Every object in OpenERP is based on OSV i.e Object Service and this service implements full Object-Relational Mapping enabling developers not to worry for the simple SQL operations.

In OpenERP, we have such ORM methods that are very useful. Here are the few basic methods:

1. read(cr, uid, ids, fields=None, context=None): returns a list of dictionaries with field values.

1. write(cr, uid, ids, values, context=None): updates a record with provided values.
   1. for line in self.browse(cr, uid, ids, context=context):
   2. if set_total:
   3. self.pool.get(‘add.penalty’).write(cr, uid, [line.id], {‘name’: line.name})
2. search(cr, uid, values, args, offset=0, limit=None, order=None, context=None, count=False): returns list of ids (tuple) on search criteria
   1. seq_id = self.pool.get(‘ir.sequence’).search(cr, uid, [(‘name’,’=’,’Sales Journal’)])[0]
3. copy(cr, uid, id, defaults, context=None): duplicates a record with default values
   1. def copy(self, cr, uid, id, default=None, context=None):
   2. default = default or {}
   3. default.update({‘invoice_ids’ : []})
   4. return super(res_partner, self).copy(cr, uid, id, default, context)
4. unlink(cr, uid, values, context=None): deletes a record with given id or ids
   1. account_move_obj.unlink(cr, uid, move_ids, context=context)
5. create(cr, uid, values, context=None): creates a new record with given values and returns the newly created record id
   1. partner_id = self.create(cr, uid,
   2. { ‘name’: ‘XYZ’,
   3. ‘description’ : ‘Contact’,
   4. ‘phone’: 123456,
   5. })
6. browse(cr, uid, values, context=None): retrieves records with required fields, returns a list of records
   1. invoice = self.pool.get(‘account.invoice’)
   2. cr.execute(“select * from account_invoice where state = ‘open’ and payment_term > 0 and cash_credit=’Credit Memo'”)
   3. invoices = invoice.browse(cr, uid, map(lambda x: x[0], cr.fetchall()) )

For more you can visit OpenERP ORM Methods