Skip to content



The show node prints useful information on whatever data it receives. In particular if it receives datasets, it will print out a quick view of datasets columns and the associated schema.

You can control the output format, regular spark-like dataset dumps, i.e.:

| id|clicked|
|  4|    1.6|
|  8|    3.9|
|-- id: integer (nullable = false)
|-- clicked: double (nullable = false)

Runtime Compatibility

  • PySpark :
  • Spark :


  type: punchline
  version: "6.0"
  runtime: spark
  tenant: default
  dag: [
        type: show
        component: show
        settings: {
            // optional title
            // Optional you can select some scolumns 
            // of your input dataset
            // select: [ "features", "l
            // Optional. Default if false
            // Optional, default is trurelevant for json
            // output format. 
            // Optional, set to falsetruncate long row values
            // Optional set to true, prints overtically
            vertical: false

            // Optional, default is 10
            // Optional, also show the schema
            show_schema: true
        subscribe: [
                component: input
                stream: data


Name Type mandatory Default value Description
json_format Boolean false false Output dataset as json
pretty Boolean false false use this when json_format is set to true
truncate Boolean false false truncate column size if values are too big
vertical Boolean false false Output dataset in vertical
num_rows Integer false 3 limit the number of rows to display
show_schema Boolean false true prints to stdout the dataset schema