README

A WordPress transients interface with support for soft-expiration (use old content until new content is available), background updating of the transients (without having to wait for a cron job), and a chainable syntax that allows for one liners.

License

TLC Transients is licensed under the GPL, version 2.0 or any later version. See LICENSE.

Examples

In this simple example, we're defining a feed-fetching callback, and then using tlc_transient with a chain to point to that callback and use it, all in one line. Note that since we haven't used background_only(), the initial load of this will cause the page to pause.

<?php // Define your callback (other examples use this) function my_callback() { return wp_remote_retrieve_body( wp_remote_get( 'http://example.com/feed.xml', array( 'timeout' => 30 ) ) ); } // Grab that feed echo tlc_transient( 'example-feed' ) ->updates_with( 'my_callback' ) ->expires_in( 300 ) ->get(); ?>

This time, we'll set background_only() in the chain. This means that if there has been a hard cache flush, or this is the first-ever request, it will return false. So your code will have to be written to gracefully degrade if the feed isn't yet available. This, of course, triggers a background update. And once it is available, it will start returning the content.

<?php echo tlc_transient( 'example-feed' ) ->updates_with( 'my_callback' ) ->expires_in( 300 ) ->background_only() ->get(); ?>

We don't have to chain, of course.

<?php $t = tlc_transient( 'example-feed' ); if ( true ) { $t->updates_with( 'my_callback' ); } else { $t->updates_with( 'some_other_callback' ); } $t->expires_in( 300 ); echo $t->get(); ?>

We can even pass parameters to our callback.

<?php // Define your callback function my_callback_with_param( $param ) { return str_replace( 'foo', $param, wp_remote_retrieve_body( wp_remote_get( 'http://example.com/feed.xml', array( 'timeout' => 30 ) ) ), ); } // Grab that feed echo tlc_transient( 'example-feed' ) ->updates_with( 'my_callback_with_param', array( 'bar' ) ) ->expires_in( 300 ) ->background_only() ->get(); ?>

Notes

Context

It should be noted that when a callback runs asynchronously, you are not in control of context. The context that existed when you registered the callback has no bearing on the context when the callback is actually run. So if there is anything that you're assuming in your callback function (whether a certain user being current, a certain post having been queried, etc), you must rewrite your calllback function so that these assumptions are not made, and instead pass in this context in the form of parameters, which your callback then uses to recreate your desired context.