- Some imports in tools moved inside the methods to break mutual-dependencies
bzr revid: vmt@openerp.com-20110207125723-ooee7d7ng5elmkso