Elasticsearch fingerprint Ingest Processor

Elasticsearch fingerprint Ingest Processor is for generating fingerprint from document content, which is similar to Logstash Fingerprint filter plugin

Idea derived from this issue.

Build

./gradlew assemble

Installation

bin/elasticsearch-plugin install file:///path/to/your/ingest-fingerprint-X.X.X.X.zip

Usage

options

Name	Required	Default	Description
fields	No	-	List of fields are used to compute fingerprint if method isn't UUID
target_field	No	fingerprint	The field to assign the fingerprint to
method	No	MD5	The hash method, one of [SHA1, SHA256, SHA512, MD5, MURMUR3 UUID]
base64_encode	No	false	Using base64 encoding for fingerprint
concatenate_all_fields	No	false	If true and method isn’t UUID, the processor concatenates all fields into one string and uses it to compute fingerprint
ignore_missing	No	false	If true and some of fields does not exist, the processor use the rest of fields to compute fingerprint. If all is missing, the processor just quietly exits without modifying the document

Example

1. Computing fingerprint from field "field1" using the default method "MD5" and write it to default target field "fingerprint".

PUT _ingest/pipeline/my_pipeline
{
    "processors": [
        {
            "fingerprint": {
                "fields": ["field1"]
            }
        }
    ]
}

2. Computing fingerprint from the fields "field1" and "field2" using the method "SHA1", encoding it by BASE64 and write it to target field "field3". Since "ignore_missing" is setted, this plugin will ignore missing fields or just exit without modifing the document when all is missing.

PUT _ingest/pipeline/my_pipeline
{
    "processors": [
        {
            "fingerprint": {
                "fields": ["field1", "field2"],
                "method": "SHA1",
                "target_field": "field3",
                "base64_encode": true,
                "ignore_missing": true
            }
        }
    ]
}

3. Computing fingerprint from the all document fields including source and meta fields using the method "SHA256", encoding it by BASE64 and write it to target field "field3"

PUT _ingest/pipeline/my_pipeline
{
    "processors": [
        {
            "fingerprint": {
                "method": "SHA256",
                "target_field": "field3",
                "base64_encode": true,
                "concatenate_all_fields": true
            }
        }
    ]
}