frootbox / search

v0.1.0 2026-11-04 11:59 UTC

Requires

• php: >=8.5
• frootbox/core: ^0.1
• frootbox/database: ^0.1

MIT

• search
• cms
• site-search
• frootbox

Published on Packagist.org on 2026-11-04 16:36 UTC

README

Search for the Frootbox CMS: visitors find published pages, news articles, events and the contents of other modules on the website, in their language, best hits first.

• Site search (frootbox/search/Search) is a block with a search field and, after a search, the hits page by page with the words marked. Pages with results never go into the page cache. Place it on a page of its own; the head of the base theme links to the first page with this block (block_page('frootbox/search/Search')).

The search uses the index of the core (search_index), the same as the global search of the backend, with the published state of every search source that describes it (Frootbox\Core\Search\PublishedSource). The index follows every action right away; it fills itself on the first search and bin/frootbox search:rebuild --website builds it anew.

Left out are drafts, locked pages, pages left out of the search on the website (with their branch if so set) and addresses that only show the content of another language.

For extensions

A search source of an extension takes part as soon as it implements PublishedSource::published() and returns a PublishedEntry with the target of each item, e.g. acme/shop/product:01J…. The target needs an address, usually from $registry->plugin().

Query

frootbox/search/website.find with q, language, limit and offset, open to anyone, also over the REST API: GET /api/v1/frootbox/search/website.find?q=coffee.

Source and issues

This package is developed in the Frootbox monorepo and mirrored here read-only. Questions, bugs and ideas go there, see the documentation.