[DDC-668] add upsert support Created: 04/Jul/10  Updated: 20/Dec/11

Status: Open
Project: Doctrine 2 - ORM
Component/s: ORM
Affects Version/s: None
Fix Version/s: 2.x
Security Level: All

Type: New Feature Priority: Major
Reporter: Lukas Kahwe Assignee: Benjamin Eberlei
Resolution: Unresolved Votes: 4
Labels: None


 Description   

Didnt find anything in the docs on this. Is D2 capable of doing an UPSERT [1] in case I am trying to persist an object that may or may not have been saved previously. Different RDBMS support different syntax for this case. Like MySQL has INSERT .. ON DUPLICATE KEY UPDATE (or even INSERT IGNORE) while the SQL standard defines a MERGE syntax which seems to be gaining support. Of course you can always fallback to a SELECT FOR UPDATE (or if you want to be hacky an INSERT which catches duplicate key violations .. but probably not a good idea since many RDBMS rollback on a failure inside a transaction).

[1] http://en.wikipedia.org/wiki/Upsert

See also http://opensource.atlassian.com/projects/hibernate/browse/HHH-3011 asking for MERGE support

Ideally there would be a way to define on a model or model instance level if merge logic should be applied.



 Comments   
Comment by Robert Burkhead [ 09/Jul/10 ]

Doctrine_Record defines a replace() method.

In the MySQL Doctrine implementation, however, it is not the same as INSERT .. ON DUPLICATE KEY UPDATE. The replace() method implemented in Doctrine_Connection_Mysql uses the REPLACE INTO syntax, which is a DELETE and then INSERT when the key exists. This is fine, except for tables that use auto-increment fields. The delete-then-insert operation yields a new auto-incremented value, whereas INSERT .. ON DUPLICTATE KEY UPDATE would not.

Comment by Lukas Kahwe [ 09/Jul/10 ]

MySQL (and SQLite) REPLACE is a no go. It causes way too much disc I/O and worse yet totally screws up the on disk data structures because of the deleting.

Comment by Benjamin Eberlei [ 31/Jul/11 ]

Scheduled for 2.2

Comment by Benjamin Eberlei [ 31/Jul/11 ]

Evaluating this makes me sad, except MySQL support for this is rather non-existant, and the oracle merge is aiming at batch operations.

Comment by Benjamin Eberlei [ 22/Oct/11 ]

Should this be done with

1. Select first, then insert
2. Catch and evaluate exception then update

I am leaning towards 1.

Comment by Guilherme Blanco [ 20/Dec/11 ]

Updating fix version

Generated at Sun Dec 21 14:37:54 UTC 2014 using JIRA 6.2.3#6260-sha1:63ef1d6dac3f4f4d7db4c1effd405ba38ccdc558.