Rewrite Manifest Files

When a table’s write pattern doesn’t align with the query pattern, metadata can be rewritten to re-group data files into manifests using rewrite_manifest action

Configurations

Attribute	Type	Description
useCaching	boolean	Use caching during the operation. Defaults to true.

Code Snippet Example 01

The below case scenario depicts rewriting manifest files of dataset stored within the Lakehouse Depot using the rewrite_manifest action.

version: v1                                   # Version
name: wf-rewrite-manifest                     # Name of the Workflow
type: workflow                                # Type of Resource (Here its workflow)
tags:                                         # Tags 
  - manifests
workflow:                                     # Workflow Section
  title: Compress iceberg manifests           # Title of the DAG
  dag:                                        # Directed Acyclic Graph (DAG)
    - name: manifest                          # Name of the Job
      title: Compress iceberg manifests       # Title of the Job
      spec:                                   # Specs
        tags:                                 # Tags
          - manifests
        stack: flare:7.0                      # Stack is Flare (Here Flare Version is 4.0)
        compute: runnable-default             # Compute
        stackSpec:                            # Flare Stack specific Section
          job:                                # Job Section
            explain: true                     # Explain
            logLevel: INFO                    # Loglevel
            inputs:                           # Inputs Section
              - name: inputDf                 # Input Dataset Name
                dataset: dataos://lakehouse:retail/pos_store_product_cust?acl=rw   # Input UDL
                format: Iceberg               # Dataset Format
            actions:                          # Flare Action
              - name: rewrite_manifest        # Action Name
                input: inputDf                # Input Dataset Name

Code Snippet Example 02

The below case scenario depicts rewriting manifest files of dataset stored within the Lakehouse using the rewrite_manifest action with useCaching as False.

version: v1
name: rewritemanifests
type: workflow
tags:
  - rewritemanifests
  - iceberg
  - actions
description: This workflow rewrite manifests of an iceberg dataset
workflow:
  title: Rewrite Manifests
  dag:
    - name: rewrite_manifests
      title: Rewrite Manifests
      description: This job rewrite manifests of an iceberg dataset
      spec:
        stack: flare:7.0
        compute: runnable-default
        stackSpec:
          job:
            explain: true
            inputs:
              - name: input
                dataset: dataos://lakehouse:retail/city
            logLevel: INFO
            actions:
              - name: rewrite_manifest
                input: input
                options:
                  useCaching: false