Indexer

Purpose

Indexer allows you to index to a content object data coming from ancestors / descendants of the current content object when published. Building a consistent, rich search index is the best way to provide relevant search results.Common exemples could include:

• Indexing a product's category name, when the category is the product's parent
• Centralizing search data to the object that matters in your hierarchy: let's say you have Speach objects, organized under Speakers. You may want to index some Speaker information in the Speach object so that you can look for Speaches by searching for author name, place...
• When you have a multipage article system, implemented using a Multi-page article parent object containing several article objects: you want the search engine to return the Multipage articles, but you want to be able to search for data stored in the articles themselves.

This extension will provide all of these for you. Just add a new "Parent indexer" attribute, and configure the classes / attributes the datatype should search for when the object is published. These attributes data will be added to the search index for the published object. A workflow will automatically take care of reindexing the requires objects upon publishing of parents / children (if you update an article, the multipage parent object should be reindexed).

Project status

The extension is far from being complete. At the moment, only parent indexing has been designed the way we want. The workflow have not been to the new structure yet, but this should be added shortly.

In the end, the extension should only provide one "indexer" datatype, suitable for both ancestors and descendants indexing.

Credits

This project is based on Jan Kudlika's article / extensions: http://ez.no/developer/articles/helper_child_nodes_user_friendly_search_results_and_preventing_full_node_views. The extension tries to provide a better integration, with a visual GUI at class datatype level that let you configure what attributes should be indexed.