---
title: "Configuration"
version: "1.6"
source: "Configuration.md"
rendered: "2026-09-22T06:15:12+00:00"
---

# Configuration {#configuration}

## Update database structure {#update-database-structure}

Use the "Analyze Database Structure" in the "Maintenance" Admin Tools section to update the database structure.

## Web service {#web-service}

### API endpoint {#api-endpoint}

To access the Algolia API, store the corresponding configuration in the file `additional.php` within the global structure "TYPO3_CONF_VARS" under "EXTENSIONS" and "typo3_search_algolia" (note the spelling) of your TYPO3 installation.

```php
// The universal messenger API endpoint
$GLOBALS['TYPO3_CONF_VARS']['EXTENSIONS']['typo3_search_algolia'] = array_merge(
    $GLOBALS['TYPO3_CONF_VARS']['EXTENSIONS']['typo3_search_algolia'] ?? [],
    [
        'appId' => 'YOUR-APP-ID',
        'apiKey' => 'YOUR-API-KEY',
    ]
);

```

| Field | Description |
| --- | --- |
| appId | Your Algolia application ID. |
| apiKey | An API key with the necessary permissions to make the request. The required access control list (ACL) to make a request is listed in each endpoint's reference. |

You can find your application ID and API key in the Algolia dashboard.

## Basics {#basics}

Next, create a data directory to store the configured index services, search engines, and, if applicable, file collections. To do this, switch to the list view in the TYPO3 backend and add a new "Folder" element to the page tree, then select the created data folder.

## Setting up index services and search engine {#setting-up-index-services-and-search-engine}

### Search engine {#search-engine}

Create a new record of type "Search Engine".

![configuration-search-engine-1](Images/Configuration-001.png)

*Fig. 1: Create a new record of type "Search Engine"*

Add a title, an optional description to the new record, and select the search engine service to use (the default available one is "Algolia Search Service"). To create a custom search engine service, refer to: [Custom Search Engine](CustomSearchEngine.html)

Add a search engine dependent configuration into the "index_name" field. In the case of Algolia searches, this is the name of an index within Algolia in which the indexed data will ultimately be stored.

![configuration-search-engine-2](Images/Configuration-002.png)

*Fig. 2: Configuration TCA of type "Search Engine"*

### Indexing Service {#indexing-service}

In the next step, the respective indexing services are created. These define which data will be indexed.

To do this, create a new record of type "Indexing Service".

![configuration-indexing-service-1](Images/Configuration-003.png)

*Fig. 3: Create a new record of type "Indexing Service"*

Give the indexing service a meaningful title, an optional description, and select the type of indexer to be used. By default, the following are available:

-   Content elements
-   Pages
-   Messages
-   Files

The display updates depending on your selection and offers additional configuration options.

Then select one of the previously created search engine configurations.

![configuration-indexing-service-1](Images/Configuration-004.png)

*Fig. 4: Configuration TCA of type "Indexing Service"*

#### Page Indexer Configuration {#page-indexer-configuration}

The following additional options are available for the page indexer:

-   Include content elements \[include_content_elements\]

    -   Enable this option to index the page's content elements in addition to the page properties. Alternatively, you can use the content element indexer, especially if the search engine has a record size limit.
    -   With this option, the content element fields stored in the TypoScript configuration  `module.tx_typo3searchalgolia.indexer.tt_content.fields` are read and stored directly in the page's index entry in the `content` attribute. A separate indexing service for the content elements is then no longer necessary. However, a data set for a page can become very large, so it's important to check whether the search engine supports this.
-   Page type \[pages_doktype\]

    -   Select the page types by which the pages selected below for indexing should be filtered. If no selection is made here, all page types will be included.
-   Single pages \[pages_single\]

    -   Select individual pages to be indexed here. Indexing will only occur for pages whose "Include in Search" option is enabled in the page properties.
-   Pages (recursively) \[pages_recursive\]

    -   Select the pages you want to recursively index, including all their subpages. Indexing will only occur for pages whose "Include in Search" option is enabled in the page properties.

##### Excluding content element columns (colPos) {#excluding-content-element-columns-colpos}

With "Include content elements" enabled, the page indexer adds the content of every visible content element found on a page to the page's `content` attribute, including content stored in columns that the site's page templates never render (for example a project-specific "unused elements" storage column). To leave such columns out, list their `colPos` values in the TypoScript option `excludeColPos`:

```typoscript
module.tx_typo3searchalgolia.indexer.pages.excludeColPos = 9999

```

The example value is site-specific. Only list columns that your page templates do not render. Column `0` is TYPO3's default main content column, so listing it on a site that renders it removes the main content from the index.

-   The value is a comma-separated list of whole numbers in plain integer notation. Whitespace around the entries is ignored. Anything else (letters, decimals, leading zeros, a plus sign) is ignored as well, so a typo can never be read as `colPos` 0\. Because such an entry is dropped silently, check the result after changing the option.
-   Any `colPos` can be listed, including columns that are not part of a backend layout, such as the child columns of container elements.
-   Only the listed columns are excluded. A column that is not listed, for example a storage column added later or the child columns of a disabled container element, stays indexed.
-   If the option is not set (the default), nothing is excluded and every `colPos` is indexed as before. Do not assign an empty value in your own TypoScript template, simply leave the option out.
-   The option applies to all indexing services of type "Pages" that have "Include content elements" enabled. The content element indexer is not affected by it and keeps indexing all content elements regardless of `colPos`. If you also run an indexing service of that type, content in the excluded columns stays searchable through it.
-   Set the option in the TypoScript of your site package, so it is versioned and identical on every environment. A value entered in the "Setup" field of a TypoScript template record in the backend is stored in the database, is not deployed with the code and has to be maintained on each environment separately. By default only administrators can edit template records.
-   When the indexing runs on the command line (queue worker, scheduler), the TypoScript is resolved for the first site root page, not for the site of the page being indexed. On a multi-site installation set the option in TypoScript that applies to all sites, such as a site package that every root template includes.
-   The pages have to be indexed again after the option was changed, content that was indexed earlier stays in the index until then.

#### Content Element Indexer Configuration {#content-element-indexer-configuration}

The following additional options are available for the content item indexer:

-   Single pages \[pages_single\]

    -   Select individual pages to be indexed here. Indexing will only occur for pages whose "Include in Search" option is enabled in the page properties.
-   Pages (recursively) \[pages_recursive\]

    -   Select the pages you want to recursively index, including all their subpages. Indexing will only occur for pages whose "Include in Search" option is enabled in the page properties.

#### File Indexer Configuration {#file-indexer-configuration}

The following additional options are available for the file indexer:

-   File collections \[file_collections\]

    -   Select the file collections that should be used for indexing. Indexing will only occur for files whose "Include in Search" option is enabled in the metadata properties.