You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
155 lines
5.8 KiB
155 lines
5.8 KiB
# -*- coding: utf-8 -*-
|
|
##############################################################################
|
|
#
|
|
# OpenERP, Open Source Management Solution
|
|
# This module copyright (C) 2014 Therp BV (<http://therp.nl>).
|
|
#
|
|
# This program is free software: you can redistribute it and/or modify
|
|
# it under the terms of the GNU Affero General Public License as
|
|
# published by the Free Software Foundation, either version 3 of the
|
|
# License, or (at your option) any later version.
|
|
#
|
|
# This program is distributed in the hope that it will be useful,
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
# GNU Affero General Public License for more details.
|
|
#
|
|
# You should have received a copy of the GNU Affero General Public License
|
|
# along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
#
|
|
##############################################################################
|
|
|
|
from openerp.osv import orm, fields
|
|
from openerp.tools.translate import _
|
|
|
|
|
|
class CleanupPurgeLineColumn(orm.TransientModel):
|
|
_inherit = 'cleanup.purge.line'
|
|
_name = 'cleanup.purge.line.column'
|
|
|
|
_columns = {
|
|
'model_id': fields.many2one(
|
|
'ir.model', 'Model',
|
|
required=True, ondelete='CASCADE'),
|
|
'wizard_id': fields.many2one(
|
|
'cleanup.purge.wizard.column', 'Purge Wizard', readonly=True),
|
|
}
|
|
|
|
def purge(self, cr, uid, ids, context=None):
|
|
"""
|
|
Unlink columns upon manual confirmation.
|
|
"""
|
|
for line in self.browse(cr, uid, ids, context=context):
|
|
if line.purged:
|
|
continue
|
|
|
|
model_pool = self.pool[line.model_id.model]
|
|
|
|
# Check whether the column actually still exists.
|
|
# Inheritance such as stock.picking.in from stock.picking
|
|
# can lead to double attempts at removal
|
|
cr.execute(
|
|
'SELECT count(attname) FROM pg_attribute '
|
|
'WHERE attrelid = '
|
|
'( SELECT oid FROM pg_class WHERE relname = %s ) '
|
|
'AND attname = %s',
|
|
(model_pool._table, line.name))
|
|
if not cr.fetchone()[0]:
|
|
continue
|
|
|
|
self.logger.info(
|
|
'Dropping column %s from table %s',
|
|
line.name, model_pool._table)
|
|
cr.execute(
|
|
"""
|
|
ALTER TABLE "%s" DROP COLUMN "%s"
|
|
""" % (model_pool._table, line.name))
|
|
line.write({'purged': True})
|
|
cr.commit()
|
|
return True
|
|
|
|
|
|
class CleanupPurgeWizardColumn(orm.TransientModel):
|
|
_inherit = 'cleanup.purge.wizard'
|
|
_name = 'cleanup.purge.wizard.column'
|
|
|
|
# List of known columns in use without corresponding fields
|
|
# Format: {table: [fields]}
|
|
blacklist = {
|
|
'wkf_instance': ['uid'], # lp:1277899
|
|
}
|
|
|
|
def default_get(self, cr, uid, fields, context=None):
|
|
res = super(CleanupPurgeWizardColumn, self).default_get(
|
|
cr, uid, fields, context=context)
|
|
if 'name' in fields:
|
|
res['name'] = _('Purge columns')
|
|
return res
|
|
|
|
def get_orphaned_columns(self, cr, uid, model_pools, context=None):
|
|
"""
|
|
From openobject-server/openerp/osv/orm.py
|
|
Iterate on the database columns to identify columns
|
|
of fields which have been removed
|
|
"""
|
|
|
|
columns = list(set([
|
|
column for model_pool in model_pools
|
|
for column in model_pool._columns
|
|
if not (isinstance(model_pool._columns[column],
|
|
fields.function) and
|
|
not model_pool._columns[column].store)
|
|
]))
|
|
columns += orm.MAGIC_COLUMNS
|
|
columns += self.blacklist.get(model_pools[0]._table, [])
|
|
|
|
cr.execute("SELECT a.attname"
|
|
" FROM pg_class c, pg_attribute a"
|
|
" WHERE c.relname=%s"
|
|
" AND c.oid=a.attrelid"
|
|
" AND a.attisdropped=%s"
|
|
" AND pg_catalog.format_type(a.atttypid, a.atttypmod)"
|
|
" NOT IN ('cid', 'tid', 'oid', 'xid')"
|
|
" AND a.attname NOT IN %s",
|
|
(model_pools[0]._table, False, tuple(columns))),
|
|
return [column[0] for column in cr.fetchall()]
|
|
|
|
def find(self, cr, uid, context=None):
|
|
"""
|
|
Search for columns that are not in the corresponding model.
|
|
|
|
Group models by table to prevent false positives for columns
|
|
that are only in some of the models sharing the same table.
|
|
Example of this is 'sale_id' not being a field of stock.picking.in
|
|
"""
|
|
res = []
|
|
model_pool = self.pool['ir.model']
|
|
model_ids = model_pool.search(cr, uid, [], context=context)
|
|
|
|
# mapping of tables to tuples (model id, [pool1, pool2, ...])
|
|
table2model = {}
|
|
|
|
for model in model_pool.browse(cr, uid, model_ids, context=context):
|
|
model_pool = self.pool.get(model.model)
|
|
if not model_pool or not model_pool._auto:
|
|
continue
|
|
table2model.setdefault(
|
|
model_pool._table, (model.id, []))[1].append(model_pool)
|
|
|
|
for table, model_spec in table2model.iteritems():
|
|
for column in self.get_orphaned_columns(
|
|
cr, uid, model_spec[1], context=context):
|
|
res.append((0, 0, {
|
|
'name': column,
|
|
'model_id': model_spec[0]}))
|
|
if not res:
|
|
raise orm.except_orm(
|
|
_('Nothing to do'),
|
|
_('No orphaned columns found'))
|
|
return res
|
|
|
|
_columns = {
|
|
'purge_line_ids': fields.one2many(
|
|
'cleanup.purge.line.column',
|
|
'wizard_id', 'Columns to purge'),
|
|
}
|