This documentation describes using the csv processor in OpenSearch ingest pipelines. Consider using the Data Prepper csv processor, which runs on the OpenSearch cluster, if your use case involves large or complex datasets.

CSV processor

The csv processor is used to parse CSVs and store them as individual fields in a document. The processor ignores empty fields.

Syntax

The following is the syntax for the csv processor:

  1. {
  2.   "csv": {
  3.     "field": "field_name",
  4.     "target_fields": ["field1, field2, ..."]
  5.   }
  6. }

copy

Configuration parameters

The following table lists the required and optional parameters for the csv processor.

ParameterRequired/OptionalDescription
fieldRequiredThe name of the field containing the data to be converted. Supports template snippets.
target_fieldsRequiredThe name of the field in which to store the parsed data.
descriptionOptionalA brief description of the processor.
empty_valueOptionalRepresents optional parameters that are not required or are not applicable.
ifOptionalA condition for running the processor.
ignore_failureOptionalSpecifies whether the processor continues execution even if it encounters errors. If set to true, failures are ignored. Default is false.
ignore_missingOptionalSpecifies whether the processor should ignore documents that do not contain the specified field. If set to true, the processor does not modify the document if the field does not exist or is null. Default is false.
on_failureOptionalA list of processors to run if the processor fails.
quoteOptionalThe character used to quote fields in the CSV data. Default is “.
separatorOptionalThe delimiter used to separate the fields in the CSV data. Default is ,.
tagOptionalAn identifier tag for the processor. Useful for debugging in order to distinguish between processors of the same type.
trimOptionalIf set to true, the processor trims white space from the beginning and end of the text. Default is false.

Using the processor

Follow these steps to use the processor in a pipeline.

Step 1: Create a pipeline

The following query creates a pipeline, named csv-processor, that splits resource_usage into three new fields named cpu_usage, memory_usage, and disk_usage:

  1. PUT _ingest/pipeline/csv-processor
  2. {
  3.   "description": "Split resource usage into individual fields",
  4.   "processors": [
  5.     {
  6.       "csv": {
  7.         "field": "resource_usage",
  8.         "target_fields": ["cpu_usage", "memory_usage", "disk_usage"],
  9.         "separator": ","
  10.       }
  11.     }
  12.   ]
  13. }

copy

Step 2 (Optional): Test the pipeline

It is recommended that you test your pipeline before you ingest documents.

To test the pipeline, run the following query:

  1. POST _ingest/pipeline/csv-processor/_simulate
  2. {
  3.   "docs": [
  4.     {
  5.       "_index": "testindex1",
  6.       "_id": "1",
  7.       "_source": {
  8.         "resource_usage": "25,4096,10",
  9.         "memory_usage": "4096",
  10.         "disk_usage": "10",
  11.         "cpu_usage": "25"
  12.       }
  13.     }
  14.   ]
  15. }

copy

Response

The following example response confirms that the pipeline is working as expected:

  1. {
  2.   "docs": [
  3.     {
  4.       "doc": {
  5.         "_index": "testindex1",
  6.         "_id": "1",
  7.         "_source": {
  8.           "memory_usage": "4096",
  9.           "disk_usage": "10",
  10.           "resource_usage": "25,4096,10",
  11.           "cpu_usage": "25"
  12.         },
  13.         "_ingest": {
  14.           "timestamp": "2023-08-22T16:40:45.024796379Z"
  15.         }
  16.       }
  17.     }
  18.   ]
  19. }

Step 3: Ingest a document

The following query ingests a document into an index named testindex1:

  1. PUT testindex1/_doc/1?pipeline=csv-processor
  2. {
  3.   "resource_usage": "25,4096,10"
  4. }

copy

Step 4 (Optional): Retrieve the document

To retrieve the document, run the following query:

  1. GET testindex1/_doc/1

copy