Workspace Support 

To enable automatic reindexing when publishing workspace records, install the workspaces extension:

composer require typo3/cms-workspaces

Without this extension, the search indexer will still work but won't automatically queue records when publishing from workspaces.

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.

// 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', 
    ]
);

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

Search engine 

Create a new record of type "Search Engine".

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

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.

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

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".

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.

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

Standard Fields 

In addition to the Indexers.md#standard-indexed-fields, the following fields are indexed by default for content elements:

Custom Fields 

Additional fields to be indexed from the content element properties can be defined using the TypoScript configuration module.tx_typo3searchalgolia.indexer.tt_content.fields. This is set by default as follows:

module {
    tx_typo3searchalgolia {
        indexer {
            tt_content {
                fields {
                    header = title
                    subheader = subTitle
                    bodytext = description
                }
            }
        }
    }
}

When to Use the Content Element Indexer 

Consider using the Content Element Indexer when:

• Your pages contain many different content elements
• You want users to find specific content within pages
• You need more granular search results than page-level indexing provides
• Your pages are large and would exceed Algolia's record size limits if indexed as a whole

Best Practices 

1. Content Element Types: Consider which content element types should be indexed. Text-based elements like text, textpic, and textmedia are typically most relevant for search.
2. Field Mapping: Customize the field mapping to ensure the most important content from your elements is properly indexed. Different content element types may need different field mappings.
3. URL Generation: Ensure that the URL generation for content elements includes proper anchors to direct users to the specific element within a page.
4. Complementary Indexing: Consider using both Page Indexer and Content Element Indexer together - pages for general content and content elements for specific details.
5. Performance Considerations: Be mindful of the number of content elements being indexed, as this can significantly increase the size of your index and affect performance.

Standard Fields 

In addition to the Indexers.md#standard-indexed-fields, the following fields are indexed by default for files:

Custom Fields 

Additional fields to be indexed from the file metadata can be defined using the TypoScript configuration module.tx_typo3searchalgolia.indexer.sys_file_metadata.fields. This is set by default as follows:

module {
    tx_typo3searchalgolia {
        indexer {
            sys_file_metadata {
                fields {
                    title = title
                    description = description
                    alternative = alternative
                    creator = author
                }

                # Comma-separated list of allowed file extensions
                extensions = pdf
            }
        }
    }
}

Supported File Types 

Currently, the File Indexer fully supports:

• PDF files (including content extraction)
• Other file types (metadata only)

The supported file extensions can be configured in the TypoScript settings as shown in the Custom Fields section.

Triggering File Indexing 

Files are indexed when:

• The indexing is triggered through the backend module
• A file is directly enqueued via the context menu
• A scheduled task runs the indexing process

Best Practices 

1. File Collections: Organize your files into logical file collections to make indexing management easier. This allows you to selectively index different types of files for different purposes.
2. Metadata Quality: Ensure your files have proper metadata (title, description, etc.) as this significantly improves search quality. Well-described files are more likely to appear in relevant search results.
3. File Size Considerations: Be mindful of very large PDF files, as content extraction can be resource-intensive. Consider setting size limits for content extraction in custom implementations.
4. File Extensions: Only enable content extraction for file types that contain searchable text. Adding non-text file types to the extensions list won't provide useful search content.
5. Security Awareness: Remember that indexed file content becomes searchable. Ensure that sensitive documents are either excluded from indexing or properly access-restricted in your search implementation.

Standard Indexed Fields 

Depending on the indexer, certain fields of a record are indexed. The following fields are common to all:

Custom Fields 

Additional fields to be indexed from the news properties can be defined using the TypoScript configuration module.tx_typo3searchalgolia.indexer.tx_news_domain_model_news.fields. This is set by default as follows:

module {
    tx_typo3searchalgolia {
        indexer {
            tx_news_domain_model_news {
                fields {
                    title = title
                    abstract = teaser
                    bodytext = description
                    author = author
                    author_email = authorEmail
                    keywords = keywords
                }
            }
        }
    }
}

Triggering News Indexing 

News articles are indexed when:

• They are created or updated
• The indexing is triggered through the backend module
• A scheduled task runs the indexing process

Best Practices 

1. News Record Quality: Ensure your news articles have proper titles, teasers, and content. Well-written news articles with descriptive titles and comprehensive content will yield better search results.
2. Metadata Usage: Make good use of the author, keywords, and categories fields in your news records. These fields can enhance search relevance and enable filtering in search results.
3. News Archive Structure: Consider how your news categories and archive structure should be reflected in search results. You might want to include category information in the indexed data for better context.
4. Pagination Handling: For news archives with many articles, consider implementing pagination in your search results to handle large result sets efficiently.
5. Date-Based Relevance: Consider configuring your search to factor in the publication date of news articles, potentially giving more weight to newer articles in search results.
6. Related Content: If your news articles relate to other content on your site, consider how to link between search results for news and other content types.

Standard Fields 

In addition to the Indexers.md#standard-indexed-fields, the following fields are indexed by default for pages:

Custom fields 

Additional fields to be indexed from the page properties can be defined using the TypoScript configuration module.tx_typo3searchalgolia.indexer.pages.fields. This is set by default as follows:

module {
    tx_typo3searchalgolia {
        indexer {
            pages {
                fields {
                    title = title
                    subtitle = subTitle
                    nav_title = navTitle
                    description = description
                    abstract = teaser
                    author = author
                    keywords = keywords
                }
            }
        }
    }
}

Triggering Page Indexing 

Pages are automatically indexed when:

• They are created or updated
• The indexing is triggered through the backend module
• A scheduled task runs the indexing process

Best Practices 

1. Selective Indexing: Only index pages that are relevant for search to keep your index size manageable.
2. Page Types: Consider which page types should be indexed. Typically, you'll want to index standard content pages but exclude special page types like folders, shortcuts, or backend modules.
3. Content Elements: Decide whether to include content elements directly in the page index. For sites with many content elements per page, it might be better to use a separate Content Element Indexer to avoid hitting Algolia's record size limits.
4. Metadata: Ensure your pages have proper metadata (titles, descriptions, keywords) for better search results.
5. Hierarchical Structure: Consider how your page hierarchy should be reflected in search results. You might want to include parent page information in child pages for context.