Runtime fields in Elasticsearch - Step-by-Step Execution

Concept Flow - Runtime fields

Search Request Received

↓

Check for Runtime Fields

↓

Evaluate Script

↓

Add Runtime Fields to Result

↓

Return Results

When a search request includes runtime fields, Elasticsearch evaluates their scripts on the fly and adds the computed values to the search results.

Execution Sample

Elasticsearch

{
  "runtime_mappings": {
    "day_of_week": {
      "type": "keyword",
      "script": "emit(doc['date'].value.dayOfWeekEnum.getDisplayName(TextStyle.FULL, Locale.ENGLISH))"
    }
  }
}

Defines a runtime field 'day_of_week' that computes the day name from a date field during search.

Execution Table

Step	Action	Input	Script Evaluation	Result
1	Receive search request	{"runtime_mappings": {"day_of_week": {...}}}	N/A	Start processing
2	Check for runtime fields	runtime_mappings present	N/A	Proceed to evaluate scripts
3	For each document	doc['date'] value	Evaluate dayOfWeekEnum.getDisplayName()	Emit day name string
4	Add runtime field to document	Emitted day name	N/A	Document enriched with 'day_of_week'
5	Return search results	All documents with runtime fields	N/A	Results sent to client

💡 All documents processed; runtime fields evaluated and added to results

Variable Tracker

Variable	Start	After Doc 1	After Doc 2	Final
doc['date'].value	N/A	2024-06-01	2024-06-02	Last document date
day_of_week (emitted)	N/A	Saturday	Sunday	Last emitted day

Key Moments - 2 Insights

Why does the runtime field script run for each document?

Because runtime fields are computed on the fly during search, the script runs for every document to produce the field's value, as shown in execution_table step 3.

Does the runtime field change the stored data in Elasticsearch?

No, runtime fields do not modify stored data; they only add computed values at query time, as seen in step 4 where the field is added to the result but not stored.

Visual Quiz - 3 Questions

Look at the execution_table, at which step does Elasticsearch evaluate the runtime field script?

A Step 2

B Step 3

C Step 4

D Step 5

💡 Hint

Check the 'Script Evaluation' column in execution_table rows

According to variable_tracker, what is the emitted value for the first document's runtime field?

A Saturday

B Sunday

C Monday

D Friday

💡 Hint

Look at 'day_of_week (emitted)' after Doc 1 in variable_tracker

If the runtime field script was removed, what would happen to the search results?

A Search would fail with an error

B Results would include the runtime field with default values

C Results would not include the runtime field

D Stored fields would be overwritten

💡 Hint

Refer to concept_flow where runtime fields are checked and evaluated only if present

Concept Snapshot

Runtime fields let you add fields computed at search time.
Define them with scripts in runtime_mappings.
Scripts run for each document during search.
They do not change stored data.
Results include these computed fields dynamically.

Full Transcript

Runtime fields in Elasticsearch are fields computed during search using scripts. When a search request includes runtime fields, Elasticsearch checks for them and evaluates their scripts for each document. For example, a runtime field can compute the day of the week from a date field. The script runs for every document, emitting the computed value, which is then added to the search result. This process does not modify stored data but enriches the results dynamically. The execution table shows the step-by-step flow from receiving the request, checking for runtime fields, evaluating scripts per document, adding the computed fields, and returning results. The variable tracker shows how document dates and emitted day names change during execution. Key moments clarify why scripts run per document and that stored data remains unchanged. The visual quiz tests understanding of when scripts run, emitted values, and effects of removing runtime fields.