Merge lp:~therp-nl/ocb-server/7.0-lp1194893-export_xmlid_wrong_table_name into lp:ocb-server

[Stefan Rijnhart (Opener)]

[Stefan Rijnhart (Opener) (stefan-opener)]

Before you ask, line 9 is unintentional. Since r._table_name is in fact equivalent to r._model._name, it does not constitute a functional change. But I would like to leave it there to resolve the confusion that made me put it there in the first place.

[Nicolas Bessi - Camptocamp (nbessi-c2c-deactivatedaccount)]

Hello,
Thanks for the MP.

line 18 is there a reason why use use _model._table instead of _model._name ?
It also appear that _table is deprecated:

        self._table = table # deprecated, use _model!

We may also fix spacing, or are you afraid of merge conflict?

Regards

Nicolas

review: Needs Information (no test, code review)

[Stefan Rijnhart (Opener) (stefan-opener)]

Why use table and not model name? This MP does not change anything about that, but I am guessing that the reason is that XML-IDs are not supposed to have more than one dot. Having XML-IDs like '__export__.res.partner_169' could be confusing.

Using browse_record._table is deprecated, that is why I am using browse_record._model. Using Model._table is not deprecated ;-)

[Nicolas Bessi - Camptocamp (nbessi-c2c-deactivatedaccount)]

Ok then LGTM

review: Approve (no test code review only)

[Holger Brunn (Therp) (hbrunn)]

review: Approve (code review)

=== modified file 'openerp/osv/orm.py'
--- openerp/osv/orm.py	2013-06-20 13:10:57 +0000
+++ openerp/osv/orm.py	2013-06-26 14:16:36 +0000
@@ -1106,7 +1106,7 @@
 
         def _get_xml_id(self, cr, uid, r):
             model_data = self.pool.get('ir.model.data')
-            data_ids = model_data.search(cr, uid, [('model', '=', r._table_name), ('res_id', '=', r['id'])])
+            data_ids = model_data.search(cr, uid, [('model', '=', r._model._name), ('res_id', '=', r['id'])])
             if len(data_ids):
                 d = model_data.read(cr, uid, data_ids, ['name', 'module'])[0]
                 if d['module']:
@@ -1116,13 +1116,13 @@
             else:
                 postfix = 0
                 while True:
-                    n = self._table+'_'+str(r['id']) + (postfix and ('_'+str(postfix)) or '' )
+                    n = r._model._table+'_'+str(r['id']) + (postfix and ('_'+str(postfix)) or '' )
                     if not model_data.search(cr, uid, [('name', '=', n)]):
                         break
                     postfix += 1
                 model_data.create(cr, SUPERUSER_ID, {
                     'name': n,
-                    'model': self._name,
+                    'model': r._model._name,
                     'res_id': r['id'],
                     'module': '__export__',
                 })