Stdout Sink Node

The stdout node serializes each record it receives with the configured encoding and prints it to the standard output of the ZephFlow process, one record at a time. Internal ZephFlow engine command for local and CLI runs. Not offered in Fleak Data Workflow and not supported in Fleak Data Workflow workflows; do not add it to a workflow.

Internal engine command

Internal ZephFlow engine command for local and CLI runs. Not offered in Fleak Data Workflow and not supported in Fleak Data Workflow workflows; do not add it to a workflow.

Key Features

• Per-record output: Each record is serialized and printed on its own
• Compact JSON: JSON encodings print a record on a single line, without pretty-printing
• Shared stream: Records and ZephFlow log lines are written to the same standard output

Config Object

The full config object for the stdout node (StdOutDto.Config):

Field	Type	Required	Default	Description
encodingType	EncodingType	Yes	—	Output format of each record. One of CSV, JSON_OBJECT, JSON_ARRAY, JSON_OBJECT_LINE. Other values are rejected

Enum values are case-sensitive: json_object fails DAG compilation.

Output Format

One Record at a Time

Each record is serialized on its own and printed followed by a newline, so the encoding applies to a single record:

encodingType	Output for records {"a":1} and {"a":2}
JSON_OBJECT	{"a":1} then {"a":2}, one per line
JSON_OBJECT_LINE	{"a":1} then {"a":2}, one per line
JSON_ARRAY	[{"a":1}] then [{"a":2}]: each record in its own one-element array
CSV	A header line and a value line for every record, each followed by an empty line

CSV lines end with \r\n. A record that fails to serialize is logged at ERROR level and skipped.

Log Lines

ZephFlow writes its log lines to the same standard output, so records appear between timestamped log lines such as 2026-10-08 22:07:07.770 [main] INFO io.fleak.zephflow.clistarter.Main - Selected metric client: noop. When another program reads the output, keep only the record lines, for example by filtering out lines that match the log format.

Job Context

The DAG definition must contain a jobContext block, which may be empty (jobContext: {}). Without it, the CLI starter fails with a NullPointerException before the DAG runs.

Examples

DAG Definition

Save as stdout_example.yml:

jobContext: {}

dag:
  - id: "stdin_source"
    commandName: "stdin"
    config:
      encodingType: "JSON_ARRAY"
    outputs:
      - "stdout_sink"

  - id: "stdout_sink"
    commandName: "stdout"
    config:
      encodingType: "JSON_ARRAY"
    outputs: []

Run it with the CLI starter:

echo '[{"id": 1, "level": "INFO"}, {"id": 2, "level": "ERROR"}]' | docker run -i \
  -v $(pwd)/stdout_example.yml:/app/stdout_example.yml \
  fleak/zephflow-clistarter:latest -f /app/stdout_example.yml

The input array is split into two records, and each is printed as its own array. Between the log lines, the output contains the stdin prompt and the records (field order within a record may differ from the input):

use an empty line to quit
[{"level":"INFO","id":1}]
[{"level":"ERROR","id":2}]

Related nodes

• stdin: Read records from standard input
• filesource: Read a single local file
• kafkasink: Write records to a Kafka topic
• s3sink: Write records to AWS S3