diff options
author | wangxiyuan <wangxiyuan@huawei.com> | 2019-05-05 10:39:17 +0800 |
---|---|---|
committer | wangxiyuan <wangxiyuan1007@gmail.com> | 2019-07-05 06:59:58 +0000 |
commit | f43954be9419e96fbe9d231aa1f5ae32cd0142eb (patch) | |
tree | ae8212f647b40cc19fc1ae169ee71015b73d5784 | |
parent | 163f4e4a6e0b5ae9d1060a1779745fab949a6088 (diff) | |
download | keystone-f43954be9419e96fbe9d231aa1f5ae32cd0142eb.tar.gz |
Drop limit columns
The service_id, region_id and resource_name columns in limit table
is useless now. These info can be got by 'registered_limit_id'
column from registered limit table. So there is no data migration
needed.
Since this is a silent change for end users, there is no need for
a release note as well.
Related-Bug: #1777892
Change-Id: Ibf4aa81dad7c5ebbd5599a55237cc5a658223432
5 files changed, 104 insertions, 90 deletions
diff --git a/keystone/common/sql/contract_repo/versions/063_contract_drop_limit_columns.py b/keystone/common/sql/contract_repo/versions/063_contract_drop_limit_columns.py new file mode 100644 index 000000000..8858ba916 --- /dev/null +++ b/keystone/common/sql/contract_repo/versions/063_contract_drop_limit_columns.py @@ -0,0 +1,23 @@ +# Licensed under the Apache License, Version 2.0 (the "License"); you may +# not use this file except in compliance with the License. You may obtain +# a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT +# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the +# License for the specific language governing permissions and limitations +# under the License. + +import sqlalchemy as sql + + +def upgrade(migrate_engine): + meta = sql.MetaData() + meta.bind = migrate_engine + + limit_table = sql.Table('limit', meta, autoload=True) + limit_table.c.service_id.drop() + limit_table.c.region_id.drop() + limit_table.c.resource_name.drop() diff --git a/keystone/common/sql/data_migration_repo/versions/063_migrate_drop_limit_columns.py b/keystone/common/sql/data_migration_repo/versions/063_migrate_drop_limit_columns.py new file mode 100644 index 000000000..8aa15c1ef --- /dev/null +++ b/keystone/common/sql/data_migration_repo/versions/063_migrate_drop_limit_columns.py @@ -0,0 +1,15 @@ +# Licensed under the Apache License, Version 2.0 (the "License"); you may +# not use this file except in compliance with the License. You may obtain +# a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT +# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the +# License for the specific language governing permissions and limitations +# under the License. + + +def upgrade(migrate_engine): + pass diff --git a/keystone/common/sql/expand_repo/versions/063_expand_drop_limit_columns.py b/keystone/common/sql/expand_repo/versions/063_expand_drop_limit_columns.py new file mode 100644 index 000000000..8aa15c1ef --- /dev/null +++ b/keystone/common/sql/expand_repo/versions/063_expand_drop_limit_columns.py @@ -0,0 +1,15 @@ +# Licensed under the Apache License, Version 2.0 (the "License"); you may +# not use this file except in compliance with the License. You may obtain +# a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT +# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the +# License for the specific language governing permissions and limitations +# under the License. + + +def upgrade(migrate_engine): + pass diff --git a/keystone/limit/backends/sql.py b/keystone/limit/backends/sql.py index 3d32c8da7..b2ce6c13b 100644 --- a/keystone/limit/backends/sql.py +++ b/keystone/limit/backends/sql.py @@ -69,15 +69,10 @@ class LimitModel(sql.ModelBase, sql.ModelDictMixin): 'registered_limit_id' ] - # TODO(wxy): Drop "service_id", "region_id" and "resource_name" columns - # in T release. internal_id = sql.Column(sql.Integer, primary_key=True, nullable=False) id = sql.Column(sql.String(length=64), nullable=False, unique=True) project_id = sql.Column(sql.String(64)) domain_id = sql.Column(sql.String(64)) - _service_id = sql.Column('service_id', sql.String(255)) - _region_id = sql.Column('region_id', sql.String(64), nullable=True) - _resource_name = sql.Column('resource_name', sql.String(255)) resource_limit = sql.Column(sql.Integer, nullable=False) description = sql.Column(sql.Text()) registered_limit_id = sql.Column(sql.String(64), @@ -89,29 +84,21 @@ class LimitModel(sql.ModelBase, sql.ModelDictMixin): def service_id(self): if self.registered_limit: return self.registered_limit.service_id - return self._service_id - - @service_id.setter - def service_id(self, value): - self._service_id = value + return None @service_id.expression def service_id(self): - return LimitModel._service_id + return RegisteredLimitModel.service_id @hybrid_property def region_id(self): if self.registered_limit: return self.registered_limit.region_id - return self._region_id - - @region_id.setter - def region_id(self, value): - self._region_id = value + return None @region_id.expression def region_id(self): - return LimitModel._region_id + return RegisteredLimitModel.region_id @hybrid_property def resource_name(self): @@ -119,25 +106,16 @@ class LimitModel(sql.ModelBase, sql.ModelDictMixin): return self.registered_limit.resource_name return self._resource_name - @resource_name.setter - def resource_name(self, value): - self._resource_name = value - @resource_name.expression def resource_name(self): - return LimitModel._resource_name - - @classmethod - def from_dict(cls, limit): - obj = super(LimitModel, cls).from_dict(limit) - with sql.session_for_read() as session: - query = session.query(RegisteredLimitModel).filter_by( - id=obj.registered_limit_id) - obj.registered_limit = query.first() - return obj + return RegisteredLimitModel.resource_name def to_dict(self): ref = super(LimitModel, self).to_dict() + if self.registered_limit: + ref['service_id'] = self.registered_limit.service_id + ref['region_id'] = self.registered_limit.region_id + ref['resource_name'] = self.registered_limit.resource_name ref.pop('internal_id') ref.pop('registered_limit_id') return ref @@ -151,16 +129,18 @@ class UnifiedLimit(base.UnifiedLimitDriverBase): # current reference between registered limit and limit. i.e. We should # ensure that there is no duplicate entry. hints = driver_hints.Hints() - hints.add_filter('service_id', unified_limit['service_id']) - hints.add_filter('resource_name', unified_limit['resource_name']) - hints.add_filter('region_id', unified_limit.get('region_id')) if is_registered_limit: + hints.add_filter('service_id', unified_limit['service_id']) + hints.add_filter('resource_name', unified_limit['resource_name']) + hints.add_filter('region_id', unified_limit.get('region_id')) with sql.session_for_read() as session: query = session.query(RegisteredLimitModel) unified_limits = sql.filter_limit_query(RegisteredLimitModel, query, hints).all() else: + hints.add_filter('registered_limit_id', + unified_limit['registered_limit_id']) is_project_limit = (True if unified_limit.get('project_id') else False) if is_project_limit: @@ -169,27 +149,8 @@ class UnifiedLimit(base.UnifiedLimitDriverBase): hints.add_filter('domain_id', unified_limit['domain_id']) with sql.session_for_read() as session: query = session.query(LimitModel) - old_unified_limits = sql.filter_limit_query(LimitModel, - query, - hints).all() - query = session.query( - LimitModel).outerjoin(RegisteredLimitModel) - query = query.filter( - RegisteredLimitModel.service_id == - unified_limit['service_id'], - RegisteredLimitModel.region_id == - unified_limit.get('region_id'), - RegisteredLimitModel.resource_name == - unified_limit['resource_name']) - if is_project_limit: - query = query.filter( - LimitModel.project_id == unified_limit['project_id']) - else: - query = query.filter( - LimitModel.domain_id == unified_limit['domain_id']) - new_unified_limits = query.all() - - unified_limits = old_unified_limits + new_unified_limits + unified_limits = sql.filter_limit_query(LimitModel, query, + hints).all() if unified_limits: msg = _('Duplicate entry') @@ -296,12 +257,16 @@ class UnifiedLimit(base.UnifiedLimitDriverBase): with sql.session_for_write() as session: new_limits = [] for limit in limits: - self._check_unified_limit_unique(limit, - is_registered_limit=False) target = self._check_and_fill_registered_limit_id(limit) + self._check_unified_limit_unique(target, + is_registered_limit=False) ref = LimitModel.from_dict(target) session.add(ref) - new_limits.append(ref.to_dict()) + new_limit = ref.to_dict() + new_limit['service_id'] = limit['service_id'] + new_limit['region_id'] = limit.get('region_id') + new_limit['resource_name'] = limit['resource_name'] + new_limits.append(new_limit) return new_limits except db_exception.DBReferenceError: raise exception.NoLimitReference() @@ -310,43 +275,18 @@ class UnifiedLimit(base.UnifiedLimitDriverBase): def update_limit(self, limit_id, limit): with sql.session_for_write() as session: ref = self._get_limit(session, limit_id) - old_dict = ref.to_dict() - old_dict.update(limit) - new_limit = LimitModel.from_dict(old_dict) - ref.resource_limit = new_limit.resource_limit - ref.description = new_limit.description + if limit.get('resource_limit'): + ref.resource_limit = limit['resource_limit'] + if limit.get('description'): + ref.description = limit['description'] return ref.to_dict() @driver_hints.truncated def list_limits(self, hints): - hint_copy = copy.deepcopy(hints) - new_format_data = [] with sql.session_for_read() as session: - query = session.query(LimitModel) - limits = sql.filter_limit_query(LimitModel, - query, - hints) - old_format_data = [s.to_dict() for s in limits] - project_filter = hint_copy.get_exact_filter_by_name('project_id') - domain_filter = hint_copy.get_exact_filter_by_name('domain_id') - if hint_copy.filters and (not (project_filter or domain_filter) - or len(hint_copy.filters) > 1): - # If the hints contain "service_id", "region_id" or - # "resource_name", we should combine the registered_limit table - # first to fetch these information. - query_new = session.query( - LimitModel).outerjoin(RegisteredLimitModel) - limits = sql.filter_limit_query(RegisteredLimitModel, - query_new, - hint_copy) - if project_filter: - limits = limits.filter( - LimitModel.project_id == project_filter['value']) - elif domain_filter: - limits = limits.filter( - LimitModel.domain_id == domain_filter['value']) - new_format_data = [s.to_dict() for s in limits] - return old_format_data + new_format_data + query = session.query(LimitModel).outerjoin(RegisteredLimitModel) + limits = sql.filter_limit_query(LimitModel, query, hints) + return [limit.to_dict() for limit in limits] def _get_limit(self, session, limit_id): query = session.query(LimitModel).filter_by(id=limit_id) diff --git a/keystone/tests/unit/test_sql_upgrade.py b/keystone/tests/unit/test_sql_upgrade.py index fe1f5b230..cfa65cf06 100644 --- a/keystone/tests/unit/test_sql_upgrade.py +++ b/keystone/tests/unit/test_sql_upgrade.py @@ -3366,6 +3366,27 @@ class FullMigration(SqlMigrateBase, unit.TestCase): self.assertEqual(trust['redelegation_count'], upgraded_trust.redelegation_count) + def test_migration_063_drop_limit_columns(self): + self.expand(62) + self.migrate(62) + self.contract(62) + + limit_table = 'limit' + self.assertTableColumns( + limit_table, + ['id', 'project_id', 'service_id', 'region_id', 'resource_name', + 'resource_limit', 'description', 'internal_id', + 'registered_limit_id', 'domain_id']) + + self.expand(63) + self.migrate(63) + self.contract(63) + + self.assertTableColumns( + limit_table, + ['id', 'project_id', 'resource_limit', 'description', + 'internal_id', 'registered_limit_id', 'domain_id']) + class MySQLOpportunisticFullMigration(FullMigration): FIXTURE = db_fixtures.MySQLOpportunisticFixture |