Configuring the Analysis Pipeline Manually

This section outlines the high-level procedure to configure an analysis pipeline with both document and semantic processors, and to display the output in the search logic.

To map data to index fields and categories, you must define:

The document processors to generate output contexts, or metas, based on the name specified in the Output to box.
The semantic processors to generate annotations.
Map all contexts and annotations to index fields or categories.

Before performing the mapping, you must create the target index fields under Index > Data Model > Advanced Schema.

You can then configure in Index > Data Processing > Mappings, a list of:
- metas and parts mapping document contexts to index fields
- annotations mapping document annotations to index fields

Important: All changes made to the analysis pipeline require clearing the index and reindexing data to avoid inconsistencies.

This task shows you how to:

Map Document Processors Manually
Map Semantic Processors Manually
Set Mapping Limits for Index Fields

Map Document Processors Manually

Add document processors to the list of current processors, and define the output contexts (or metas) for them.
1. Go to Index > Data processing > pipeline name > Document Processors.
2. Drag the appropriate processor to the list.
3. Configure the processor.
Create index fields to store the contexts.
1. Go to Index > Data Model > Advanced Schema.
2. Click Add Field.

Define mappings for the new context generated by the document processors.

Go to Index > Data processing > pipeline name > Document Processors.
Select the Mappings subtab.
Click Add mapping source and create a mapping with the same name as the output context.
Click Add mapping target and select either Index field or Category field (for facet).

Under Details, configure indexing options for the index field or category field. The following table describes the most important ones.


Option	Description
Searchable	Preindexes the data for efficient lookup. This allows users to search for documents based on the available values in this index field.
Retrievable	Displays the content of the field in the hit content of the search results.
Indexing options	For each target field, you can configure the form of words to be indexed. Choose one of these: index exact form index lowercase form index normalized form (this removes accents and specifies the contents to lowercase) You can also choose to index the position of separators to enable search within a string. Select this option for use with a "split" type prefix handler.

Map Semantic Processors Manually

Drag semantic processors to the list of current processors
1. Go to Index > Data processing > pipeline name > Semantic Processors.
2. Drag the appropriate processor to the list.
3. Configure the processor.
Define mappings for the new annotations generated by the semantic processors.
1. On the Mappings subtab, select Add mapping source.
2. To create annotation mappings, follow Map Document Processors Manually > Step 3.
To display category or index fields in the search results:
- To display categories as facet in the search results, add a facet to your search logic.
- To display index fields in the hit content of the search results, create a hit meta (or add the field to an existing hit meta) in the search logic.
Click Apply.
Clear the index and reindex your data.

Set Mapping Limits for Index Fields

You can configure a list of field retrieval limits to define the size limits for the index fields.

Go to Index > Data Processing > Mapping Limits.
In Max indexed words, specify the maximum number of words that are searchable for the field.
For example, if you index 500 words in this field, but set a limit to 100, then only the first 100 words are searchable.
In Max size stored for display, specify the maximum number of bytes that are retrieved for the field.