100

I don't design schemas everyday, but when I do, I try to setup cascade updates/deletes correctly to make administration easier. I understand how cascades work, but I can never remember which table is which.

For example, if I have two tables - Parent and Child - with a foreign key on Child that references Parent and has ON DELETE CASCADE, which records trigger a cascade and which records get deleted by the cascade? My first guess would be the Child records get deleted when Parent records are deleted, since Child records depend on Parent records, but the ON DELETE is ambiguous; it could mean delete the Parent record when the Child record is deleted, or it could mean delete the Child record when the Parent is deleted. So which is it?

I wish the syntax was ON PARENT DELETE, CASCADE, ON FOREIGN DELETE, CASCADE or something similar to remove the ambiguity. Does anyone have any mnemonics for remembering this?

141

If you like the Parent and Child terms and you feel they are easy to be remembered, you may like the translation of ON DELETE CASCADE to Leave No Orphans!

Which means that when a Parent row is deleted (killed), no orphan row should stay alive in the Child table. All children of the parent row are killed (deleted), too. If any of these children has grandchildren (in another table through another foreign key) and there is ON DELETE CASCADE defined, these should be killed, too (and all descendants, as long as there is a cascade effect defined.)

The FOREIGN KEY constraint itself could also be described as Allow No Orphans! (in the first place). No Child should ever be allowed (written) in the child table if it hasn't a Parent (a row in the parent table).

For consistency, the ON DELETE RESTRICT can be translated to the (less aggresive) You Can't Kill Parents! Only childless rows can be killed (deleted.)

  • 3
    I feel something is still missing in the analogy. Can't a child have more than one parent? In this case will killing one parent make the child an orphan? – Jus12 Nov 6 '14 at 13:31
  • 7
    @Jus12 No, foreign key constraints work with 1 parent only. It's not a good analogy regarding this aspect. – ypercubeᵀᴹ Nov 6 '14 at 13:33
  • 1
    @ypercube: Is this not allowed? Order(custID, itemID, orderID) where custID refers to a primary key in Customers table and itemID refers to a primary key in Items table. Won't Order have two parents? – Jus12 Nov 6 '14 at 15:32
  • 4
    @Jus12 That is allowed of course but it would be 2 foreign key constraints. Then every child (order) would have a parent (customer) and a parent (item). The behaviours of the 2 FKs might differ though. (So, for example, it could be that killing customers would kill all their (orders) children but killing items would not kill their orders.) – ypercubeᵀᴹ Nov 6 '14 at 15:45
  • 1
    The parent analogy can still work if we don't say "orphan". If there are two references to two separate parents on a child entry, this can still be seen as a child of a divorced couple. Restrict: "I won't let you kill my mom" Cascade: "If you kill my dad, I will also die" – Christopher McGowan Feb 28 '15 at 18:15
32

For example, if I have two tables - Parent and Child - where Child records are owned by Parent records, which table needs the ON DELETE CASCADE?

ON DELETE CASCADE is an optional clause in a foreign key declaration. So it goes with the foreign key declaration. (Meaning, in the "child" table.)

...it could mean delete the Parent record when the Child record is deleted, or it could mean delete the Child record when the Parent is deleted. So which is it?

One way to interpret a foreign key declaration is, "All valid values for this column come from 'that_column' in 'that_table'." When you delete a row in the "child" table, nobody cares. It doesn't affect data integrity.

When you delete a row from the "parent" table--from "that_table"--you remove a valid value from the possible values for the "child" table. To maintain data integrity, you have to do something to the "child" table. Cascading deletes is one thing you can do.

2

SQL:2011 Spec

There are five options for ON DELETE, and ON UPDATE that can apply to the FOREIGN KEY. These are called <referential actions>, directly from the SQL:2011 spec

  • ON DELETE CASCADE: if a row of the referenced table is deleted, then all matching rows in the referencing table are deleted.
  • ON DELETE SET NULL: if a row of the referenced table is deleted, then all referencing columns in all matching rows of the referencing table to be set to null.
  • ON DELETE SET DEFAULT: if a row of the referenced table is deleted, then all referencing columns in all matching rows of the referencing table to be set to the column’s default value.
  • ON DELETE RESTRICT: it is prohibited to delete a row of the referenced table if that row has any matching rows in the referencing table.
  • ON DELETE NO ACTION (the default): there is no referential delete action; the referential constraint only specifies a constraint check.

The foreign key establishes the dependent relationship. The <referential action> determines what happens when the relationship is dissolved.

Example / Metaphor / Explanation

For this example, we will accept the common model of society and economy: where every business is a company that maintains a relationship to the bourgeoisie through a fatcat_owner.

