odoo
/
odoo
2
0
Fork 0
Code Pull Requests Releases Activity

[REF] ir_property: changed 'value_reference' and 'res_id' fields into char instead of reference fields because the code was fuzzy, the performances very bad (at least since 'view_load' attribute has been removed), and because there were no reason to have them as references (even from an useability point of view) 

bzr revid: qdp-launchpad@openerp.com-20130624085049-elmhvzkufntlzbf8
This commit is contained in:
Quentin (OpenERP) 2013-06-24 10:50:49 +02:00
parent de8272eb5f
commit 601ee8e030
1 changed files with 7 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
openerp/addons/base/res/ir_property.py
View File

@ -22,7 +22,7 @@
import time
from openerp.osv import osv, fields
from openerp.osv.orm import browse_record
from openerp.osv.orm import browse_record, browse_null
from openerp.tools.misc import attrgetter
# -------------------------------------------------------------------------
@ -32,27 +32,10 @@ from openerp.tools.misc import attrgetter
class ir_property(osv.osv):
_name = 'ir.property'
def _models_field_get(self, cr, uid, field_key, field_value, context=None):
get = attrgetter(field_key, field_value)
obj = self.pool.get('ir.model.fields')
ids = obj.search(cr, uid, [], context=context)
res = set()
for o in obj.browse(cr, uid, ids, context=context):
res.add(get(o))
return list(res)
def _models_get(self, cr, uid, context=None):
return self._models_field_get(cr, uid, 'model', 'model_id.name', context)
def _models_get2(self, cr, uid, context=None):
return self._models_field_get(cr, uid, 'relation', 'relation', context)
_columns = {
'name': fields.char('Name', size=128, select=1),
'res_id': fields.reference('Resource', selection=_models_get, size=128,
help="If not set, acts as a default value for new resources", select=1),
'res_id': fields.char('Resource', size=128, help="If not set, acts as a default value for new resources", select=1),
'company_id': fields.many2one('res.company', 'Company', select=1),
'fields_id': fields.many2one('ir.model.fields', 'Field', ondelete='cascade', required=True, select=1),
@ -60,7 +43,7 @@ class ir_property(osv.osv):
'value_integer' : fields.integer('Value'),
'value_text' : fields.text('Value'), # will contain (char, text)
'value_binary' : fields.binary('Value'),
'value_reference': fields.reference('Value', selection=_models_get2, size=128),
'value_reference': fields.char('Value', size=128),
'value_datetime' : fields.datetime('Value'),
'type' : fields.selection([('char', 'Char'),
@ -128,7 +111,6 @@ class ir_property(osv.osv):
values[field] = value
return values
def write(self, cr, uid, ids, values, context=None):
return super(ir_property, self).write(cr, uid, ids, self._update_values(cr, uid, ids, values), context=context)
@ -147,7 +129,10 @@ class ir_property(osv.osv):
elif record.type == 'binary':
return record.value_binary
elif record.type == 'many2one':
return record.value_reference
if not record.value_reference:
return browse_null()
model, resource_id = record.value_reference.split(',')
return self.pool.get(model).browse(cr, uid, int(resource_id), context=context)
elif record.type == 'datetime':
return record.value_datetime
elif record.type == 'date':