Skip to main content
Version: 4.1

Transactions

Moodle allows data manipulation to take place within a database transaction, known as a Delegated transaction. This allows you to perform CRUD1 operations, and roll them back if a failure takes place.

General principles​

  1. These delegated transactions work in a way that, when nested, the outer levels have control over the inner ones.
  2. Code should not rely on a rollback happening. It is only a measure to reduce (not to eliminate) DB2 garbled information
  3. Any code using transactions that result in unfinished, unbalanced, or finished twice transactions will generate a transaction_exception and the DB will perform a rollback
  4. If one transaction (at any level) has been marked for rollback() there will not be any method to change it. Finally Moodle will perform the DB rollback
  5. If one transaction (at any level) has been marked for allow_commit() it will be possible to change that status to rollback() in any outer level
  6. It will be optional to catch exceptions when using transactions, but if they are caught, then it is mandatory to mark the transaction for rollback()
  7. Any explicit rollback() call will pass the exception originating from it, as in rollback($exception), to be re-thrown

The API​

  1. All the handling must go, exclusively, to a moodle_database object, leaving real drivers only implementing (protected) the old begin/commit/rollback_sql() functions

  2. One array of objects of type moodle_transaction will be stored / checked from $DB

  3. $DB will be the responsible to instantiate / accumulate / pair / compare moodle_transactions

  4. Each moodle_transaction will be able to set the global mark for rollback. Commit won't change anything

  5. Inner-most commit/rollback will printout one complete stack of moodle_transactions information if we are under DEBUG_DEVELOPER and the new setting delegatedtransactionsdebug is enabled

  6. Normal usage of the moodle_transaction will be:

    $transaction = $DB->start_delegated_transaction();
    // Perform some $DB stuff
    $transaction->allow_commit();

  7. If, for any reason, the developer needs to catch exceptions when using transactions, it will be mandatory to use it in this way:

    try {
        $transaction = $DB->start_delegated_transaction();
        // Perform some $DB stuff.
        $transaction->allow_commit();
    } catch (Exception $e) {
        // Extra cleanup steps.
        // Re-throw exception after commiting.
        $transaction->rollback($e);
    }

  8. In order to be able to keep some parts of code out from top transactions completely, if we know it can lead to problems, we can use:

    // Check to confirm we aren't using transactions at this point.
    // This will throw an exception if a transaction is found.
    $DB->transactions_forbidden();

The Flow​

The flow of transactions in diagram format

  1. Any default exception handler will:
    1. Catch uncaught transaction_exception exceptions
    2. Properly perform the DB rollback
    3. debug/error/log honouring related settings
    4. inform with as many details as possible (token, place... whatever)
  2. Any "footer" (meaning some place before ending <html> output) will:
    1. Detect "in-transaction" status
    2. Let execution continue, transaction is automatically rolled back in $DB->dispose()
    3. inform with as many details as possible (token, place... whatever)
  3. $DB->dispose() will:
    1. Detect "in-transaction" status
    2. log error (not possible to honour settings!)
    3. Properly perform the full DB rollback

Footnotes​

  1. Create Read Update Delete ↩

  2. The Moodle database ↩