Number of input and output link

Number of input link	:	One
Number of output link	:	One

Post-Join Assembly () is used to expand a composite record produced by a join or pre-join operation into one or more individual key–value records.

It takes an input record in the form:

[key, list(values)]

and extracts it into multiple output records:

[key, value]
    

(one record for each value in the list).

Purpose of Post-Join Assembly

The Post-Join Assembly is typically used to:

• Flatten the result of a join or Pre-Join Assembly
• Convert list-based join results back into row-based records
• Enable further transformations, aggregations, or filtering
• Prepare data for output sinks (files, databases, streams)

How It Works

• Receives an input record consisting of:
  • A key
  • A list of values
• Iterates over the list.
• Emits one output record per value, each paired with the same key.
• Sends the expanded records downstream.

Example

Input Record

[A, [10, 20, 30]]

Output Records

Key	Value
A	10
A	20
A	30

Typical Usage Flow

Source
  ↓
Transform Assembly
  ↓
Pre-Join Assembly
  ↓
Join Assembly
  ↓
Post-Join Assembly
  ↓
Transform / Group / Sink

Key Characteristics

• Converts list-based records into flat records
• Emits one or more records per input
• Preserves the original key for each emitted record
• Commonly used after Join or Pre-Join operations
• Complements the Pre-Join Assembly

Summary

The Post-Join Assembly performs the inverse operation of the Pre-Join Assembly by unpacking aggregated lists into individual records. This step restores a relational, record-oriented structure suitable for downstream processing and output.

Supported Engines

Map Reduce	Spark 1.x RDD	Spark 2.x RDD	Spark 2.x Dataset	Apache Beam 2.29.0 Batch	Apache Beam 2.29.0 Streaming	Apache Flink 1.12.0 Batch	Apache Flink 1.12.0 Streaming	Apache Tez 0.9.1	PySpark