ArchiveArchive

How To Force Search API To Reindex a Node / an Entity

Oct 27, 2014 · Updated: Nov 04, 2017 · by Tim Kamanin

By default Search API (Drupal 7) reindexes a node when the node gets updated. But what if you want to reindex a node / an entity on demand or via some other hook i.e. outside of update cycle? Turned out it is a quite simple exercise. You just need to execute this function call whenever you want to reindex a node / an entity:

search_api_track_item_change('node', array($nid));

See this snippet at dropbucket: http://dropbucket.org/node/1600 search_api_track_item_change marks the items with the specified IDs as "dirty", i.e., as needing to be reindexed. You need to supply this function with two arguments: entity_type ('node' in our example) and an array of entity_ids you want to be reindexed. Once you've done this, Search API will take care of the rest as if you've just updated your node / entity. Additional tip: In some cases, it 's worth to clear field_cache for an entity before sending it to reindex:

// Clear field cache for the node.
cache_clear_all('field:node:' . $nid, 'cache_field');

// Reindex the node.
search_api_track_item_change('node', array($nid));

This is the case, when you manually save / update entity values via sql queries and then want to reindex the result (for example, radioactivity module doesn't save / update a node, it directly manipulates data is sql tables). That way you'll ensure that search_api reindexes fresh node / entity and not the cached one.

Hey, if you've found this useful, please share the post to help other folks find it:

There's even more:

Subscribe for updates

  • via Twitter: @timonweb
  • old school RSS:
  • or evergreen email ↓ ↓ ↓