Configuration
The following section explains the configuration of the cartridge’s content indexing job, including its configuration parameters and default attributes.
Content indexing job structure
The Content Indexing job’s general structure and workflow mirror those of product indexing jobs.
The cartridge’s content indexing uses the Salesforce B2C Content Search Model (ContentSearchModel) to browse content assets.
This is why the job has two steps: SFCCContentSearchIndex and algoliaContentIndex. Job first refreshes the SFCC content search index before extracting content assets and sends them to Algolia.
Indexed content attributes
The content job fully reindexes all content data/records with each job run. It is possible to configure at the job level which attributes to send using the attributeList parameter.
By default, the content indexing job sends the following attributes to Algolia: id, name, description, url, body, page, and algolia_chunk_order.
Unlike for the product indexing jobs where there is a Custom Preference for defining defaults (that can be overridden per job), content attributes sent to Algolia are only configured at the job level by utilizing the attributeList parameter.
The job intentionally ignores and strips out certain HTML tags, such as the <script> tag, as indexing content from these tags is not meaningful for search purposes.
Base attributes
The base set of attributes is always included in content indexing, regardless of whether these are specified in attributeList.
| Attribute | Description |
|---|---|
| id | Content Asset or Page ID (on Salesforce B2C) |
| objectId | A unique identifier on Algolia’s side for the Content Asset or Page (id + "_" + algolia_chunk_order) |
| algolia_chunk_order | Used for sorting chunks that belong to the same content asset/page |
Configurable attributes
You have the option to send additional content record attributes to Algolia by listing them in the Attribute List as a job parameter.
The default attributes include name, description, url, body, and page. Expand this list as needed for your use case (for example, pageTitle).
| Attribute | Description |
|---|---|
| name | The name of the content asset or page |
| description | A description of the content asset or page |
| url | The URL that provides access to the page or content asset |
| body | The primary content displayed to users on the storefront |
| page | Indicates whether the item is a (Page Designer) page (true if so) |
| classificationFolder | The classification folder for the content (not applicable for Page Designer pages if the page is not assigned to a folder) |
| pageDescription | A description metadata field, often used for SEO purposes |
| pageKeywords | A list of keywords, often used for SEO purposes but potentially also useful for site search |
| pageTitle | A page title, used for SEO purposes and to define the browser’s page title |
| template | Applied for rendering content on Salesforce B2C |
Customize indexed attributes
It is not possible to exclude algolia_chunk_order and objectId from the attributes being sent. However, you can specify which attributes should always be included by using defaultAttributes in cartridges/int_algolia/cartridge/scripts/algolia/lib/algoliaContentConfig.js.
Index unsearchable page designer content
By default, the cartridge conforms to Salesforce B2C’s standard indexing policy, which indexes only the searchable attributes of Page Designer components. However, this can be overridden on a code level with the indexOnlySearchables boolean flag in file [specify file]. If indexOnlySearchables is set to false, all attributes become indexable without the need for developer support for marking an attribute as searchable, provided that the page is marked as searchable in the Business Manager (BM). Nonetheless, more granular control can be retained over which specific attribute types are indexed (indexableAttributes) and which are explicitly excluded (ignoredAttributes).
Content indexing job configuration
The process has two steps:
- SFCCContentSearchIndex: Prepares and updates the SFCC content index before browsing and sending content data to Algolia (in the second step). This is a predefined Salesforce B2C Commerce job step that doesn’t requires any additional configuration.
- algoliaContentIndex: Extracts data from content assets and Page Designer pages and indexes it with Algolia.
The job configuration is accessible in the Job Steps tab. Set the scope to Specific Sites and select all sites for which the job should be executed.
Content indexing job parameters
| Parameter | Description |
|---|---|
| attributeList | A comma-separated list of attributes to be indexed. |
| failureThresholdPercentage | The maximum percentage of records allowed to have errors during indexing. For example, if the threshold is 10%, the indexing job fails if more than 10% of your records have an error during indexing. The temporary index will not be moved to production if the failure rate exceeds this threshold. The default is 0%. |
| splittingTag | The HTML tag used for splitting content asset records into chunks. Without specifying this delimiter tag, there will be no semantic splitting, but only record size-based splitting. Example: h1 |
| includedContent | Defines the indexing scope: allContents indexes both content assets and Page Designer components, contentAssets exclusively indexes content assets, and pageDesignerComponents indexes only the page designer components. |
Content assets are indexed to a single attribute: body. Page Designer pages are more complex, as they can consist of multiple regions and components, each with several attributes.
The structure of pages, regions, and components is adaptable. To effectively manage this structure, the cartridge uses the pageDesignerHelper.js file, which helps you optimize extracting data from a single page.
pageDesignerHelper.js utilizes a recursive approach to effectively manage and structure the hierarchical, tree-like data. For a comprehensive overview, see Salesforce’s Page Designer documentation.
Attribute selection and performance
The job employs a retry strategy akin to those used by other API clients.
The number and type of content attributes sent to Algolia impacts job performance: the more attributes you select, the longer the indexing takes. Additionally, enabling indexing of unsearchable Page Designer components might lead to longer indexing jobs. For detailed insights, refer to the performance considerations section.
Content search UI
The cartridge provides a content search UI that you can use to display content results on your storefront.
