Hi,

for the answer just create a file.sql with your requests inside, exemple :

DELETE FROM ir_ui_view_sc WHERE user_id = 1 and name = 'Clients';

add in your module_name/__openerp__.py :

in section

'update_xml': ["file.sql",],

for v7 :

'data': ["file.sql",],

use insert , update ... Bye

You can create a method init(self, cr) in your model and it will be called upon module installation or upgrade.

You can check for update or install mode by checking the the existence of data or structures that should have been created in initialization or you can just use common sql or replace mechanism to use the same sql for both states.

Most reports use this method to create required views. Check this example from addons/account/report/account_report:

class report_account_receivable(osv.osv):
    _name = "report.account.receivable"
    _description = "Receivable accounts"
    _auto = False
    _columns = {
        'name': fields.char('Week of Year', size=7, readonly=True),
        'type': fields.selection(_code_get, 'Account Type', required=True),
        'balance':fields.float('Balance', readonly=True),
        'debit':fields.float('Debit', readonly=True),
        'credit':fields.float('Credit', readonly=True),
    }
    _order = 'name desc'

    def init(self, cr):
        tools.drop_view_if_exists(cr, 'report_account_receivable')
        cr.execute("""
            create or replace view report_account_receivable as (
                select
                    min(l.id) as id,
                    to_char(date,'YYYY:IW') as name,
                    sum(l.debit-l.credit) as balance,
                    sum(l.debit) as debit,
                    sum(l.credit) as credit,
                    a.type
                from
                    account_move_line l
                left join
                    account_account a on (l.account_id=a.id)
                where
                    l.state <> 'draft'
                group by
                    to_char(date,'YYYY:IW'), a.type
            )""")

As long as we write less SQL sentences, the ORM can make more optimizations for us in the future and the maintenance will be easier.

There are cases where you have made strong changes to your models already used in production and the ORM can't make the automatic inference of the sql statements needed to give the database a logical state corresponding to your new code (ex: fix your table structure, move data between columns, calculate quickly default values for new not null columns using sql expressions, etc..).If your case is one of those data maintenance problems, you can use the migration scripts as specified in the MigrationManager descriptions: http://bit.ly/1UqiO8h.This mechanism is used extensively inside openupgrade: http://bit.ly/1QDc40R  You can find a lot of samples there.

In your case, just create a migration folder with a -pre or -post update script:

            <moduledir>
            `-- migrations
                |-- 8.0
                |   |-- pre-migration.py

In the  pre-migration.py script you can check the version and put both queries. The first two digits passed to the script is always the odoo version the others come from your module version ( __openerp__.py of your module)

You can install the module specifying one version 1.1 and then change the module version to 1.2, 1.3, etc... in

def migrate(cr, version):
    if version == '8.0.1.1':
         cr.execute("""UPDATE ... FROM account_analytic_journal WHERE .... """)
    else
         cr.execute(anothersql)