SOLR Search for TYPO3

What Is SOLR?

Apache Solr is a standalone, open-source search engine software from the Apache Software Foundation. Via a matching TYPO3 extension, it can be integrated as a search engine for your own website, going far beyond plain full-text search: autocomplete, spell correction, support for synonyms and word stems, result highlighting, faceted search, and custom weighting of individual areas (e.g. giving current news more weight) are all part of its feature set.

To use SOLR in TYPO3, you need:

• your own Solr instance reachable over the internet
• the TYPO3 extension solr
• for searching within files (PDF, etc.), additionally the extensions tika and solr_fal

SOLR and AI

Both the TYPO3 and Solr communities are currently actively developing ways to add AI capabilities to classic search – for example via the "Learning To Rank" (LTR) module, available from version 13.1.0 of the "Apache Solr for TYPO3 – Enterprise Search" extension. Going forward, search is evolving from plain keyword matching toward a smarter, more conversational search experience. More information on Solr is available at lucene.apache.org/solr.

By Comparison: TYPO3's Standard Search

For smaller websites, TYPO3 already ships with indexed_search, a built-in, easy-to-set-up search function. The first time a page is called up, its content is automatically indexed in the database – regardless of whether it's a regular page, an extension detail view like News, or a custom output.

For larger websites, however, indexed_search quickly hits its limits: the index tables can grow very large under heavy use of news or event plugins, noticeably slowing the system down. It also lacks important features like faceted search, typo tolerance, stop words, or result weighting. Since it's a plain full-text search, a query like "TYPO3 agency" only matches the exact word sequence – phrasings like "agency for TYPO3" wouldn't be found.

SOLR and Privacy

With the TYPO3 extension "solr", traffic doesn't flow directly from the visitor's browser to the Solr server, but always via your own web server: TYPO3 receives the search request, communicates with the Solr server in the background, and returns the processed response to the browser. TYPO3 thus acts as an intermediary.

This has several advantages: direct access to the Solr server stays hidden, reducing the attack surface. The web server can monitor and, if needed, filter all requests. And search queries or results can still be adjusted by TYPO3 before being delivered.

Making Files Searchable Too

With the additional extensions tika and solr_fal, PDF and other files on the server can also be indexed and made searchable – this also runs via a regular cron job. For files to appear meaningfully in search results, their metadata (title, description) should be properly maintained. solr_fal also brings additional features, such as displaying "most popular searches" and "recently searched", search statistics in the backend, and the ability to editorially maintain synonyms.