Wildcard records cannot be added, when at least one is already created

[Pavel Kirpichyov]

Scenario 1:
create domain: test.com.
create record: *.test.com.
create record: mail.test.com.

Scenario 2:
create domain: test.com.
create record: mail.test.com.
create record: *.test.com.

Scenario 2 will failed at step 3, becouse find_recordset method will find previous record recordset (mail.test.com) instead of new one. The reason for this behavior is feature in sqlalchemy storage:
https://github.com/openstack/designate/blob/bc17db433cbb3f3d3c6135db5d747d5a7b0e1c94/designate/sqlalchemy/base.py#L100
which replaces * for %, but here it's wrong behavior

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/166806

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/166806
Committed: https://git.openstack.org/cgit/openstack/designate/commit/?id=e59271d97f82d452992337f9deff109f79e7f75f
Submitter: Jenkins
Branch: master

commit e59271d97f82d452992337f9deff109f79e7f75f
Author: Endre Karlson <email address hidden>
Date: Mon Mar 23 13:52:17 2015 +0100

    Fix handling of wildcards in api / storage

    We need to move the handling of '*' to '%' up to the API layer since if not it
    causes a issue described in the bug. This change addresses this by removing the
    replace of '*' to '%' in the storage layer to the API and also fixes any other
    places '*' is used for wildcard search.

    Closes-Bug: #1391426

    Change-Id: Ib75e8359b8040effea3dfdbf9c2baf05c3ed208b