This project is no longer maintained and has been archived.


Transactions


Introduction

A database transaction is a unit of interaction with a database management system or similar system that is treated in a coherent and reliable way independent of other transactions that must be either entirely completed or aborted. Ideally, a database system will guarantee all of the ACID (Atomicity, Consistency, Isolation, and Durability) properties for each transaction.

  • Atomicity refers to the ability of the DBMS to guarantee that either all of the tasks of a transaction are performed or none of them are. The transfer of funds can be completed or it can fail for a multitude of reasons, but atomicity guarantees that one account won't be debited if the other is not credited as well.
  • Consistency refers to the database being in a legal state when the transaction begins and when it ends. This means that a transaction can't break the rules, or //integrity constraints//, of the database. If an integrity constraint states that all accounts must have a positive balance, then any transaction violating this rule will be aborted.
  • Isolation refers to the ability of the application to make operations in a transaction appear isolated from all other operations. This means that no operation outside the transaction can ever see the data in an intermediate state; a bank manager can see the transferred funds on one account or the other, but never on both - even if she ran her query while the transfer was still being processed. More formally, isolation means the transaction history (or schedule) is serializable. For performance reasons, this ability is the most often relaxed constraint. See the isolation article for more details.
  • Durability refers to the guarantee that once the user has been notified of success, the transaction will persist, and not be undone. This means it will survive system failure, and that the database system has checked the integrity constraints and won't need to abort the transaction. Typically, all transactions are written into a log that can be played back to recreate the system to its state right before the failure. A transaction can only be deemed committed after it is safely in the log.

In Doctrine all operations are wrapped in transactions by default. There are some things that should be noticed about how Doctrine works internally:

  • Doctrine uses application level transaction nesting.
  • Doctrine always executes INSERT / UPDATE / DELETE queries at the end of transaction (when the outermost commit is called). The operations are performed in the following order: all inserts, all updates and last all deletes. Doctrine knows how to optimize the deletes so that delete operations of the same component are gathered in one query.

First we need to begin a new transation:

$ conn->beginTransaction();

Next perform some operations which result in queries being executed:

$ user       = new User();
$ user->name = 'New user';
$ user->save();

$ user       = Doctrine_Core::getTable('User')->find(5);
$ user->name = 'Modified user';
$ user->save();

Now we can commit all the queries by using the commit() method:

$ conn->commit();

Nesting

You can easily nest transactions with the Doctrine DBAL. Check the code below for a simple example demonstrating nested transactions.

First lets create a standard PHP function named saveUserAndGroup():

function saveUserAndGroup(Doctrin_Connection $conn, User $user, Group $group)
{
    $conn->beginTransaction();

    $user->save();

    $group->save();

    $conn->commit();
}

Now we make use of the function inside another transaction:

try {
    $conn->beginTransaction();

    saveUserAndGroup($conn,$user,$group);
    saveUserAndGroup($conn,$user2,$group2);
    saveUserAndGroup($conn,$user3,$group3);

    $conn->commit();

} catch(Doctrine_Exception $e) {
    $conn->rollback();
}

Notice how the three calls to saveUserAndGroup() are wrapped in a transaction, and each call to the function starts its own nested transaction.


Savepoints

Doctrine supports transaction savepoints. This means you can set named transactions and have them nested.

The Doctrine_Transaction::beginTransaction($savepoint) sets a named transaction savepoint with a name of $savepoint. If the current transaction has a savepoint with the same name, the old savepoint is deleted and a new one is set.

try {
    $conn->beginTransaction();
    // do some operations here

    // creates a new savepoint called mysavepoint
    $conn->beginTransaction('mysavepoint');
    try {
        // do some operations here

        $conn->commit('mysavepoint');
    } catch(Exception $e) {
        $conn->rollback('mysavepoint');
    }
    $conn->commit();
} catch(Exception $e) {
    $conn->rollback();
}

The Doctrine_Transaction::rollback($savepoint) rolls back a transaction to the named savepoint. Modifications that the current transaction made to rows after the savepoint was set are undone in the rollback.

Mysql, for example, does not release the row locks that were stored in memory after the savepoint.

Savepoints that were set at a later time than the named savepoint are deleted.

The Doctrine_Transaction::commit($savepoint) removes the named savepoint from the set of savepoints of the current transaction.

All savepoints of the current transaction are deleted if you execute a commit or if a rollback is being called without savepoint name parameter.

try {
    $conn->beginTransaction();
    // do some operations here

    // creates a new savepoint called mysavepoint
    $conn->beginTransaction('mysavepoint');

    // do some operations here

    $conn->commit();   // deletes all savepoints
} catch(Exception $e) {
    $conn->rollback(); // deletes all savepoints
}

Isolation Levels

A transaction isolation level sets the default transactional behavior. As the name 'isolation level' suggests, the setting determines how isolated each transation is, or what kind of locks are associated with queries inside a transaction. The four available levels are (in ascending order of strictness):

READ UNCOMMITTED
Barely transactional, this setting allows for so-called 'dirty reads', where queries inside one transaction are affected by uncommitted changes in another transaction.
READ COMMITTED
Committed updates are visible within another transaction. This means identical queries within a transaction can return differing results. This is the default in some DBMS's.
REPEATABLE READ
Within a transaction, all reads are consistent. This is the default of Mysql INNODB engine.
SERIALIZABLE
Updates are not permitted in other transactions if a transaction has run an ordinary SELECT query.

To get the transaction module use the following code:

$ tx = $conn->transaction;

Set the isolation level to READ COMMITTED:

$ tx->setIsolation('READ COMMITTED');

Set the isolation level to SERIALIZABLE:

$ tx->setIsolation('SERIALIZABLE');

Some drivers (like Mysql) support the fetching of current transaction isolation level. It can be done as follows:

$ level = $tx->getIsolation();

Conclusion

Transactions are a great feature for ensuring the quality and consistency of your database. Now that you know about transactions we are ready to move on and learn about the events sub-framework.

The events sub-framework is a great feature that allows you to hook in to core methods of Doctrine and alter the operations of internal functionality without modifying one line of core code.