Skip to content

Mapping

Dynamic/Explicit Field Mapping

  • Dynamic: Elasticsearch dynamically adds the field to the type mapping by default when it detects a new field in a document.
  • Explicit: Explicitly defining all possible fields in an index and their types

Normalizer

A normalizer in Elasticsearch ensures consistent indexing and searching for exact matching in fields. It preprocesses text, making it suitable for scenarios like case-insensitive sorting and aggregations.

  • It produces only one output
  • It doesn't change the main value in the document
  • For the aggregations, ES uses the result of it
  • Use normalizers with keyword fields for exact matching
  • Avoid normalizers for text fields, as normalizers are designed for keyword fields

Elasticsearch doesn't apply a default normalizer for keyword fields if none is specified. If you don't explicitly define a normalizer for a keyword field, it essentially means that no normalization will be applied, and the exact values will be indexed as they are.

Analyzer

An analyzer defines how text fields are processed during indexing and searching, allowing customization for language-specific stemming, synonym handling, and other transformations.

  • It produces multiple outputs
  • It doesn't change the main value in the document
  • For the search, ES uses the results of it
  • Apply analyzers to text fields for language-specific processing
  • Choose analyzers based on your language and search requirements

If you don't specify an analyzer for a text field in Elasticsearch, the default analyzer applied is typically the "standard" analyzer. The "standard" analyzer is a good general-purpose choice and includes tokenization, lowercasing, and removal of stop words.