I need to implement the following classification in SQL:
- Multiple tests in the group Or Multiple
- Testing can occur in many groups
- (Children) Groups may be in multiple (parent) groups
.. Which means 2x: N relationships. The following table groups can store trees:
TEST_TABLE
test_id (pk), test_nameGROUP_TABLE < Group_id (pk), group_name
TEST_IN_GROUP_TABLE
test_id (fk), group_id (fk)GROUP_RELATIONS_TABLE
Parent_group_id (fk), child_group_id (fk)but:
- How to list all the tests in the group? How modify tables me impossibly difficult to find, because the group can be tested or sub-group with other tests
- or choose it? I can make an additional child_count column in GROUP_TABLE, it will be filled automatically by the trigger but with this it is also hard for me.
If you used the general table expression first group, then using that A group of groups, creating group_ids to be included on the test table through TEST_IN_GROUP_TABLE for all tests
;. When as select group_id from (GROUP_TABLE with CTE where group_id = 1 union select all GROUP_RELATIONS_TABLE inner child_group_id include CTE tg.group_id = cte.group_id inner TEST_TABLE not to join the inner TEST_IN_GROUP_TABLE TG t.test_id = Tg.test_id
- or choose it? I can make an additional child_count column in GROUP_TABLE, it will be filled automatically by the trigger but with this it is also hard for me.
- How to list all the tests in the group? How modify tables me impossibly difficult to find, because the group can be tested or sub-group with other tests
Comments
Post a Comment