Product Indexes

A product index is an index of product data on a solution – typically all standard and custom product fields as well as a lot of generated fields with additional information which is useful when publishing products to frontend (1), rendering facets (2), and for backend tasks in e.g. PIM.

To create a product index:

  1. Go to Settings > Repositories and open/create a repository
  2. Click Add index in the toolbar
  3. Name it
  4. Click OK

This creates an empty index (Figure 1.2) – you should now add instances to it.

An instance is a file in the file archive – it is this file which is being searched when a query is executed. Since instances are often rebuilt to make sure they contain the most recent changes to product data you want to create at least 2 instances. This ensures that there’s always an instance to search, even if one instance is currently being rebuilt.

Create two instances:

  1. Click Add instance
  2. Provide a name – you could call one instance ‘A’ and the other instance ‘B’
  3. Specify a folder to place the instance file under
  4. Click OK

Once created an instance will look like:

When an instance is built a set of index files are generated under System/Indexes/YourIndexName/YourInstanceName – but before you can build it you must create a build configuration.

So now that you have two instances you want to build them – to do so you need to create a build definition. Each type of index has a specific builder associated with it – in the case of a product index this builder is helpfully called the ProductIndexBuilder.

To add the build configuration:

  1. Click Add build
  2. Provide a name
  3. Select the ProductIndexBuilder
  4. Select a builder action:
    1. Full rebuilds the whole index
    2. Update rebuilds the products which have edited in the timespan between the current time and the HoursToUpdate builder setting
    3. UpdateWithIds is a mode used by the system to update smaller batches of products as they are saved in e.g. PIM
  5. Review the builder settings
  6. Set up notifications if appropriate
  7. Click OK

The following builder settings are available – please review carefully to see if any of them are relevant for your setup:

Setting

Value

Comments

BulkSize

Integer – default is 500

The number of products being built at a time

DoNotAnalyzeDefaultFields

Boolean - defaults to False

If True, schema extender fields not set to analyzed by default

DoNotStoreDefaultFields

Boolean – defaults to False

If True, schema extender fields are not set to stored by default

DoNotFailOnMismatchingProductCount

Boolean - defaults to False

If true, building an index will not fail even if the product count before indexing and after indexing is different. This may be desirable if an import job happens while the index is being built.

EmptyStringReplacement

String – default is an empty string

NULL values are not indexed by Lucene, so to be able to locate an empty field you need to index it with a dummy value – this dummy value can be specified here.

HandleInheritedCategoryValues

Boolean - defaults to False

If True, inherited product category values are indexed. This is very slow, so please don't set this to true unless you really need to.

HoursToUpdate

An integer – not set by default

If combined with the builder action Update, only the products updated within the hours specified here are rebuilt

MaxProductsToIndex

Integer – default is 2147483647

The maximum number of products to index

OnlyIndexActiveProducts

Boolean – defaults is False

If set to True, only active products are indexed

ShopsToIndex

Comma-separated list of shop IDs

This setting makes it possible to create indexes which only contain products from the specified shops/warehouses.

SkipAllExtendedFields

Boolean – defaults to False

If set to True, the fields "CampaignStartTime", "CampaignEndTime", "CampaignShowProductsAfterExpiration", "IsVariant", "ManufacturerName", "AssortmentIDs", and "StockLocationProductAvailable" are skipped

SkipCategoryFields

Boolean – defaults to False

If set to True, all product category fields are skipped

SkipDetailImages

Boolean – defaults to false

If set to true, Details images are not indexed

SkipExtenders

Boolean – defaults to False

If set to True, no custom Extenders can extend (update, remove, add) the fields in the index

SkipGrouping

Boolean – defaults to False

If set to True, the fields "GroupIDs", "ShopIDs", "GroupNames",
"GroupNumbers", "GroupDescriptions", "PrimaryGroupSort", "ParentGroupIDs", and "ParentGroupNames" are skipped

SkipGroupSorting

Boolean – defaults to false

If set to true, group sorting fields are not indexed – this may improve performance.

SkipImages

Boolean - defaults to False

If true, image paths are not indexed

SkipImagePatternImages

Boolean – defaults to false

If true, image pattern images are not indexed

SkipPrices

Boolean – defaults to false

If true, product prices are not indexed

SkipRelatedProducts

Boolean – defaults to false

If true, related products are not indexed

So now you’ve specified how you want the index to be built – you should now specify what you want to include in the index.

Lucene indexes are composed of small documents, with each document divided into named fields which contain either content which can be searched or data which can be retrieved. Each field added to the index can therefore be stored, indexed, and analyzed depending on what you want to use it for:

  • Stored fields have their values stored in the index
  • Indexed fields can be searched, the value is stored as a single value
  • Analyzed fields have their values run through an analyzer and split into tokens (words)

Generally speaking, a field you want to display in frontend must be indexed and a field where you want the user to search for parts of the value in free-text search must be analyzed. Field which are to be published using the query publisher should be stored. Fields you want to display as facets should be indexed but not analyzed.

To make things (a lot) easier for you we’ve created a default set of fields typically used in product indexes – this default field set is defined in something called the ProductIndexSchemaExtender.

To add the fields from the schema extender to the index:

  • Click Add field
  • Select the schema extender field type
  • Select the ProductIndexSchemaExtender
  • Click OK

This adds a whole bunch of fields to the index (Figure 4.1).

Most of these fields are standard product fields and various types of custom fields such as product category fields, but there’s also a whole bunch of field which are generated whenever the index is rebuilt e.g. the BoughtWithProducts field, the GroupIDs field, and many others. For each field you can see the Name, System Name, Source and Type – and whether or not the field is stored, indexed and analyzed.

As you can see, all String-type fields are analyzed by default, which is ideal if you want to include them in e.g. a free-text query – but it may be a problem if you want to e.g. create facets based on the field. Fortunately you can also add fields to the index manually – see the Custom Fields article.

Once you’ve added instances, a build configuration, and a set of fields to the index you should build it – to do so click a yellow button with an instance name on it (Figure 5.1).

Of course, you don’t want to do this manually every time – you want to do a combination of the following:

  • Rebuild the index every time an integration job has been executed
  • Rebuild the index every time a product is saved in PIM
  • Rebuild the index on a schedule – see the article on tasks

If you have more than three instances defined you can also choose between two different balancing methods for selecting an alternative index when the primary index is being built:

  1. Click Balancer in the toolbar
  2. Choose between the two modes:
    1. ActivePassive selects the next instance on the list of instances – so if instance A is unavailable (building, has failed), instance B will be used unless it’s unavailable, in which case instance C will be used, and so on.
    2. LastUpdated selects the most recently built index

On solutions with heavy traffic and frequent product data updates we recommend using the LastUpdated mode to ensure that visitors are always shown the most recently updated product data. On solutions with only two instances (the vast majority of solutions) it is not necessary to select a balancer mode, as the “other index” will always be used when an index is unavailable.