Elasticsearch Filter DSL Full Practice: Filter Query, Sor...

Overview

This article details practical usage of Elasticsearch Filter DSL, covering filter query, sort pagination, highlight display and batch operations.

Filter DSL

Basic Concepts

Filter DSL is Elasticsearch’s filter query language, different from query:

• Doesn’t participate in relevance scoring: Only filters documents that meet criteria
• Faster execution: No score calculation needed, can be cached in memory
• Applicable scenarios: Log analysis, data classification and other scenarios that don’t need match degree calculation

Usage Example

POST /book/_search
{
  "query": {
    "bool": {
      "must": { "match_all": {} },
      "filter": {
        "range": {
          "price": { "gte": 200, "lte": 1000 }
        }
      }
    }
  }
}

Query Sort

Default Relevance Sort

POST /book/_search
{
  "query": { "match": {"description":"solr"} }
}

Field Value Sort

POST /book/_search
{
  "query": { "match_all": {} },
  "sort": [{ "price": { "order": "desc" } }]
}

Multi-level Sort

POST /book/_search
{
  "query": { "match_all": {} },
  "sort": [
    { "price": { "order": "desc" } },
    { "timestamp": { "order": "desc" } }
  ]
}

Pagination Query

POST /book/_search
{
  "query": { "match_all": {} },
  "sort": [{ "price": { "order": "desc" } }],
  "size": 2,
  "from": 0
}

Result Highlight

POST /book/_search
{
  "query": { "match": { "name": "elasticsearch" } },
  "highlight": {
    "pre_tags": "<font color='pink'>",
    "post_tags": "</font>",
    "fields": [{ "name": {} }]
  }
}

Batch Operations

Batch Query (_mget)

POST /_mget
{
  "docs": [
    { "_index": "book", "_id": 1 },
    { "_index": "book", "_id": 2 }
  ]
}

Batch Query Same Index

POST /book/_mget
{
  "docs": [{ "_id": 2 }, { "_id": 3 }]
}

Batch Add/Delete/Update (_bulk)

POST /_bulk
{ "delete": { "_index": "book", "_id": "1" }}
{ "create": { "_index": "book", "_id": "5" }}
{ "name": "test14", "price": 100.99 }
{ "update": { "_index": "book", "_id": "2" }}
{ "doc": { "name": "test" } }

Notes

• Bulk request size: A few thousand operations per request, size a few MB
• Best practice: 10k-50k documents, about 5-15MB, default not exceed 100MB

Common Error Quick Reference

Issue	Solution
Filter seems not working	Check field type, ensure numeric/date fields use range
Sort not as expected	Use .keyword for text field sorting
Deep pagination timeout	Consider using search_after/scroll instead of from/size
Highlight no output	Ensure highlight.fields matches query field
bulk returns 400	Check JSON format, ensure action line and data line strictly paired