3.0.1 - Retry transaction when commit fails with a RetryableException

What's Changed

• fix: Retry transaction when commit fails with a RetryableException by @paxuclus in #21

Full Changelog: 3.0.0...3.0.1

3.0.0 - Add option to skip execution of scheduled jobs

What's Changed

• feat: Add option to skip execution of scheduled jobs by @paxuclus in #20

Full Changelog: 2.2.1...3.0.0

2.2.1 - Add indixes for claiming lots of jobs

Adding new claim values to a huge list of jobs ready to be worked on
will always require ordering the jobs table by date before updating
the very first record.

Retrieveing the one previously claimed record should start with the
claim row to segment all nun-claimed records away to allow the search
mechanism to ignore all those values.

2.2.0 - Require PHP 8.2 and Flow 8.3

Merge pull request #18 from netlogix/bugfix/reconnect-and-retry-when-…

2.1.1 - fix: DBAL connection retries when connection was lost

Try and retry an SQL query in case of connection timeouts. This avoids
multiple PING requests in rapid succession. Since all query performed
are meant to be atomic anyway, there should be no lost data and no data
duplication.

2.1.0 - Use Custom Database Connection

The scheduler used to rely on the applications database connection.

When app processes start transaction that take a couple of seconds,
every scheduling call done within this app process is captured by
this transaction, which potentially locks the scheduler table for
worker processes.

This change now uses the same database credentials as the entity manager
but creates a new connection object. All SQL queries issued by the
scheduler are meant to be atomic. Having the buried within application
transactions hinders the synchronization of multiple parallel scheduler
instances.

2.0.0 - Scheduled Groups and Auto Scheduling

• Scheduled Jobs must be assigned to Groups.
• Groups must have individual workers.
• Even though parallel workers are available for a long time, multiple workers only increase the number of jobs per second. Multiple groups, on the other hand, allow for "job type A must not delay job type B" settings.
• There is a default group that is named "default-group".
• Introduce "copy to scheduler" interface to allow every regular jobqueue job to be self-scheduling
• Increase session wait timeout to one hour when working on scheduled jobs to allow for jobs that work for a long time.

Breaking: Several methods require the "groupName" argument.

1.3.0 - Add option to keep failed jobs after reaching their numberOfRetries

What's Changed

• Build(deps): Bump actions/checkout from 3 to 4 by @dependabot in #12
• feat: Add option to keep failed jobs after reaching their numberOfRetries by @paxuclus in #15

Full Changelog: 1.2.4...1.3.0

1.2.4 - Make phpunit dataProviders static

What's Changed

• BUGFIX: Make phpunit dataProviders static by @paxuclus in #11

Full Changelog: 1.2.3...1.2.4

1.2.3 - Add missing return type hints to doctrine migrations

What's Changed

• BUGFIX: Add return type hint to migration methods by @nlx-lars in #10

Full Changelog: 1.2.2...1.2.3