CREATE TABLE bourgeoisie(
  fatcat_owner varchar(100) PRIMARY KEY
);
INSERT INTO bourgeoisie(fatcat_owner) VALUES
  ( 'Koch Brothers' );

CREATE TABLE business (
  name         varchar(100),
  fatcat_owner varchar(100) REFERENCES bourgeoisie
);
INSERT INTO business(name, fatcat_owner)
  VALUES ('Georgia-Pacific', 'Koch Brothers');

If all businesses are directly affected by bourgeoisie by way of their fatcat_owner then what do you do after the workers' revolution when that you purge the fatcat_owners and have a classless society?

-- Viva la revolución 
BEGIN;
  DELETE FROM bourgeoisie;
END;

You have a few options here,

  • Stop the revolution. In SQL parlance, RESTRICT. Some people believe this is the lesser evil, but they're usually wrong.
  • Allow it to go on. If so when the revolution happens SQL gives you four options,

    • SET NULL -- leave it blank. Who knows, maybe capitalism is restored the bourgeoisie comes up and oligarchs fill the roll of the fatcat_owners. Important note, the column must be NULLABLE (not NOT NULL) or this can never happen.
    • SET DEFAULT -- perhaps you had a DEFAULT that handled this? A DEFAULT can call a function. Maybe your schema is revolution-ready, already.
    • CASCADE -- there is no damage control. If the bourgeoisie goes, so does the business. If a business must have a fatcat_pig, then sometimes it makes more sense to lose the data rather than have a non-business in a business table.
    • NO ACTION -- this is essentially a method of delaying the check, in MySQL it's no different than RESTRICT, but in PostgreSQL, you'd be able to do

      -- Not a real revolution.
      -- requires constraint be DEFERRABLE INITIALLY DEFERRED
      BEGIN;
        SET CONSTRAINTS ALL DEFERRED;
        DELETE FROM bourgeoisie;
        INSERT INTO bourgeoisie VALUES ( 'Putin' );
        UPDATE business SET fatcat_pig = 'Putin';
      END;
      

      In such a system, the constraint is validated only before the transaction commits. This may result in stopping the revolution, but you can recover in the transaction -- for some degree of "recover."

  • Does referenced table means parent table and referencing table means child table? – sg552 Nov 11 '18 at 17:19
  • @sg552 Yes, you understood it correctly. – informatik01 May 16 at 21:58
1

A simple mnemonic would be

ON DELETE of parent CASCADE [by deleting] here

That tells you which deletes (deletes of the parent) get cascaded, where the ON DELETE CASCADE statement goes (on the child), and what gets deleted (the child).

-3

well, perhaps we can rationalize the syntax. Let's take a Python example:

class Parent(self):
    # define parent's fields

class Child(self):    
    # define child's fields
    parent_pk_is_childs_foreign_key = models.ForeignKey(Parent, on_delete=models.CASCADE)

what this line says is on_delete of the Parent (which is accidentally mentioned in the statement), please cascade the deletion onto the child. That is why the CASCADE statement is defined at the child level, it marks those children that need to be deleted

For example if you had another class

class GrownUpChild(self):    
        # define grown up child's fields
        parent_pk_is_childs_foreign_key = models.ForeignKey(Parent, on_delete=models.DO_NOTHING)

this structure would clearly show which of the children need to be removed (Child) and which are to stay (GrownUpChild) albeit orphaned

[Edit: Given the context of the discussion, specifically in cases of on_delete=models.CASCADE etc, ] in fact it is often a desired behavior to leave children of a deleted parent, due to auditing and reporting reasons, as well as recovering accidental deletions. [ of course enterprise level software will be built around such behavior and will flag deleted records as deleted=1 instead of actually deleting them and also will not include them in any queries for the front end, minus some specially designed reports. In addition it will have a function of purging the deleted==1 records from the database, which will usually be executed by the UI administrator, often avoiding any involvement from the database administrator's side.]

  • 1
    'In fact it is often a desired behavior to leave children of a deleted parent, due to auditing and reporting reasons, as well as recovering accidental deletions' - that would be disastrous in a (sane) database. – dezso Jul 20 '18 at 10:22
  • @dezso thanks for your input. However, multiple enterprise level CRM systems do exactly that. – George Mogilevsky Jul 23 '18 at 2:22
  • TBH that doesn't make it any more sensible. I once got an assignment to fix shit resulting from such an approach - no joy, except the paycheck. – dezso Jul 24 '18 at 8:22
  • you sound like a savvy database admin :) I can totally hear your point. The software I mentioned above that does it, actually has a function to remove those deleted = 1 manually so it is up to the admin of the app to make this call. Normally database administrator is not even involved in maintaining this aspect. And besides, the entire software's database class is built around the concept, so it always checks for deleted flag in crud operations – George Mogilevsky Jul 24 '18 at 14:38
  • Yes, that is a known and sane pattern - but then you should possibly change the wording above to reflect this. – dezso Jul 25 '18 at 9:09

protected by Community Oct 28 at 13:52

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.